Image-Text-to-Text
Transformers
Safetensors
English
qwen3_5
decision-model
system-one
calibrated-probabilities
typed-decisions
ainode
merged-lora
conversational
Eval Results (legacy)
Instructions to use frontier-infra/jebadiah-9b-v2 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use frontier-infra/jebadiah-9b-v2 with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("image-text-to-text", model="frontier-infra/jebadiah-9b-v2") messages = [ { "role": "user", "content": [ {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"}, {"type": "text", "text": "What animal is on the candy?"} ] }, ] pipe(text=messages)# pip install -U transformers accelerate # Load model directly from transformers import AutoProcessor, AutoModelForMultimodalLM processor = AutoProcessor.from_pretrained("frontier-infra/jebadiah-9b-v2") model = AutoModelForMultimodalLM.from_pretrained("frontier-infra/jebadiah-9b-v2", device_map="auto") messages = [ { "role": "user", "content": [ {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"}, {"type": "text", "text": "What animal is on the candy?"} ] }, ] inputs = processor.apply_chat_template( messages, add_generation_prompt=True, tokenize=True, return_dict=True, return_tensors="pt", ).to(model.device) outputs = model.generate(**inputs, max_new_tokens=256) print(processor.decode(outputs[0][inputs["input_ids"].shape[-1]:])) - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- vLLM
How to use frontier-infra/jebadiah-9b-v2 with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "frontier-infra/jebadiah-9b-v2" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "frontier-infra/jebadiah-9b-v2", "messages": [ { "role": "user", "content": [ { "type": "text", "text": "Describe this image in one sentence." }, { "type": "image_url", "image_url": { "url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg" } } ] } ] }'Use Docker
docker model run hf.co/frontier-infra/jebadiah-9b-v2
- SGLang
How to use frontier-infra/jebadiah-9b-v2 with SGLang:
Install from pip and serve model
# Install SGLang from pip: pip install sglang # Start the SGLang server: python3 -m sglang.launch_server \ --model-path "frontier-infra/jebadiah-9b-v2" \ --host 0.0.0.0 \ --port 30000 # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:30000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "frontier-infra/jebadiah-9b-v2", "messages": [ { "role": "user", "content": [ { "type": "text", "text": "Describe this image in one sentence." }, { "type": "image_url", "image_url": { "url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg" } } ] } ] }'Use Docker images
docker run --gpus all \ --shm-size 32g \ -p 30000:30000 \ -v ~/.cache/huggingface:/root/.cache/huggingface \ --env "HF_TOKEN=<secret>" \ --ipc=host \ lmsysorg/sglang:latest \ python3 -m sglang.launch_server \ --model-path "frontier-infra/jebadiah-9b-v2" \ --host 0.0.0.0 \ --port 30000 # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:30000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "frontier-infra/jebadiah-9b-v2", "messages": [ { "role": "user", "content": [ { "type": "text", "text": "Describe this image in one sentence." }, { "type": "image_url", "image_url": { "url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg" } } ] } ] }' - Docker Model Runner
How to use frontier-infra/jebadiah-9b-v2 with Docker Model Runner:
docker model run hf.co/frontier-infra/jebadiah-9b-v2
Jebadiah 9B v2: model card, merge verification, standalone script
Browse files- README.md +236 -0
- eval/merge-verify.json +2700 -0
- scripts/decide_standalone.py +3 -0
README.md
ADDED
|
@@ -0,0 +1,236 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
base_model: Qwen/Qwen3.5-9B
|
| 3 |
+
base_model_relation: finetune
|
| 4 |
+
library_name: transformers
|
| 5 |
+
license: apache-2.0
|
| 6 |
+
language:
|
| 7 |
+
- en
|
| 8 |
+
tags:
|
| 9 |
+
- decision-model
|
| 10 |
+
- system-one
|
| 11 |
+
- calibrated-probabilities
|
| 12 |
+
- typed-decisions
|
| 13 |
+
- ainode
|
| 14 |
+
- merged-lora
|
| 15 |
+
datasets:
|
| 16 |
+
- LocalLLaMA/typed-decisions
|
| 17 |
+
- nvidia/HelpSteer2
|
| 18 |
+
- mteb/summeval
|
| 19 |
+
model-index:
|
| 20 |
+
- name: jebadiah-9b-v2
|
| 21 |
+
results:
|
| 22 |
+
- task:
|
| 23 |
+
type: text-classification
|
| 24 |
+
name: typed decisions (choice, noul, score)
|
| 25 |
+
dataset:
|
| 26 |
+
name: Jevals suite 0.1.0, PubMedQA (noul)
|
| 27 |
+
type: jevals-pubmedqa
|
| 28 |
+
metrics:
|
| 29 |
+
- type: accuracy
|
| 30 |
+
value: 90.3
|
| 31 |
+
- type: decision_score_jevals
|
| 32 |
+
value: 68.6
|
| 33 |
+
- task:
|
| 34 |
+
type: text-classification
|
| 35 |
+
name: typed decisions (choice, noul, score)
|
| 36 |
+
dataset:
|
| 37 |
+
name: Jevals suite 0.1.0, Banking77 (choice, 77-way)
|
| 38 |
+
type: jevals-banking77
|
| 39 |
+
metrics:
|
| 40 |
+
- type: accuracy
|
| 41 |
+
value: 70.7
|
| 42 |
+
- type: decision_score_jevals
|
| 43 |
+
value: 58.2
|
| 44 |
+
- task:
|
| 45 |
+
type: text-classification
|
| 46 |
+
name: typed decisions (choice, noul, score)
|
| 47 |
+
dataset:
|
| 48 |
+
name: Jevals suite 0.1.0, HelpSteer2 helpfulness (score)
|
| 49 |
+
type: jevals-helpsteer2
|
| 50 |
+
metrics:
|
| 51 |
+
- type: decision_score_jevals
|
| 52 |
+
value: 10.4
|
| 53 |
+
- task:
|
| 54 |
+
type: text-classification
|
| 55 |
+
name: typed decisions (choice, noul, score)
|
| 56 |
+
dataset:
|
| 57 |
+
name: Nimble public human-labelled subsets (13, macro accuracy)
|
| 58 |
+
type: nimble-public
|
| 59 |
+
metrics:
|
| 60 |
+
- type: accuracy
|
| 61 |
+
value: 77.0
|
| 62 |
+
---
|
| 63 |
+
|
| 64 |
+
# Jebadiah 9B v2
|
| 65 |
+
|
| 66 |
+
|
| 67 |
+
Jebadiah (Jeb for short) is Frontier Infra's open System One style decision model, trained and served with AINode: it answers typed questions with a
|
| 68 |
+
probability over the option labels instead of generating text, calibrated per question type when the shipped
|
| 69 |
+
temperatures are applied. Three question types, the
|
| 70 |
+
same three Jev uses: **choice** (pick one of N), **noul** (a yes or no statement, returned as P(yes)) and
|
| 71 |
+
**score** (place the state on an ordered rubric). It is trained with AINode's trainer on public data only
|
| 72 |
+
and served by AINode's `/v1/decide` and `/v1/systemone` routes on any NVIDIA GPU. The routes answer Jev's
|
| 73 |
+
three question types; the request schema is AINode's own (documented on the route), not Jev's exact wire format.
|
| 74 |
+
|
| 75 |
+
<img src="jeb-banner.png" alt="They call me Jeb. He does not talk much. He just decides." width="100%">
|
| 76 |
+
|
| 77 |
+
This repository holds **v2 9B** as full bf16 weights: the v2 LoRA merged into `Qwen/Qwen3.5-9B` (pinned revision
|
| 78 |
+
`c2022362`), so it loads with transformers alone, no PEFT and no separate base download.
|
| 79 |
+
|
| 80 |
+
**What changed from v1 is one thing: the base checkpoint.** v1 was trained on the Base checkpoint
|
| 81 |
+
`Qwen/Qwen3.5-9B-Base` (revision `68c46c4b`). v2 is trained on the **chat** checkpoint `Qwen/Qwen3.5-9B` with thinking off
|
| 82 |
+
(the chat template rendered with `enable_thinking=false`, the prompt contract v1 already used). Everything else
|
| 83 |
+
in the recipe is unchanged from [v1](https://huggingface.co/frontier-infra/jebadiah-9b-v1): the same public pool (11,013 records,
|
| 84 |
+
15,813 questions) and the same train and calibration split, the same LoRA (rank 16, alpha 32, every linear
|
| 85 |
+
projection), the same one epoch, the same learning rate 1e-4 and seed, the same objective (cross-entropy over the
|
| 86 |
+
option-label logits with the ordinal score target) and the same temperature fit. That base swap is what the gain
|
| 87 |
+
is: headline **73.9** against v1's 73.3. The run trained on one NVIDIA A100 80GB PCIe in
|
| 88 |
+
90 minutes. v2 is a fixed version; newer versions are published as new repositories and never
|
| 89 |
+
replace this one.
|
| 90 |
+
|
| 91 |
+
Made in Texas.
|
| 92 |
+
|
| 93 |
+
**Read this first.** Served through AINode today, `/v1/systemone` and `/v1/decide` return the model's raw distribution; the per-type temperatures in `temperatures.json` are applied offline in every calibrated number on this card (AINode issue 276 tracks applying them on the route). To reproduce the numbers, or to get the calibrated answer yourself, use the standalone script in `scripts/` (below).
|
| 94 |
+
|
| 95 |
+
## Results
|
| 96 |
+
|
| 97 |
+
Measured by us with AINode's bench, one logit read per question, the same rendered prompt for every
|
| 98 |
+
model. Accuracy is the share of questions whose top label is the human label. Decision Score is Jevals'
|
| 99 |
+
metric: 100 = perfect, 0 = guessing the label base rates, below 0 = worse than that. These are our
|
| 100 |
+
numbers on the public suites, not rows on the Jevals board. Headline (macro over the zero-shot public
|
| 101 |
+
sets): **73.9** (v1: 73.3). Its parts: Jevals PubMedQA 90.3 (v1 89.7), Jevals Banking77 70.7 (v1 70.0), Jevals HelpSteer2 40.7 (v1 40.3), Nimble 324 81.2 (v1 78.7), Kev transfer-v4 83.8 (v1 84.0), Nimble public macro 77.0 (v1 77.0).
|
| 102 |
+
|
| 103 |
+
The v1 columns are the published [v1](https://huggingface.co/frontier-infra/jebadiah-9b-v1) records, same sets, same bench, same repeats.
|
| 104 |
+
|
| 105 |
+
| Set (questions) | Type | Accuracy | v1 accuracy | Floor (majority label) | Decision Score (Jevals) | v1 Decision Score | ECE, temperatures applied | Flips over identical repeats |
|
| 106 |
+
|---|---|---:|---:|---:|---:|---:|---:|---:|
|
| 107 |
+
| Jevals PubMedQA (300) | noul | 90.3 | 89.7 | 62.0 | 68.6 | 66.8 | 0.025 | 0.0% |
|
| 108 |
+
| Jevals Banking77 (300, 77 options) | choice | 70.7 | 70.0 | 1.3 | 58.2 | 56.9 | 0.076 | 0.7% |
|
| 109 |
+
| Jevals HelpSteer2 helpfulness (300, 5 levels) | score | 40.7 | 40.3 | 41.7 | 10.4 | 11.9 | 0.039 (raw 0.066) | 2.3% |
|
| 110 |
+
| Nimble held-out eval (324) | mixed | 81.2 | 78.7 | 17.6 | 65.0 | 65.0 | 0.085 | 0.3% |
|
| 111 |
+
| Kev transfer-v4 test (764) | mixed | 83.8 | 84.0 | 21.5 | 67.9 | 67.0 | 0.025 | 0.3% |
|
| 112 |
+
| Kev decision-v7 test (1,440) | mixed | 81.2 | 80.8 | 20.3 | 71.4 | 70.6 | 0.050 | n/a |
|
| 113 |
+
| typed-decisions test (2,000) | mixed | 78.3 | 78.9 | 15.3 | 56.5 | 56.4 | 0.156 | 0.1% |
|
| 114 |
+
|
| 115 |
+
Nimble's 13 public human-labelled subsets (3,880 questions), macro accuracy: **77.0** (v1: 77.0).
|
| 116 |
+
For scale, Bespoke's published table puts Nimble-9B at 74.8 and Jev at 76.0 on the same subsets with their scorer.
|
| 117 |
+
|
| 118 |
+
| Nimble public subset (questions) | Accuracy | v1 accuracy | Decision Score (Jevals) | v1 Decision Score |
|
| 119 |
+
|---|---:|---:|---:|---:|
|
| 120 |
+
| aegis2 (250) | 78.4 | 79.2 | 38.0 | 35.8 |
|
| 121 |
+
| boolq (300) | 86.0 | 85.7 | 54.6 | 57.7 |
|
| 122 |
+
| civil_comments (300) | 86.0 | 86.7 | -7.4 | -4.8 |
|
| 123 |
+
| helpsteer2 (249) | 43.4 | 41.0 | 13.3 | 15.2 |
|
| 124 |
+
| massive-de-DE (350) | 84.6 | 84.9 | 74.3 | 73.3 |
|
| 125 |
+
| massive-en-US (350) | 85.4 | 85.7 | 77.3 | 76.0 |
|
| 126 |
+
| multinli (299) | 84.9 | 87.6 | 67.2 | 70.6 |
|
| 127 |
+
| paws (250) | 83.6 | 85.6 | 54.0 | 57.2 |
|
| 128 |
+
| pubmedqa (250) | 72.8 | 68.8 | 38.5 | 34.5 |
|
| 129 |
+
| squad2 (299) | 79.9 | 80.3 | 42.9 | 43.6 |
|
| 130 |
+
| summeval-consistency (144) | 87.5 | 87.5 | 42.6 | 36.0 |
|
| 131 |
+
| summeval-relevance (240) | 52.5 | 51.7 | 23.4 | 25.9 |
|
| 132 |
+
| vitaminc-dev (599) | 75.6 | 76.3 | 41.8 | 39.0 |
|
| 133 |
+
|
| 134 |
+
Where v2 is not better. Jevals HelpSteer2 Decision Score is 10.4 against v1's 11.9, and its top level flips over identical repeats on 2.3% of questions against 0.3%. The Nimble public macro is flat (77.0 against 77.0): MultiNLI (84.9 against 87.6) and PAWS (83.6 against 85.6) give back what PubMedQA (72.8 against 68.8) and HelpSteer2 gain. ECE on the Nimble 324 set rises from 0.063 to 0.085. The headline gain is 0.6 points, most of it from the Nimble 324 set (81.2 against 78.7).
|
| 135 |
+
|
| 136 |
+
Two things to read carefully, unchanged from v1 because the pool is unchanged. HelpSteer2 and SummEval are not
|
| 137 |
+
zero-shot for Jeb: the pool trains on the HelpSteer2 **train** split (the Jevals and Nimble items come from
|
| 138 |
+
validation, and no item overlaps) and on SummEval articles that Nimble does not score (its 15 evaluation articles
|
| 139 |
+
are held out entirely), so those rows are held-out items of a seen rubric. And the in-distribution sets
|
| 140 |
+
(typed-decisions, Kev decision-v7) are the test splits of sources in the pool.
|
| 141 |
+
|
| 142 |
+
Every eval record (per question: option keys, probabilities, pick, label, repeat, option order) is in
|
| 143 |
+
`eval/`, with `eval/results.json` carrying the full metric set. The records were scored on the unmerged adapter;
|
| 144 |
+
the merged weights here reproduce them (see Training, merge).
|
| 145 |
+
|
| 146 |
+
## Robustness to irrelevant content (nonce test)
|
| 147 |
+
|
| 148 |
+
The nonce pass (a fresh UUID planted in the state or appended to the instructions, three passes per question on the four head sets, as on the v1 card) had not finished for this model when this card was written. It is pending; no robustness number is claimed here until its records exist.
|
| 149 |
+
|
| 150 |
+
## How it decides
|
| 151 |
+
|
| 152 |
+
**Which temperature fit, and why.** `temperatures.json` carries two fits on the calibration split (689 records, 921 questions: choice 225, noul 147, score 549). The `hard` fit minimises NLL against the argmax label and sharpens (T 0.79 / 0.67 / 0.83). The `train` fit minimises NLL against the training target, the soft or ordinal distribution the model was taught, and softens (T 1.19 / 1.09 / 1.22). v2 applies `train`, as v1 does, because the ordinal target is what makes rubric probabilities honest, and the `hard` fit is exactly the sharpening that made v0's HelpSteer2 calibration worse. The price is visible in the same file: on the calibration split measured against hard labels, ECE moves from 0.074 to 0.089 (choice), 0.064 to 0.064 (noul) and 0.092 to 0.128 (score) while NLL on the training target falls (0.4594 to 0.4542, 0.3776 to 0.3768, 1.1017 to 1.0929). The ECE column in the results table is measured on the evaluation sets, not on this split, with the applied temperatures. Both fits are in the file; a consumer who gates on the argmax may prefer `hard`, and either way should refit on its own data.
|
| 153 |
+
|
| 154 |
+
The prompt is AINode's own decide rendering (`ainode.api.decide.build_messages`, source commit
|
| 155 |
+
`e5c08938`, `prompt_source_sha256` in `prompt_contract.json`), through the chat template with
|
| 156 |
+
thinking off. The option labels are single tokens; the model's answer is the distribution over those label
|
| 157 |
+
tokens at the last prompt position, read in fp32 from the last hidden state and then temperature scaled
|
| 158 |
+
per type (`temperatures.json`: choice 1.19, noul 1.09, score 1.22). Nothing is generated.
|
| 159 |
+
|
| 160 |
+
The supported way to run it is AINode, which renders the prompt exactly as trained (a worked
|
| 161 |
+
`/v1/systemone` request and response is in the AINode docs issue tracker until the page lands):
|
| 162 |
+
|
| 163 |
+
```bash
|
| 164 |
+
curl -sS https://<your-ainode>/v1/systemone -H "Authorization: Bearer <key>" -H "Content-Type: application/json" -d '{
|
| 165 |
+
"model": "frontier-infra/jebadiah-9b-v2",
|
| 166 |
+
"state": {"ticket": "Customer says the invoice total does not match the quote."},
|
| 167 |
+
"questions": {
|
| 168 |
+
"route": {"type": "choice", "instructions": "Which team should take this ticket?", "criteria": {"billing": "an invoice, a charge or a refund", "support": "a product question", "sales": "a quote or a renewal"}},
|
| 169 |
+
"urgent": {"type": "noul", "instructions": "The customer is blocked from working.", "criteria": {"true": "work has stopped", "false": "it can wait"}}
|
| 170 |
+
}
|
| 171 |
+
}'
|
| 172 |
+
```
|
| 173 |
+
|
| 174 |
+
The weights keep the base repository's exact layout (`Qwen3_5ForConditionalGeneration`, the same `config.json`,
|
| 175 |
+
tokenizer and chat template as `Qwen/Qwen3.5-9B` at `c2022362`); only the LoRA-targeted language-model weights differ.
|
| 176 |
+
The vision tower is the base's, untouched and unused.
|
| 177 |
+
|
| 178 |
+
### Standalone, reproducible
|
| 179 |
+
|
| 180 |
+
`scripts/` in this repository carries the trainer's own copies of the renderer and the logit read: `ainode_prompt_verbatim.py` is AINode's decide rendering copied verbatim from [getainode/ainode at e5c08938](https://github.com/getainode/ainode/blob/e5c089386e0239c9eb270eeb490d181722b8da5b/ainode/api/decide.py) (the sha256 of its copied definitions is the `prompt_source_sha256` in `prompt_contract.json`), `jebadiah_prompt.py` turns a `/v1/systemone` question into that prompt with single-token labels, `jebadiah_model.py` reads the label-token logits at the answer position in fp32, and `decide_standalone.py` runs one request end to end. These four files are byte-identical to v1's except `decide_standalone.py`, which takes the merged model directory instead of a base plus an adapter:
|
| 181 |
+
|
| 182 |
+
```bash
|
| 183 |
+
pip install torch transformers==5.17.0 huggingface_hub
|
| 184 |
+
huggingface-cli download frontier-infra/jebadiah-9b-v2 --local-dir jeb
|
| 185 |
+
python jeb/scripts/decide_standalone.py --model jeb --request jeb/scripts/example-request.json # add --no-temperatures for the raw distribution
|
| 186 |
+
```
|
| 187 |
+
|
| 188 |
+
It prints the answers in the route's shape (choice: `choice`, `confidence`, `probabilities`; noul: `noul` = P(true); score: `score`, `legend`, `probabilities`) plus the temperatures it applied. Runs on CPU too (slow without the DeltaNet kernels), and `--device mps` on Apple silicon, which is how it was checked before publishing.
|
| 189 |
+
|
| 190 |
+
Thresholds belong to the caller: act on a high probability, confirm or escalate on a middle one, hand a
|
| 191 |
+
low one to a person or a bigger model. The model never refuses; policy is built from decisions.
|
| 192 |
+
|
| 193 |
+
## Training
|
| 194 |
+
|
| 195 |
+
- **Base:** `Qwen/Qwen3.5-9B` at revision `c202236235762e1c871ad0ccb60c8ee5ba337b9a`, the chat checkpoint, prompted with thinking off. (v1: `Qwen/Qwen3.5-9B-Base`.)
|
| 196 |
+
- **Objective:** cross-entropy over the candidate option-label logits at the answer position. Choice and
|
| 197 |
+
noul use the source's gold distribution where it provides one (typed-decisions) and the hard label
|
| 198 |
+
otherwise; score questions use an ordinal kernel around the human label (adjacent level 0.2 of the
|
| 199 |
+
label's weight), and the score temperature is fitted on that same target. Unchanged from v1.
|
| 200 |
+
- **Adapter, before merging:** LoRA r=16, alpha=32, dropout 0.05 on every linear projection including the Gated DeltaNet
|
| 201 |
+
projections, 43.3M trainable parameters. bf16, SDPA attention, micro-batch 8, learning rate 1e-4,
|
| 202 |
+
one epoch, 1,862 steps, gradient checkpointing, seed 17. Unchanged from v1.
|
| 203 |
+
- **Data (public only), 11,013 records and 15,813 questions in the pool, unchanged from v1:** `LocalLLaMA/typed-decisions` train (Apache-2.0,
|
| 204 |
+
revision `ea930645`); the Kev v7 training sources whose licenses permit derived weights (BoolQ, MNLI,
|
| 205 |
+
DBpedia14 and Kev's contrastive and composition sets, Kev suites revision `a88f56db`); `nvidia/HelpSteer2`
|
| 206 |
+
train split (CC BY 4.0, 2,573 score questions: helpfulness in the Jevals task's exact shape for 1,609
|
| 207 |
+
rows, plus correctness, coherence, complexity and verbosity on 241 of them); SummEval (MIT, via
|
| 208 |
+
`mteb/summeval`, 1,664 score questions over four dimensions on 26 articles, with Nimble's 15 evaluation
|
| 209 |
+
articles excluded). 95/5 family split into 10,324 training records (14,892 training questions) and 689 calibration records (921 calibration questions).
|
| 210 |
+
Nimble's train set is excluded (no license stated); the Jevals test items are excluded from training.
|
| 211 |
+
No private data of any kind.
|
| 212 |
+
- **Compute:** one NVIDIA A100 80GB PCIe, 90 min training, 16 min evaluation, torch 2.11, PEFT 0.21. (v1 trained on an H100 PCIe.)
|
| 213 |
+
- **Merge:** W' = W + (alpha / r) B A = W + 2.0 B A on every LoRA-targeted weight, done shard by shard on the CPU
|
| 214 |
+
from the safetensors files in fp32 and stored in bf16: 248 of 248 LoRA pairs merged, none unmatched, in 35 s.
|
| 215 |
+
Every other tensor is the base's, bit for bit. Checked after merging: the merged weights load with no missing or
|
| 216 |
+
unexpected keys, and re-scoring 260 test questions (Jevals PubMedQA, Banking77 and HelpSteer2, Nimble 324;
|
| 217 |
+
repeat 0 in the run's own option order, temperatures applied) on Apple MPS in bf16 gives the same pick as the run's
|
| 218 |
+
records on **257 of 260** (choice 83/84, noul 89/89, score 85/87); the 3 that differ were near ties in the run's own record (top-two margin at most 0.0106), and 1 of them already changed pick between the run's own repeats. The largest probability difference on any
|
| 219 |
+
option is 0.0291 (median 0.0021), which is bf16 across two devices
|
| 220 |
+
(CUDA with the unmerged adapter for the records, MPS with merged weights here). Verification records: `eval/merge-verify.json`.
|
| 221 |
+
|
| 222 |
+
## Limitations
|
| 223 |
+
|
| 224 |
+
- Single-hop judgments only. Split a chain of inference into hops.
|
| 225 |
+
- AINode caps a choice question at 20 options; Banking77's 77 options were scored with an extended
|
| 226 |
+
single-token alphabet for the benchmark only.
|
| 227 |
+
- English data, plus Massive's German subset scoring well by accident of the base model.
|
| 228 |
+
- Calibration was fitted on the training distribution. Refit the temperatures on your own data before
|
| 229 |
+
trusting a threshold, and remember the served route applies none yet.
|
| 230 |
+
|
| 231 |
+
## Versioning and license
|
| 232 |
+
|
| 233 |
+
v2 is frozen. Later versions land as `jebadiah-<size>-v<N>` repositories; [v1](https://huggingface.co/frontier-infra/jebadiah-9b-v1) stays
|
| 234 |
+
as it is. The weights are Apache-2.0, the base model `Qwen/Qwen3.5-9B` is Apache-2.0. Evaluation data: Jevals suite 0.1.0 (CC-BY-4.0, "Jevals (jevals.com),
|
| 235 |
+
release 2026-09-18"), Nimble public subsets (Bespoke Labs), Kev test sets and typed-decisions test,
|
| 236 |
+
each under its own license.
|
eval/merge-verify.json
ADDED
|
@@ -0,0 +1,2700 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"summary": {
|
| 3 |
+
"merged": "jebadiah-9b-v2",
|
| 4 |
+
"run": "9b-chat-v1",
|
| 5 |
+
"device": "mps",
|
| 6 |
+
"dtype": "bf16 weights, fp32 candidate head",
|
| 7 |
+
"temperatures": {
|
| 8 |
+
"choice": 1.1863,
|
| 9 |
+
"noul": 1.0903,
|
| 10 |
+
"score": 1.2162
|
| 11 |
+
},
|
| 12 |
+
"questions": 260,
|
| 13 |
+
"identical_picks": 257,
|
| 14 |
+
"by_type": {
|
| 15 |
+
"choice": {
|
| 16 |
+
"identical": 83,
|
| 17 |
+
"questions": 84
|
| 18 |
+
},
|
| 19 |
+
"noul": {
|
| 20 |
+
"identical": 89,
|
| 21 |
+
"questions": 89
|
| 22 |
+
},
|
| 23 |
+
"score": {
|
| 24 |
+
"identical": 85,
|
| 25 |
+
"questions": 87
|
| 26 |
+
}
|
| 27 |
+
},
|
| 28 |
+
"by_set": {
|
| 29 |
+
"jevals-pubmedqa": 80,
|
| 30 |
+
"jevals-banking77": 70,
|
| 31 |
+
"jevals-helpsteer2": 80,
|
| 32 |
+
"nimble-eval": 30
|
| 33 |
+
},
|
| 34 |
+
"max_abs_prob_diff": {
|
| 35 |
+
"median": 0.002068277301788324,
|
| 36 |
+
"p95": 0.012686089736938522,
|
| 37 |
+
"max": 0.029109271339416543
|
| 38 |
+
},
|
| 39 |
+
"differing": [
|
| 40 |
+
{
|
| 41 |
+
"set": "jevals-banking77",
|
| 42 |
+
"id": "jevals-banking77:banking77-882",
|
| 43 |
+
"qid": "decision",
|
| 44 |
+
"type": "choice",
|
| 45 |
+
"pick": "supported_cards_and_currencies",
|
| 46 |
+
"pick_run": "fiat_currency_support",
|
| 47 |
+
"same": false,
|
| 48 |
+
"max_abs_dp": 0.01041107138633729,
|
| 49 |
+
"run_top2_margin": 0.010566,
|
| 50 |
+
"run_picks_over_repeats": [
|
| 51 |
+
"fiat_currency_support",
|
| 52 |
+
"fiat_currency_support"
|
| 53 |
+
]
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"set": "jevals-helpsteer2",
|
| 57 |
+
"id": "jevals-helpsteer2:helpsteer2-972",
|
| 58 |
+
"qid": "decision",
|
| 59 |
+
"type": "score",
|
| 60 |
+
"pick": "3",
|
| 61 |
+
"pick_run": "2",
|
| 62 |
+
"same": false,
|
| 63 |
+
"max_abs_dp": 0.007170431488037116,
|
| 64 |
+
"run_top2_margin": 0.00078,
|
| 65 |
+
"run_picks_over_repeats": [
|
| 66 |
+
"2",
|
| 67 |
+
"2",
|
| 68 |
+
"2",
|
| 69 |
+
"2",
|
| 70 |
+
"2"
|
| 71 |
+
]
|
| 72 |
+
},
|
| 73 |
+
{
|
| 74 |
+
"set": "jevals-helpsteer2",
|
| 75 |
+
"id": "jevals-helpsteer2:helpsteer2-945",
|
| 76 |
+
"qid": "decision",
|
| 77 |
+
"type": "score",
|
| 78 |
+
"pick": "2",
|
| 79 |
+
"pick_run": "3",
|
| 80 |
+
"same": false,
|
| 81 |
+
"max_abs_dp": 0.003689604001998914,
|
| 82 |
+
"run_top2_margin": 0.000376,
|
| 83 |
+
"run_picks_over_repeats": [
|
| 84 |
+
"3",
|
| 85 |
+
"3",
|
| 86 |
+
"3",
|
| 87 |
+
"3",
|
| 88 |
+
"2"
|
| 89 |
+
]
|
| 90 |
+
}
|
| 91 |
+
],
|
| 92 |
+
"load_s": 42.8,
|
| 93 |
+
"total_s": 702.9,
|
| 94 |
+
"reference": "the run's eval-adapter rows, repeat 0, order seed 0 (CUDA A100, bf16, PEFT adapter unmerged)",
|
| 95 |
+
"differing_max_run_margin": 0.010566,
|
| 96 |
+
"differing_flipped_in_run_repeats": 1
|
| 97 |
+
},
|
| 98 |
+
"rows": [
|
| 99 |
+
{
|
| 100 |
+
"set": "jevals-pubmedqa",
|
| 101 |
+
"id": "jevals-pubmedqa:pubmedqa-577",
|
| 102 |
+
"qid": "decision",
|
| 103 |
+
"type": "noul",
|
| 104 |
+
"pick": "false",
|
| 105 |
+
"pick_run": "false",
|
| 106 |
+
"same": true,
|
| 107 |
+
"max_abs_dp": 0.002361614906311038
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"set": "jevals-pubmedqa",
|
| 111 |
+
"id": "jevals-pubmedqa:pubmedqa-229",
|
| 112 |
+
"qid": "decision",
|
| 113 |
+
"type": "noul",
|
| 114 |
+
"pick": "false",
|
| 115 |
+
"pick_run": "false",
|
| 116 |
+
"same": true,
|
| 117 |
+
"max_abs_dp": 0.0007400083885192865
|
| 118 |
+
},
|
| 119 |
+
{
|
| 120 |
+
"set": "jevals-pubmedqa",
|
| 121 |
+
"id": "jevals-pubmedqa:pubmedqa-874",
|
| 122 |
+
"qid": "decision",
|
| 123 |
+
"type": "noul",
|
| 124 |
+
"pick": "true",
|
| 125 |
+
"pick_run": "true",
|
| 126 |
+
"same": true,
|
| 127 |
+
"max_abs_dp": 0.0007248036618232728
|
| 128 |
+
},
|
| 129 |
+
{
|
| 130 |
+
"set": "jevals-pubmedqa",
|
| 131 |
+
"id": "jevals-pubmedqa:pubmedqa-214",
|
| 132 |
+
"qid": "decision",
|
| 133 |
+
"type": "noul",
|
| 134 |
+
"pick": "true",
|
| 135 |
+
"pick_run": "true",
|
| 136 |
+
"same": true,
|
| 137 |
+
"max_abs_dp": 8.721527481081459e-05
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"set": "jevals-pubmedqa",
|
| 141 |
+
"id": "jevals-pubmedqa:pubmedqa-758",
|
| 142 |
+
"qid": "decision",
|
| 143 |
+
"type": "noul",
|
| 144 |
+
"pick": "true",
|
| 145 |
+
"pick_run": "true",
|
| 146 |
+
"same": true,
|
| 147 |
+
"max_abs_dp": 0.00020452411460876452
|
| 148 |
+
},
|
| 149 |
+
{
|
| 150 |
+
"set": "jevals-pubmedqa",
|
| 151 |
+
"id": "jevals-pubmedqa:pubmedqa-967",
|
| 152 |
+
"qid": "decision",
|
| 153 |
+
"type": "noul",
|
| 154 |
+
"pick": "false",
|
| 155 |
+
"pick_run": "false",
|
| 156 |
+
"same": true,
|
| 157 |
+
"max_abs_dp": 0.0011608515167236266
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"set": "jevals-pubmedqa",
|
| 161 |
+
"id": "jevals-pubmedqa:pubmedqa-405",
|
| 162 |
+
"qid": "decision",
|
| 163 |
+
"type": "noul",
|
| 164 |
+
"pick": "true",
|
| 165 |
+
"pick_run": "true",
|
| 166 |
+
"same": true,
|
| 167 |
+
"max_abs_dp": 0.000412434085845903
|
| 168 |
+
},
|
| 169 |
+
{
|
| 170 |
+
"set": "jevals-pubmedqa",
|
| 171 |
+
"id": "jevals-pubmedqa:pubmedqa-970",
|
| 172 |
+
"qid": "decision",
|
| 173 |
+
"type": "noul",
|
| 174 |
+
"pick": "true",
|
| 175 |
+
"pick_run": "true",
|
| 176 |
+
"same": true,
|
| 177 |
+
"max_abs_dp": 4.357640852034085e-05
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"set": "jevals-pubmedqa",
|
| 181 |
+
"id": "jevals-pubmedqa:pubmedqa-842",
|
| 182 |
+
"qid": "decision",
|
| 183 |
+
"type": "noul",
|
| 184 |
+
"pick": "true",
|
| 185 |
+
"pick_run": "true",
|
| 186 |
+
"same": true,
|
| 187 |
+
"max_abs_dp": 0.00011573467636105672
|
| 188 |
+
},
|
| 189 |
+
{
|
| 190 |
+
"set": "jevals-pubmedqa",
|
| 191 |
+
"id": "jevals-pubmedqa:pubmedqa-697",
|
| 192 |
+
"qid": "decision",
|
| 193 |
+
"type": "noul",
|
| 194 |
+
"pick": "true",
|
| 195 |
+
"pick_run": "true",
|
| 196 |
+
"same": true,
|
| 197 |
+
"max_abs_dp": 0.002668783897876739
|
| 198 |
+
},
|
| 199 |
+
{
|
| 200 |
+
"set": "jevals-pubmedqa",
|
| 201 |
+
"id": "jevals-pubmedqa:pubmedqa-130",
|
| 202 |
+
"qid": "decision",
|
| 203 |
+
"type": "noul",
|
| 204 |
+
"pick": "true",
|
| 205 |
+
"pick_run": "true",
|
| 206 |
+
"same": true,
|
| 207 |
+
"max_abs_dp": 0.0007475724792480598
|
| 208 |
+
},
|
| 209 |
+
{
|
| 210 |
+
"set": "jevals-pubmedqa",
|
| 211 |
+
"id": "jevals-pubmedqa:pubmedqa-880",
|
| 212 |
+
"qid": "decision",
|
| 213 |
+
"type": "noul",
|
| 214 |
+
"pick": "true",
|
| 215 |
+
"pick_run": "true",
|
| 216 |
+
"same": true,
|
| 217 |
+
"max_abs_dp": 0.0019318389310836737
|
| 218 |
+
},
|
| 219 |
+
{
|
| 220 |
+
"set": "jevals-pubmedqa",
|
| 221 |
+
"id": "jevals-pubmedqa:pubmedqa-463",
|
| 222 |
+
"qid": "decision",
|
| 223 |
+
"type": "noul",
|
| 224 |
+
"pick": "true",
|
| 225 |
+
"pick_run": "true",
|
| 226 |
+
"same": true,
|
| 227 |
+
"max_abs_dp": 0.0015171572542190537
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"set": "jevals-pubmedqa",
|
| 231 |
+
"id": "jevals-pubmedqa:pubmedqa-16",
|
| 232 |
+
"qid": "decision",
|
| 233 |
+
"type": "noul",
|
| 234 |
+
"pick": "true",
|
| 235 |
+
"pick_run": "true",
|
| 236 |
+
"same": true,
|
| 237 |
+
"max_abs_dp": 0.008817474800109859
|
| 238 |
+
},
|
| 239 |
+
{
|
| 240 |
+
"set": "jevals-pubmedqa",
|
| 241 |
+
"id": "jevals-pubmedqa:pubmedqa-531",
|
| 242 |
+
"qid": "decision",
|
| 243 |
+
"type": "noul",
|
| 244 |
+
"pick": "true",
|
| 245 |
+
"pick_run": "true",
|
| 246 |
+
"same": true,
|
| 247 |
+
"max_abs_dp": 0.0028381280999183606
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"set": "jevals-pubmedqa",
|
| 251 |
+
"id": "jevals-pubmedqa:pubmedqa-604",
|
| 252 |
+
"qid": "decision",
|
| 253 |
+
"type": "noul",
|
| 254 |
+
"pick": "false",
|
| 255 |
+
"pick_run": "false",
|
| 256 |
+
"same": true,
|
| 257 |
+
"max_abs_dp": 0.004145436416626014
|
| 258 |
+
},
|
| 259 |
+
{
|
| 260 |
+
"set": "jevals-pubmedqa",
|
| 261 |
+
"id": "jevals-pubmedqa:pubmedqa-409",
|
| 262 |
+
"qid": "decision",
|
| 263 |
+
"type": "noul",
|
| 264 |
+
"pick": "true",
|
| 265 |
+
"pick_run": "true",
|
| 266 |
+
"same": true,
|
| 267 |
+
"max_abs_dp": 0.0076537508029937695
|
| 268 |
+
},
|
| 269 |
+
{
|
| 270 |
+
"set": "jevals-pubmedqa",
|
| 271 |
+
"id": "jevals-pubmedqa:pubmedqa-388",
|
| 272 |
+
"qid": "decision",
|
| 273 |
+
"type": "noul",
|
| 274 |
+
"pick": "true",
|
| 275 |
+
"pick_run": "true",
|
| 276 |
+
"same": true,
|
| 277 |
+
"max_abs_dp": 3.363730621341876e-05
|
| 278 |
+
},
|
| 279 |
+
{
|
| 280 |
+
"set": "jevals-pubmedqa",
|
| 281 |
+
"id": "jevals-pubmedqa:pubmedqa-811",
|
| 282 |
+
"qid": "decision",
|
| 283 |
+
"type": "noul",
|
| 284 |
+
"pick": "true",
|
| 285 |
+
"pick_run": "true",
|
| 286 |
+
"same": true,
|
| 287 |
+
"max_abs_dp": 0.002041696565628026
|
| 288 |
+
},
|
| 289 |
+
{
|
| 290 |
+
"set": "jevals-pubmedqa",
|
| 291 |
+
"id": "jevals-pubmedqa:pubmedqa-991",
|
| 292 |
+
"qid": "decision",
|
| 293 |
+
"type": "noul",
|
| 294 |
+
"pick": "true",
|
| 295 |
+
"pick_run": "true",
|
| 296 |
+
"same": true,
|
| 297 |
+
"max_abs_dp": 0.00605180963134766
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"set": "jevals-pubmedqa",
|
| 301 |
+
"id": "jevals-pubmedqa:pubmedqa-11",
|
| 302 |
+
"qid": "decision",
|
| 303 |
+
"type": "noul",
|
| 304 |
+
"pick": "false",
|
| 305 |
+
"pick_run": "false",
|
| 306 |
+
"same": true,
|
| 307 |
+
"max_abs_dp": 0.003769225850105279
|
| 308 |
+
},
|
| 309 |
+
{
|
| 310 |
+
"set": "jevals-pubmedqa",
|
| 311 |
+
"id": "jevals-pubmedqa:pubmedqa-192",
|
| 312 |
+
"qid": "decision",
|
| 313 |
+
"type": "noul",
|
| 314 |
+
"pick": "true",
|
| 315 |
+
"pick_run": "true",
|
| 316 |
+
"same": true,
|
| 317 |
+
"max_abs_dp": 0.004384308416366556
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"set": "jevals-pubmedqa",
|
| 321 |
+
"id": "jevals-pubmedqa:pubmedqa-337",
|
| 322 |
+
"qid": "decision",
|
| 323 |
+
"type": "noul",
|
| 324 |
+
"pick": "true",
|
| 325 |
+
"pick_run": "true",
|
| 326 |
+
"same": true,
|
| 327 |
+
"max_abs_dp": 0.011962067066192628
|
| 328 |
+
},
|
| 329 |
+
{
|
| 330 |
+
"set": "jevals-pubmedqa",
|
| 331 |
+
"id": "jevals-pubmedqa:pubmedqa-867",
|
| 332 |
+
"qid": "decision",
|
| 333 |
+
"type": "noul",
|
| 334 |
+
"pick": "false",
|
| 335 |
+
"pick_run": "false",
|
| 336 |
+
"same": true,
|
| 337 |
+
"max_abs_dp": 0.000688766813755036
|
| 338 |
+
},
|
| 339 |
+
{
|
| 340 |
+
"set": "jevals-pubmedqa",
|
| 341 |
+
"id": "jevals-pubmedqa:pubmedqa-441",
|
| 342 |
+
"qid": "decision",
|
| 343 |
+
"type": "noul",
|
| 344 |
+
"pick": "false",
|
| 345 |
+
"pick_run": "false",
|
| 346 |
+
"same": true,
|
| 347 |
+
"max_abs_dp": 0.0005272247047424239
|
| 348 |
+
},
|
| 349 |
+
{
|
| 350 |
+
"set": "jevals-pubmedqa",
|
| 351 |
+
"id": "jevals-pubmedqa:pubmedqa-980",
|
| 352 |
+
"qid": "decision",
|
| 353 |
+
"type": "noul",
|
| 354 |
+
"pick": "true",
|
| 355 |
+
"pick_run": "true",
|
| 356 |
+
"same": true,
|
| 357 |
+
"max_abs_dp": 0.0003238794937133749
|
| 358 |
+
},
|
| 359 |
+
{
|
| 360 |
+
"set": "jevals-pubmedqa",
|
| 361 |
+
"id": "jevals-pubmedqa:pubmedqa-736",
|
| 362 |
+
"qid": "decision",
|
| 363 |
+
"type": "noul",
|
| 364 |
+
"pick": "true",
|
| 365 |
+
"pick_run": "true",
|
| 366 |
+
"same": true,
|
| 367 |
+
"max_abs_dp": 0.0025702828493118224
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"set": "jevals-pubmedqa",
|
| 371 |
+
"id": "jevals-pubmedqa:pubmedqa-569",
|
| 372 |
+
"qid": "decision",
|
| 373 |
+
"type": "noul",
|
| 374 |
+
"pick": "true",
|
| 375 |
+
"pick_run": "true",
|
| 376 |
+
"same": true,
|
| 377 |
+
"max_abs_dp": 0.00014876399350166256
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"set": "jevals-pubmedqa",
|
| 381 |
+
"id": "jevals-pubmedqa:pubmedqa-359",
|
| 382 |
+
"qid": "decision",
|
| 383 |
+
"type": "noul",
|
| 384 |
+
"pick": "false",
|
| 385 |
+
"pick_run": "false",
|
| 386 |
+
"same": true,
|
| 387 |
+
"max_abs_dp": 0.0013036476001739433
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"set": "jevals-pubmedqa",
|
| 391 |
+
"id": "jevals-pubmedqa:pubmedqa-401",
|
| 392 |
+
"qid": "decision",
|
| 393 |
+
"type": "noul",
|
| 394 |
+
"pick": "true",
|
| 395 |
+
"pick_run": "true",
|
| 396 |
+
"same": true,
|
| 397 |
+
"max_abs_dp": 0.0001761184187531479
|
| 398 |
+
},
|
| 399 |
+
{
|
| 400 |
+
"set": "jevals-pubmedqa",
|
| 401 |
+
"id": "jevals-pubmedqa:pubmedqa-754",
|
| 402 |
+
"qid": "decision",
|
| 403 |
+
"type": "noul",
|
| 404 |
+
"pick": "false",
|
| 405 |
+
"pick_run": "false",
|
| 406 |
+
"same": true,
|
| 407 |
+
"max_abs_dp": 0.003703943653106667
|
| 408 |
+
},
|
| 409 |
+
{
|
| 410 |
+
"set": "jevals-pubmedqa",
|
| 411 |
+
"id": "jevals-pubmedqa:pubmedqa-858",
|
| 412 |
+
"qid": "decision",
|
| 413 |
+
"type": "noul",
|
| 414 |
+
"pick": "false",
|
| 415 |
+
"pick_run": "false",
|
| 416 |
+
"same": true,
|
| 417 |
+
"max_abs_dp": 0.001998045045375818
|
| 418 |
+
},
|
| 419 |
+
{
|
| 420 |
+
"set": "jevals-pubmedqa",
|
| 421 |
+
"id": "jevals-pubmedqa:pubmedqa-789",
|
| 422 |
+
"qid": "decision",
|
| 423 |
+
"type": "noul",
|
| 424 |
+
"pick": "false",
|
| 425 |
+
"pick_run": "false",
|
| 426 |
+
"same": true,
|
| 427 |
+
"max_abs_dp": 7.641987609863232e-05
|
| 428 |
+
},
|
| 429 |
+
{
|
| 430 |
+
"set": "jevals-pubmedqa",
|
| 431 |
+
"id": "jevals-pubmedqa:pubmedqa-52",
|
| 432 |
+
"qid": "decision",
|
| 433 |
+
"type": "noul",
|
| 434 |
+
"pick": "false",
|
| 435 |
+
"pick_run": "false",
|
| 436 |
+
"same": true,
|
| 437 |
+
"max_abs_dp": 0.004233029576301561
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"set": "jevals-pubmedqa",
|
| 441 |
+
"id": "jevals-pubmedqa:pubmedqa-119",
|
| 442 |
+
"qid": "decision",
|
| 443 |
+
"type": "noul",
|
| 444 |
+
"pick": "true",
|
| 445 |
+
"pick_run": "true",
|
| 446 |
+
"same": true,
|
| 447 |
+
"max_abs_dp": 0.00014103620529171135
|
| 448 |
+
},
|
| 449 |
+
{
|
| 450 |
+
"set": "jevals-pubmedqa",
|
| 451 |
+
"id": "jevals-pubmedqa:pubmedqa-148",
|
| 452 |
+
"qid": "decision",
|
| 453 |
+
"type": "noul",
|
| 454 |
+
"pick": "true",
|
| 455 |
+
"pick_run": "true",
|
| 456 |
+
"same": true,
|
| 457 |
+
"max_abs_dp": 3.7405700683557086e-06
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"set": "jevals-pubmedqa",
|
| 461 |
+
"id": "jevals-pubmedqa:pubmedqa-519",
|
| 462 |
+
"qid": "decision",
|
| 463 |
+
"type": "noul",
|
| 464 |
+
"pick": "true",
|
| 465 |
+
"pick_run": "true",
|
| 466 |
+
"same": true,
|
| 467 |
+
"max_abs_dp": 3.914723384380349e-05
|
| 468 |
+
},
|
| 469 |
+
{
|
| 470 |
+
"set": "jevals-pubmedqa",
|
| 471 |
+
"id": "jevals-pubmedqa:pubmedqa-812",
|
| 472 |
+
"qid": "decision",
|
| 473 |
+
"type": "noul",
|
| 474 |
+
"pick": "false",
|
| 475 |
+
"pick_run": "false",
|
| 476 |
+
"same": true,
|
| 477 |
+
"max_abs_dp": 1.720184409618325e-05
|
| 478 |
+
},
|
| 479 |
+
{
|
| 480 |
+
"set": "jevals-pubmedqa",
|
| 481 |
+
"id": "jevals-pubmedqa:pubmedqa-276",
|
| 482 |
+
"qid": "decision",
|
| 483 |
+
"type": "noul",
|
| 484 |
+
"pick": "true",
|
| 485 |
+
"pick_run": "true",
|
| 486 |
+
"same": true,
|
| 487 |
+
"max_abs_dp": 0.001692070934534079
|
| 488 |
+
},
|
| 489 |
+
{
|
| 490 |
+
"set": "jevals-pubmedqa",
|
| 491 |
+
"id": "jevals-pubmedqa:pubmedqa-505",
|
| 492 |
+
"qid": "decision",
|
| 493 |
+
"type": "noul",
|
| 494 |
+
"pick": "false",
|
| 495 |
+
"pick_run": "false",
|
| 496 |
+
"same": true,
|
| 497 |
+
"max_abs_dp": 0.029109271339416543
|
| 498 |
+
},
|
| 499 |
+
{
|
| 500 |
+
"set": "jevals-pubmedqa",
|
| 501 |
+
"id": "jevals-pubmedqa:pubmedqa-333",
|
| 502 |
+
"qid": "decision",
|
| 503 |
+
"type": "noul",
|
| 504 |
+
"pick": "true",
|
| 505 |
+
"pick_run": "true",
|
| 506 |
+
"same": true,
|
| 507 |
+
"max_abs_dp": 0.01489011363983156
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"set": "jevals-pubmedqa",
|
| 511 |
+
"id": "jevals-pubmedqa:pubmedqa-746",
|
| 512 |
+
"qid": "decision",
|
| 513 |
+
"type": "noul",
|
| 514 |
+
"pick": "true",
|
| 515 |
+
"pick_run": "true",
|
| 516 |
+
"same": true,
|
| 517 |
+
"max_abs_dp": 0.00012200270804762874
|
| 518 |
+
},
|
| 519 |
+
{
|
| 520 |
+
"set": "jevals-pubmedqa",
|
| 521 |
+
"id": "jevals-pubmedqa:pubmedqa-644",
|
| 522 |
+
"qid": "decision",
|
| 523 |
+
"type": "noul",
|
| 524 |
+
"pick": "false",
|
| 525 |
+
"pick_run": "false",
|
| 526 |
+
"same": true,
|
| 527 |
+
"max_abs_dp": 9.002473068242534e-05
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"set": "jevals-pubmedqa",
|
| 531 |
+
"id": "jevals-pubmedqa:pubmedqa-248",
|
| 532 |
+
"qid": "decision",
|
| 533 |
+
"type": "noul",
|
| 534 |
+
"pick": "true",
|
| 535 |
+
"pick_run": "true",
|
| 536 |
+
"same": true,
|
| 537 |
+
"max_abs_dp": 0.00016415084588527681
|
| 538 |
+
},
|
| 539 |
+
{
|
| 540 |
+
"set": "jevals-pubmedqa",
|
| 541 |
+
"id": "jevals-pubmedqa:pubmedqa-764",
|
| 542 |
+
"qid": "decision",
|
| 543 |
+
"type": "noul",
|
| 544 |
+
"pick": "false",
|
| 545 |
+
"pick_run": "false",
|
| 546 |
+
"same": true,
|
| 547 |
+
"max_abs_dp": 0.00045313702392578425
|
| 548 |
+
},
|
| 549 |
+
{
|
| 550 |
+
"set": "jevals-pubmedqa",
|
| 551 |
+
"id": "jevals-pubmedqa:pubmedqa-711",
|
| 552 |
+
"qid": "decision",
|
| 553 |
+
"type": "noul",
|
| 554 |
+
"pick": "true",
|
| 555 |
+
"pick_run": "true",
|
| 556 |
+
"same": true,
|
| 557 |
+
"max_abs_dp": 0.00029391733694076466
|
| 558 |
+
},
|
| 559 |
+
{
|
| 560 |
+
"set": "jevals-pubmedqa",
|
| 561 |
+
"id": "jevals-pubmedqa:pubmedqa-941",
|
| 562 |
+
"qid": "decision",
|
| 563 |
+
"type": "noul",
|
| 564 |
+
"pick": "true",
|
| 565 |
+
"pick_run": "true",
|
| 566 |
+
"same": true,
|
| 567 |
+
"max_abs_dp": 0.0019045123214721515
|
| 568 |
+
},
|
| 569 |
+
{
|
| 570 |
+
"set": "jevals-pubmedqa",
|
| 571 |
+
"id": "jevals-pubmedqa:pubmedqa-651",
|
| 572 |
+
"qid": "decision",
|
| 573 |
+
"type": "noul",
|
| 574 |
+
"pick": "false",
|
| 575 |
+
"pick_run": "false",
|
| 576 |
+
"same": true,
|
| 577 |
+
"max_abs_dp": 0.0021859156303405958
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"set": "jevals-pubmedqa",
|
| 581 |
+
"id": "jevals-pubmedqa:pubmedqa-846",
|
| 582 |
+
"qid": "decision",
|
| 583 |
+
"type": "noul",
|
| 584 |
+
"pick": "true",
|
| 585 |
+
"pick_run": "true",
|
| 586 |
+
"same": true,
|
| 587 |
+
"max_abs_dp": 0.0003319750289917378
|
| 588 |
+
},
|
| 589 |
+
{
|
| 590 |
+
"set": "jevals-pubmedqa",
|
| 591 |
+
"id": "jevals-pubmedqa:pubmedqa-906",
|
| 592 |
+
"qid": "decision",
|
| 593 |
+
"type": "noul",
|
| 594 |
+
"pick": "false",
|
| 595 |
+
"pick_run": "false",
|
| 596 |
+
"same": true,
|
| 597 |
+
"max_abs_dp": 0.00033138166332245117
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"set": "jevals-pubmedqa",
|
| 601 |
+
"id": "jevals-pubmedqa:pubmedqa-931",
|
| 602 |
+
"qid": "decision",
|
| 603 |
+
"type": "noul",
|
| 604 |
+
"pick": "false",
|
| 605 |
+
"pick_run": "false",
|
| 606 |
+
"same": true,
|
| 607 |
+
"max_abs_dp": 0.013560935886383074
|
| 608 |
+
},
|
| 609 |
+
{
|
| 610 |
+
"set": "jevals-pubmedqa",
|
| 611 |
+
"id": "jevals-pubmedqa:pubmedqa-74",
|
| 612 |
+
"qid": "decision",
|
| 613 |
+
"type": "noul",
|
| 614 |
+
"pick": "true",
|
| 615 |
+
"pick_run": "true",
|
| 616 |
+
"same": true,
|
| 617 |
+
"max_abs_dp": 0.000264806226700544
|
| 618 |
+
},
|
| 619 |
+
{
|
| 620 |
+
"set": "jevals-pubmedqa",
|
| 621 |
+
"id": "jevals-pubmedqa:pubmedqa-58",
|
| 622 |
+
"qid": "decision",
|
| 623 |
+
"type": "noul",
|
| 624 |
+
"pick": "true",
|
| 625 |
+
"pick_run": "true",
|
| 626 |
+
"same": true,
|
| 627 |
+
"max_abs_dp": 0.006453835803985619
|
| 628 |
+
},
|
| 629 |
+
{
|
| 630 |
+
"set": "jevals-pubmedqa",
|
| 631 |
+
"id": "jevals-pubmedqa:pubmedqa-878",
|
| 632 |
+
"qid": "decision",
|
| 633 |
+
"type": "noul",
|
| 634 |
+
"pick": "true",
|
| 635 |
+
"pick_run": "true",
|
| 636 |
+
"same": true,
|
| 637 |
+
"max_abs_dp": 3.280313563346765e-05
|
| 638 |
+
},
|
| 639 |
+
{
|
| 640 |
+
"set": "jevals-pubmedqa",
|
| 641 |
+
"id": "jevals-pubmedqa:pubmedqa-217",
|
| 642 |
+
"qid": "decision",
|
| 643 |
+
"type": "noul",
|
| 644 |
+
"pick": "false",
|
| 645 |
+
"pick_run": "false",
|
| 646 |
+
"same": true,
|
| 647 |
+
"max_abs_dp": 0.007298624530792219
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"set": "jevals-pubmedqa",
|
| 651 |
+
"id": "jevals-pubmedqa:pubmedqa-966",
|
| 652 |
+
"qid": "decision",
|
| 653 |
+
"type": "noul",
|
| 654 |
+
"pick": "false",
|
| 655 |
+
"pick_run": "false",
|
| 656 |
+
"same": true,
|
| 657 |
+
"max_abs_dp": 0.0016663671798705648
|
| 658 |
+
},
|
| 659 |
+
{
|
| 660 |
+
"set": "jevals-pubmedqa",
|
| 661 |
+
"id": "jevals-pubmedqa:pubmedqa-304",
|
| 662 |
+
"qid": "decision",
|
| 663 |
+
"type": "noul",
|
| 664 |
+
"pick": "true",
|
| 665 |
+
"pick_run": "true",
|
| 666 |
+
"same": true,
|
| 667 |
+
"max_abs_dp": 0.0006484546394348589
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"set": "jevals-pubmedqa",
|
| 671 |
+
"id": "jevals-pubmedqa:pubmedqa-60",
|
| 672 |
+
"qid": "decision",
|
| 673 |
+
"type": "noul",
|
| 674 |
+
"pick": "true",
|
| 675 |
+
"pick_run": "true",
|
| 676 |
+
"same": true,
|
| 677 |
+
"max_abs_dp": 0.00080796931374073
|
| 678 |
+
},
|
| 679 |
+
{
|
| 680 |
+
"set": "jevals-pubmedqa",
|
| 681 |
+
"id": "jevals-pubmedqa:pubmedqa-624",
|
| 682 |
+
"qid": "decision",
|
| 683 |
+
"type": "noul",
|
| 684 |
+
"pick": "true",
|
| 685 |
+
"pick_run": "true",
|
| 686 |
+
"same": true,
|
| 687 |
+
"max_abs_dp": 3.0353588104214957e-05
|
| 688 |
+
},
|
| 689 |
+
{
|
| 690 |
+
"set": "jevals-pubmedqa",
|
| 691 |
+
"id": "jevals-pubmedqa:pubmedqa-547",
|
| 692 |
+
"qid": "decision",
|
| 693 |
+
"type": "noul",
|
| 694 |
+
"pick": "true",
|
| 695 |
+
"pick_run": "true",
|
| 696 |
+
"same": true,
|
| 697 |
+
"max_abs_dp": 0.0085917929992676
|
| 698 |
+
},
|
| 699 |
+
{
|
| 700 |
+
"set": "jevals-pubmedqa",
|
| 701 |
+
"id": "jevals-pubmedqa:pubmedqa-374",
|
| 702 |
+
"qid": "decision",
|
| 703 |
+
"type": "noul",
|
| 704 |
+
"pick": "false",
|
| 705 |
+
"pick_run": "false",
|
| 706 |
+
"same": true,
|
| 707 |
+
"max_abs_dp": 0.0004448536033629935
|
| 708 |
+
},
|
| 709 |
+
{
|
| 710 |
+
"set": "jevals-pubmedqa",
|
| 711 |
+
"id": "jevals-pubmedqa:pubmedqa-774",
|
| 712 |
+
"qid": "decision",
|
| 713 |
+
"type": "noul",
|
| 714 |
+
"pick": "true",
|
| 715 |
+
"pick_run": "true",
|
| 716 |
+
"same": true,
|
| 717 |
+
"max_abs_dp": 0.00010372769165034601
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"set": "jevals-pubmedqa",
|
| 721 |
+
"id": "jevals-pubmedqa:pubmedqa-872",
|
| 722 |
+
"qid": "decision",
|
| 723 |
+
"type": "noul",
|
| 724 |
+
"pick": "false",
|
| 725 |
+
"pick_run": "false",
|
| 726 |
+
"same": true,
|
| 727 |
+
"max_abs_dp": 0.002048760414123524
|
| 728 |
+
},
|
| 729 |
+
{
|
| 730 |
+
"set": "jevals-pubmedqa",
|
| 731 |
+
"id": "jevals-pubmedqa:pubmedqa-524",
|
| 732 |
+
"qid": "decision",
|
| 733 |
+
"type": "noul",
|
| 734 |
+
"pick": "true",
|
| 735 |
+
"pick_run": "true",
|
| 736 |
+
"same": true,
|
| 737 |
+
"max_abs_dp": 0.0004191811428070069
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"set": "jevals-pubmedqa",
|
| 741 |
+
"id": "jevals-pubmedqa:pubmedqa-404",
|
| 742 |
+
"qid": "decision",
|
| 743 |
+
"type": "noul",
|
| 744 |
+
"pick": "false",
|
| 745 |
+
"pick_run": "false",
|
| 746 |
+
"same": true,
|
| 747 |
+
"max_abs_dp": 0.00030116800451278694
|
| 748 |
+
},
|
| 749 |
+
{
|
| 750 |
+
"set": "jevals-pubmedqa",
|
| 751 |
+
"id": "jevals-pubmedqa:pubmedqa-829",
|
| 752 |
+
"qid": "decision",
|
| 753 |
+
"type": "noul",
|
| 754 |
+
"pick": "false",
|
| 755 |
+
"pick_run": "false",
|
| 756 |
+
"same": true,
|
| 757 |
+
"max_abs_dp": 7.498416900619809e-06
|
| 758 |
+
},
|
| 759 |
+
{
|
| 760 |
+
"set": "jevals-pubmedqa",
|
| 761 |
+
"id": "jevals-pubmedqa:pubmedqa-386",
|
| 762 |
+
"qid": "decision",
|
| 763 |
+
"type": "noul",
|
| 764 |
+
"pick": "false",
|
| 765 |
+
"pick_run": "false",
|
| 766 |
+
"same": true,
|
| 767 |
+
"max_abs_dp": 0.0005611235961914263
|
| 768 |
+
},
|
| 769 |
+
{
|
| 770 |
+
"set": "jevals-pubmedqa",
|
| 771 |
+
"id": "jevals-pubmedqa:pubmedqa-675",
|
| 772 |
+
"qid": "decision",
|
| 773 |
+
"type": "noul",
|
| 774 |
+
"pick": "true",
|
| 775 |
+
"pick_run": "true",
|
| 776 |
+
"same": true,
|
| 777 |
+
"max_abs_dp": 0.00023019430702924705
|
| 778 |
+
},
|
| 779 |
+
{
|
| 780 |
+
"set": "jevals-pubmedqa",
|
| 781 |
+
"id": "jevals-pubmedqa:pubmedqa-555",
|
| 782 |
+
"qid": "decision",
|
| 783 |
+
"type": "noul",
|
| 784 |
+
"pick": "true",
|
| 785 |
+
"pick_run": "true",
|
| 786 |
+
"same": true,
|
| 787 |
+
"max_abs_dp": 4.606273269658345e-05
|
| 788 |
+
},
|
| 789 |
+
{
|
| 790 |
+
"set": "jevals-pubmedqa",
|
| 791 |
+
"id": "jevals-pubmedqa:pubmedqa-963",
|
| 792 |
+
"qid": "decision",
|
| 793 |
+
"type": "noul",
|
| 794 |
+
"pick": "true",
|
| 795 |
+
"pick_run": "true",
|
| 796 |
+
"same": true,
|
| 797 |
+
"max_abs_dp": 0.0004458992661237719
|
| 798 |
+
},
|
| 799 |
+
{
|
| 800 |
+
"set": "jevals-pubmedqa",
|
| 801 |
+
"id": "jevals-pubmedqa:pubmedqa-314",
|
| 802 |
+
"qid": "decision",
|
| 803 |
+
"type": "noul",
|
| 804 |
+
"pick": "true",
|
| 805 |
+
"pick_run": "true",
|
| 806 |
+
"same": true,
|
| 807 |
+
"max_abs_dp": 0.00030999475097659435
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"set": "jevals-pubmedqa",
|
| 811 |
+
"id": "jevals-pubmedqa:pubmedqa-895",
|
| 812 |
+
"qid": "decision",
|
| 813 |
+
"type": "noul",
|
| 814 |
+
"pick": "false",
|
| 815 |
+
"pick_run": "false",
|
| 816 |
+
"same": true,
|
| 817 |
+
"max_abs_dp": 0.004996557075500507
|
| 818 |
+
},
|
| 819 |
+
{
|
| 820 |
+
"set": "jevals-pubmedqa",
|
| 821 |
+
"id": "jevals-pubmedqa:pubmedqa-342",
|
| 822 |
+
"qid": "decision",
|
| 823 |
+
"type": "noul",
|
| 824 |
+
"pick": "false",
|
| 825 |
+
"pick_run": "false",
|
| 826 |
+
"same": true,
|
| 827 |
+
"max_abs_dp": 0.006861803737640337
|
| 828 |
+
},
|
| 829 |
+
{
|
| 830 |
+
"set": "jevals-pubmedqa",
|
| 831 |
+
"id": "jevals-pubmedqa:pubmedqa-656",
|
| 832 |
+
"qid": "decision",
|
| 833 |
+
"type": "noul",
|
| 834 |
+
"pick": "false",
|
| 835 |
+
"pick_run": "false",
|
| 836 |
+
"same": true,
|
| 837 |
+
"max_abs_dp": 0.0004893547134399157
|
| 838 |
+
},
|
| 839 |
+
{
|
| 840 |
+
"set": "jevals-pubmedqa",
|
| 841 |
+
"id": "jevals-pubmedqa:pubmedqa-79",
|
| 842 |
+
"qid": "decision",
|
| 843 |
+
"type": "noul",
|
| 844 |
+
"pick": "true",
|
| 845 |
+
"pick_run": "true",
|
| 846 |
+
"same": true,
|
| 847 |
+
"max_abs_dp": 2.925255966190754e-05
|
| 848 |
+
},
|
| 849 |
+
{
|
| 850 |
+
"set": "jevals-pubmedqa",
|
| 851 |
+
"id": "jevals-pubmedqa:pubmedqa-715",
|
| 852 |
+
"qid": "decision",
|
| 853 |
+
"type": "noul",
|
| 854 |
+
"pick": "false",
|
| 855 |
+
"pick_run": "false",
|
| 856 |
+
"same": true,
|
| 857 |
+
"max_abs_dp": 0.0007816578140258379
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"set": "jevals-pubmedqa",
|
| 861 |
+
"id": "jevals-pubmedqa:pubmedqa-368",
|
| 862 |
+
"qid": "decision",
|
| 863 |
+
"type": "noul",
|
| 864 |
+
"pick": "false",
|
| 865 |
+
"pick_run": "false",
|
| 866 |
+
"same": true,
|
| 867 |
+
"max_abs_dp": 0.0002648297873735425
|
| 868 |
+
},
|
| 869 |
+
{
|
| 870 |
+
"set": "jevals-pubmedqa",
|
| 871 |
+
"id": "jevals-pubmedqa:pubmedqa-226",
|
| 872 |
+
"qid": "decision",
|
| 873 |
+
"type": "noul",
|
| 874 |
+
"pick": "true",
|
| 875 |
+
"pick_run": "true",
|
| 876 |
+
"same": true,
|
| 877 |
+
"max_abs_dp": 0.000312848361968987
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"set": "jevals-pubmedqa",
|
| 881 |
+
"id": "jevals-pubmedqa:pubmedqa-743",
|
| 882 |
+
"qid": "decision",
|
| 883 |
+
"type": "noul",
|
| 884 |
+
"pick": "false",
|
| 885 |
+
"pick_run": "false",
|
| 886 |
+
"same": true,
|
| 887 |
+
"max_abs_dp": 0.0001364775619506764
|
| 888 |
+
},
|
| 889 |
+
{
|
| 890 |
+
"set": "jevals-pubmedqa",
|
| 891 |
+
"id": "jevals-pubmedqa:pubmedqa-871",
|
| 892 |
+
"qid": "decision",
|
| 893 |
+
"type": "noul",
|
| 894 |
+
"pick": "false",
|
| 895 |
+
"pick_run": "false",
|
| 896 |
+
"same": true,
|
| 897 |
+
"max_abs_dp": 0.002156544670104954
|
| 898 |
+
},
|
| 899 |
+
{
|
| 900 |
+
"set": "jevals-banking77",
|
| 901 |
+
"id": "jevals-banking77:banking77-1749",
|
| 902 |
+
"qid": "decision",
|
| 903 |
+
"type": "choice",
|
| 904 |
+
"pick": "failed_transfer",
|
| 905 |
+
"pick_run": "failed_transfer",
|
| 906 |
+
"same": true,
|
| 907 |
+
"max_abs_dp": 0.023940279014587396
|
| 908 |
+
},
|
| 909 |
+
{
|
| 910 |
+
"set": "jevals-banking77",
|
| 911 |
+
"id": "jevals-banking77:banking77-669",
|
| 912 |
+
"qid": "decision",
|
| 913 |
+
"type": "choice",
|
| 914 |
+
"pick": "pending_top_up",
|
| 915 |
+
"pick_run": "pending_top_up",
|
| 916 |
+
"same": true,
|
| 917 |
+
"max_abs_dp": 0.007160110675811793
|
| 918 |
+
},
|
| 919 |
+
{
|
| 920 |
+
"set": "jevals-banking77",
|
| 921 |
+
"id": "jevals-banking77:banking77-2687",
|
| 922 |
+
"qid": "decision",
|
| 923 |
+
"type": "choice",
|
| 924 |
+
"pick": "transfer_timing",
|
| 925 |
+
"pick_run": "transfer_timing",
|
| 926 |
+
"same": true,
|
| 927 |
+
"max_abs_dp": 0.0005568832015990788
|
| 928 |
+
},
|
| 929 |
+
{
|
| 930 |
+
"set": "jevals-banking77",
|
| 931 |
+
"id": "jevals-banking77:banking77-610",
|
| 932 |
+
"qid": "decision",
|
| 933 |
+
"type": "choice",
|
| 934 |
+
"pick": "receiving_money",
|
| 935 |
+
"pick_run": "receiving_money",
|
| 936 |
+
"same": true,
|
| 937 |
+
"max_abs_dp": 0.005873244781494091
|
| 938 |
+
},
|
| 939 |
+
{
|
| 940 |
+
"set": "jevals-banking77",
|
| 941 |
+
"id": "jevals-banking77:banking77-2279",
|
| 942 |
+
"qid": "decision",
|
| 943 |
+
"type": "choice",
|
| 944 |
+
"pick": "failed_transfer",
|
| 945 |
+
"pick_run": "failed_transfer",
|
| 946 |
+
"same": true,
|
| 947 |
+
"max_abs_dp": 0.003507100162506105
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"set": "jevals-banking77",
|
| 951 |
+
"id": "jevals-banking77:banking77-2981",
|
| 952 |
+
"qid": "decision",
|
| 953 |
+
"type": "choice",
|
| 954 |
+
"pick": "topping_up_by_card",
|
| 955 |
+
"pick_run": "topping_up_by_card",
|
| 956 |
+
"same": true,
|
| 957 |
+
"max_abs_dp": 0.004674171613693212
|
| 958 |
+
},
|
| 959 |
+
{
|
| 960 |
+
"set": "jevals-banking77",
|
| 961 |
+
"id": "jevals-banking77:banking77-1242",
|
| 962 |
+
"qid": "decision",
|
| 963 |
+
"type": "choice",
|
| 964 |
+
"pick": "passcode_forgotten",
|
| 965 |
+
"pick_run": "passcode_forgotten",
|
| 966 |
+
"same": true,
|
| 967 |
+
"max_abs_dp": 0.0011834933509826673
|
| 968 |
+
},
|
| 969 |
+
{
|
| 970 |
+
"set": "jevals-banking77",
|
| 971 |
+
"id": "jevals-banking77:banking77-2986",
|
| 972 |
+
"qid": "decision",
|
| 973 |
+
"type": "choice",
|
| 974 |
+
"pick": "apple_pay_or_google_pay",
|
| 975 |
+
"pick_run": "apple_pay_or_google_pay",
|
| 976 |
+
"same": true,
|
| 977 |
+
"max_abs_dp": 0.011354429265022281
|
| 978 |
+
},
|
| 979 |
+
{
|
| 980 |
+
"set": "jevals-banking77",
|
| 981 |
+
"id": "jevals-banking77:banking77-2557",
|
| 982 |
+
"qid": "decision",
|
| 983 |
+
"type": "choice",
|
| 984 |
+
"pick": "wrong_exchange_rate_for_cash_withdrawal",
|
| 985 |
+
"pick_run": "wrong_exchange_rate_for_cash_withdrawal",
|
| 986 |
+
"same": true,
|
| 987 |
+
"max_abs_dp": 0.0010959678363799963
|
| 988 |
+
},
|
| 989 |
+
{
|
| 990 |
+
"set": "jevals-banking77",
|
| 991 |
+
"id": "jevals-banking77:banking77-2076",
|
| 992 |
+
"qid": "decision",
|
| 993 |
+
"type": "choice",
|
| 994 |
+
"pick": "card_not_working",
|
| 995 |
+
"pick_run": "card_not_working",
|
| 996 |
+
"same": true,
|
| 997 |
+
"max_abs_dp": 0.00905289724349978
|
| 998 |
+
},
|
| 999 |
+
{
|
| 1000 |
+
"set": "jevals-banking77",
|
| 1001 |
+
"id": "jevals-banking77:banking77-437",
|
| 1002 |
+
"qid": "decision",
|
| 1003 |
+
"type": "choice",
|
| 1004 |
+
"pick": "fiat_currency_support",
|
| 1005 |
+
"pick_run": "fiat_currency_support",
|
| 1006 |
+
"same": true,
|
| 1007 |
+
"max_abs_dp": 0.0004598371599912644
|
| 1008 |
+
},
|
| 1009 |
+
{
|
| 1010 |
+
"set": "jevals-banking77",
|
| 1011 |
+
"id": "jevals-banking77:banking77-2717",
|
| 1012 |
+
"qid": "decision",
|
| 1013 |
+
"type": "choice",
|
| 1014 |
+
"pick": "cash_withdrawal_not_recognised",
|
| 1015 |
+
"pick_run": "cash_withdrawal_not_recognised",
|
| 1016 |
+
"same": true,
|
| 1017 |
+
"max_abs_dp": 0.0032614902958869874
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"set": "jevals-banking77",
|
| 1021 |
+
"id": "jevals-banking77:banking77-1407",
|
| 1022 |
+
"qid": "decision",
|
| 1023 |
+
"type": "choice",
|
| 1024 |
+
"pick": "compromised_card",
|
| 1025 |
+
"pick_run": "compromised_card",
|
| 1026 |
+
"same": true,
|
| 1027 |
+
"max_abs_dp": 0.0003835960665047166
|
| 1028 |
+
},
|
| 1029 |
+
{
|
| 1030 |
+
"set": "jevals-banking77",
|
| 1031 |
+
"id": "jevals-banking77:banking77-51",
|
| 1032 |
+
"qid": "decision",
|
| 1033 |
+
"type": "choice",
|
| 1034 |
+
"pick": "card_linking",
|
| 1035 |
+
"pick_run": "card_linking",
|
| 1036 |
+
"same": true,
|
| 1037 |
+
"max_abs_dp": 0.0003765650672912235
|
| 1038 |
+
},
|
| 1039 |
+
{
|
| 1040 |
+
"set": "jevals-banking77",
|
| 1041 |
+
"id": "jevals-banking77:banking77-1566",
|
| 1042 |
+
"qid": "decision",
|
| 1043 |
+
"type": "choice",
|
| 1044 |
+
"pick": "declined_cash_withdrawal",
|
| 1045 |
+
"pick_run": "declined_cash_withdrawal",
|
| 1046 |
+
"same": true,
|
| 1047 |
+
"max_abs_dp": 0.007276423375129709
|
| 1048 |
+
},
|
| 1049 |
+
{
|
| 1050 |
+
"set": "jevals-banking77",
|
| 1051 |
+
"id": "jevals-banking77:banking77-1817",
|
| 1052 |
+
"qid": "decision",
|
| 1053 |
+
"type": "choice",
|
| 1054 |
+
"pick": "declined_card_payment",
|
| 1055 |
+
"pick_run": "declined_card_payment",
|
| 1056 |
+
"same": true,
|
| 1057 |
+
"max_abs_dp": 0.008562368899822231
|
| 1058 |
+
},
|
| 1059 |
+
{
|
| 1060 |
+
"set": "jevals-banking77",
|
| 1061 |
+
"id": "jevals-banking77:banking77-1244",
|
| 1062 |
+
"qid": "decision",
|
| 1063 |
+
"type": "choice",
|
| 1064 |
+
"pick": "passcode_forgotten",
|
| 1065 |
+
"pick_run": "passcode_forgotten",
|
| 1066 |
+
"same": true,
|
| 1067 |
+
"max_abs_dp": 0.021577001438140897
|
| 1068 |
+
},
|
| 1069 |
+
{
|
| 1070 |
+
"set": "jevals-banking77",
|
| 1071 |
+
"id": "jevals-banking77:banking77-1169",
|
| 1072 |
+
"qid": "decision",
|
| 1073 |
+
"type": "choice",
|
| 1074 |
+
"pick": "unable_to_verify_identity",
|
| 1075 |
+
"pick_run": "unable_to_verify_identity",
|
| 1076 |
+
"same": true,
|
| 1077 |
+
"max_abs_dp": 0.0033154890518188873
|
| 1078 |
+
},
|
| 1079 |
+
{
|
| 1080 |
+
"set": "jevals-banking77",
|
| 1081 |
+
"id": "jevals-banking77:banking77-2430",
|
| 1082 |
+
"qid": "decision",
|
| 1083 |
+
"type": "choice",
|
| 1084 |
+
"pick": "top_up_by_card_charge",
|
| 1085 |
+
"pick_run": "top_up_by_card_charge",
|
| 1086 |
+
"same": true,
|
| 1087 |
+
"max_abs_dp": 0.015199531869888316
|
| 1088 |
+
},
|
| 1089 |
+
{
|
| 1090 |
+
"set": "jevals-banking77",
|
| 1091 |
+
"id": "jevals-banking77:banking77-3073",
|
| 1092 |
+
"qid": "decision",
|
| 1093 |
+
"type": "choice",
|
| 1094 |
+
"pick": "country_support",
|
| 1095 |
+
"pick_run": "country_support",
|
| 1096 |
+
"same": true,
|
| 1097 |
+
"max_abs_dp": 0.00039297395324711104
|
| 1098 |
+
},
|
| 1099 |
+
{
|
| 1100 |
+
"set": "jevals-banking77",
|
| 1101 |
+
"id": "jevals-banking77:banking77-10",
|
| 1102 |
+
"qid": "decision",
|
| 1103 |
+
"type": "choice",
|
| 1104 |
+
"pick": "lost_or_stolen_card",
|
| 1105 |
+
"pick_run": "lost_or_stolen_card",
|
| 1106 |
+
"same": true,
|
| 1107 |
+
"max_abs_dp": 0.018477167479515083
|
| 1108 |
+
},
|
| 1109 |
+
{
|
| 1110 |
+
"set": "jevals-banking77",
|
| 1111 |
+
"id": "jevals-banking77:banking77-578",
|
| 1112 |
+
"qid": "decision",
|
| 1113 |
+
"type": "choice",
|
| 1114 |
+
"pick": "contactless_not_working",
|
| 1115 |
+
"pick_run": "contactless_not_working",
|
| 1116 |
+
"same": true,
|
| 1117 |
+
"max_abs_dp": 0.001786226730346674
|
| 1118 |
+
},
|
| 1119 |
+
{
|
| 1120 |
+
"set": "jevals-banking77",
|
| 1121 |
+
"id": "jevals-banking77:banking77-1000",
|
| 1122 |
+
"qid": "decision",
|
| 1123 |
+
"type": "choice",
|
| 1124 |
+
"pick": "top_up_reverted",
|
| 1125 |
+
"pick_run": "top_up_reverted",
|
| 1126 |
+
"same": true,
|
| 1127 |
+
"max_abs_dp": 0.004602908008575435
|
| 1128 |
+
},
|
| 1129 |
+
{
|
| 1130 |
+
"set": "jevals-banking77",
|
| 1131 |
+
"id": "jevals-banking77:banking77-2650",
|
| 1132 |
+
"qid": "decision",
|
| 1133 |
+
"type": "choice",
|
| 1134 |
+
"pick": "top_up_failed",
|
| 1135 |
+
"pick_run": "top_up_failed",
|
| 1136 |
+
"same": true,
|
| 1137 |
+
"max_abs_dp": 0.0017825746154784827
|
| 1138 |
+
},
|
| 1139 |
+
{
|
| 1140 |
+
"set": "jevals-banking77",
|
| 1141 |
+
"id": "jevals-banking77:banking77-1341",
|
| 1142 |
+
"qid": "decision",
|
| 1143 |
+
"type": "choice",
|
| 1144 |
+
"pick": "topping_up_by_card",
|
| 1145 |
+
"pick_run": "topping_up_by_card",
|
| 1146 |
+
"same": true,
|
| 1147 |
+
"max_abs_dp": 0.0066273999137878015
|
| 1148 |
+
},
|
| 1149 |
+
{
|
| 1150 |
+
"set": "jevals-banking77",
|
| 1151 |
+
"id": "jevals-banking77:banking77-3019",
|
| 1152 |
+
"qid": "decision",
|
| 1153 |
+
"type": "choice",
|
| 1154 |
+
"pick": "verify_my_identity",
|
| 1155 |
+
"pick_run": "verify_my_identity",
|
| 1156 |
+
"same": true,
|
| 1157 |
+
"max_abs_dp": 0.01950413126373296
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"set": "jevals-banking77",
|
| 1161 |
+
"id": "jevals-banking77:banking77-2199",
|
| 1162 |
+
"qid": "decision",
|
| 1163 |
+
"type": "choice",
|
| 1164 |
+
"pick": "transfer_fee_charged",
|
| 1165 |
+
"pick_run": "transfer_fee_charged",
|
| 1166 |
+
"same": true,
|
| 1167 |
+
"max_abs_dp": 0.0006088438873290691
|
| 1168 |
+
},
|
| 1169 |
+
{
|
| 1170 |
+
"set": "jevals-banking77",
|
| 1171 |
+
"id": "jevals-banking77:banking77-1719",
|
| 1172 |
+
"qid": "decision",
|
| 1173 |
+
"type": "choice",
|
| 1174 |
+
"pick": "failed_transfer",
|
| 1175 |
+
"pick_run": "failed_transfer",
|
| 1176 |
+
"same": true,
|
| 1177 |
+
"max_abs_dp": 0.003761896965026823
|
| 1178 |
+
},
|
| 1179 |
+
{
|
| 1180 |
+
"set": "jevals-banking77",
|
| 1181 |
+
"id": "jevals-banking77:banking77-1055",
|
| 1182 |
+
"qid": "decision",
|
| 1183 |
+
"type": "choice",
|
| 1184 |
+
"pick": "balance_not_updated_after_cheque_or_cash_deposit",
|
| 1185 |
+
"pick_run": "balance_not_updated_after_cheque_or_cash_deposit",
|
| 1186 |
+
"same": true,
|
| 1187 |
+
"max_abs_dp": 0.0021776530952453133
|
| 1188 |
+
},
|
| 1189 |
+
{
|
| 1190 |
+
"set": "jevals-banking77",
|
| 1191 |
+
"id": "jevals-banking77:banking77-1227",
|
| 1192 |
+
"qid": "decision",
|
| 1193 |
+
"type": "choice",
|
| 1194 |
+
"pick": "unable_to_verify_identity",
|
| 1195 |
+
"pick_run": "unable_to_verify_identity",
|
| 1196 |
+
"same": true,
|
| 1197 |
+
"max_abs_dp": 0.02141378558158874
|
| 1198 |
+
},
|
| 1199 |
+
{
|
| 1200 |
+
"set": "jevals-banking77",
|
| 1201 |
+
"id": "jevals-banking77:banking77-2237",
|
| 1202 |
+
"qid": "decision",
|
| 1203 |
+
"type": "choice",
|
| 1204 |
+
"pick": "fiat_currency_support",
|
| 1205 |
+
"pick_run": "fiat_currency_support",
|
| 1206 |
+
"same": true,
|
| 1207 |
+
"max_abs_dp": 0.0026266516380309968
|
| 1208 |
+
},
|
| 1209 |
+
{
|
| 1210 |
+
"set": "jevals-banking77",
|
| 1211 |
+
"id": "jevals-banking77:banking77-2633",
|
| 1212 |
+
"qid": "decision",
|
| 1213 |
+
"type": "choice",
|
| 1214 |
+
"pick": "disposable_card_limits",
|
| 1215 |
+
"pick_run": "disposable_card_limits",
|
| 1216 |
+
"same": true,
|
| 1217 |
+
"max_abs_dp": 0.00017110443878176618
|
| 1218 |
+
},
|
| 1219 |
+
{
|
| 1220 |
+
"set": "jevals-banking77",
|
| 1221 |
+
"id": "jevals-banking77:banking77-2362",
|
| 1222 |
+
"qid": "decision",
|
| 1223 |
+
"type": "choice",
|
| 1224 |
+
"pick": "verify_top_up",
|
| 1225 |
+
"pick_run": "verify_top_up",
|
| 1226 |
+
"same": true,
|
| 1227 |
+
"max_abs_dp": 0.0018698823356628758
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"set": "jevals-banking77",
|
| 1231 |
+
"id": "jevals-banking77:banking77-157",
|
| 1232 |
+
"qid": "decision",
|
| 1233 |
+
"type": "choice",
|
| 1234 |
+
"pick": "card_payment_wrong_exchange_rate",
|
| 1235 |
+
"pick_run": "card_payment_wrong_exchange_rate",
|
| 1236 |
+
"same": true,
|
| 1237 |
+
"max_abs_dp": 0.019179564340591426
|
| 1238 |
+
},
|
| 1239 |
+
{
|
| 1240 |
+
"set": "jevals-banking77",
|
| 1241 |
+
"id": "jevals-banking77:banking77-396",
|
| 1242 |
+
"qid": "decision",
|
| 1243 |
+
"type": "choice",
|
| 1244 |
+
"pick": "card_acceptance",
|
| 1245 |
+
"pick_run": "card_acceptance",
|
| 1246 |
+
"same": true,
|
| 1247 |
+
"max_abs_dp": 0.017113198209762592
|
| 1248 |
+
},
|
| 1249 |
+
{
|
| 1250 |
+
"set": "jevals-banking77",
|
| 1251 |
+
"id": "jevals-banking77:banking77-472",
|
| 1252 |
+
"qid": "decision",
|
| 1253 |
+
"type": "choice",
|
| 1254 |
+
"pick": "lost_or_stolen_card",
|
| 1255 |
+
"pick_run": "lost_or_stolen_card",
|
| 1256 |
+
"same": true,
|
| 1257 |
+
"max_abs_dp": 0.005541816398620614
|
| 1258 |
+
},
|
| 1259 |
+
{
|
| 1260 |
+
"set": "jevals-banking77",
|
| 1261 |
+
"id": "jevals-banking77:banking77-1528",
|
| 1262 |
+
"qid": "decision",
|
| 1263 |
+
"type": "choice",
|
| 1264 |
+
"pick": "passcode_forgotten",
|
| 1265 |
+
"pick_run": "passcode_forgotten",
|
| 1266 |
+
"same": true,
|
| 1267 |
+
"max_abs_dp": 8.043476058542721e-05
|
| 1268 |
+
},
|
| 1269 |
+
{
|
| 1270 |
+
"set": "jevals-banking77",
|
| 1271 |
+
"id": "jevals-banking77:banking77-2432",
|
| 1272 |
+
"qid": "decision",
|
| 1273 |
+
"type": "choice",
|
| 1274 |
+
"pick": "getting_spare_card",
|
| 1275 |
+
"pick_run": "getting_spare_card",
|
| 1276 |
+
"same": true,
|
| 1277 |
+
"max_abs_dp": 0.0015329588203430444
|
| 1278 |
+
},
|
| 1279 |
+
{
|
| 1280 |
+
"set": "jevals-banking77",
|
| 1281 |
+
"id": "jevals-banking77:banking77-819",
|
| 1282 |
+
"qid": "decision",
|
| 1283 |
+
"type": "choice",
|
| 1284 |
+
"pick": "extra_charge_on_statement",
|
| 1285 |
+
"pick_run": "extra_charge_on_statement",
|
| 1286 |
+
"same": true,
|
| 1287 |
+
"max_abs_dp": 0.0037726735286712776
|
| 1288 |
+
},
|
| 1289 |
+
{
|
| 1290 |
+
"set": "jevals-banking77",
|
| 1291 |
+
"id": "jevals-banking77:banking77-1504",
|
| 1292 |
+
"qid": "decision",
|
| 1293 |
+
"type": "choice",
|
| 1294 |
+
"pick": "direct_debit_payment_not_recognised",
|
| 1295 |
+
"pick_run": "direct_debit_payment_not_recognised",
|
| 1296 |
+
"same": true,
|
| 1297 |
+
"max_abs_dp": 0.00045620281219482806
|
| 1298 |
+
},
|
| 1299 |
+
{
|
| 1300 |
+
"set": "jevals-banking77",
|
| 1301 |
+
"id": "jevals-banking77:banking77-978",
|
| 1302 |
+
"qid": "decision",
|
| 1303 |
+
"type": "choice",
|
| 1304 |
+
"pick": "card_acceptance",
|
| 1305 |
+
"pick_run": "card_acceptance",
|
| 1306 |
+
"same": true,
|
| 1307 |
+
"max_abs_dp": 0.005841841838598254
|
| 1308 |
+
},
|
| 1309 |
+
{
|
| 1310 |
+
"set": "jevals-banking77",
|
| 1311 |
+
"id": "jevals-banking77:banking77-2236",
|
| 1312 |
+
"qid": "decision",
|
| 1313 |
+
"type": "choice",
|
| 1314 |
+
"pick": "receiving_money",
|
| 1315 |
+
"pick_run": "receiving_money",
|
| 1316 |
+
"same": true,
|
| 1317 |
+
"max_abs_dp": 0.0032306124305725437
|
| 1318 |
+
},
|
| 1319 |
+
{
|
| 1320 |
+
"set": "jevals-banking77",
|
| 1321 |
+
"id": "jevals-banking77:banking77-1925",
|
| 1322 |
+
"qid": "decision",
|
| 1323 |
+
"type": "choice",
|
| 1324 |
+
"pick": "card_swallowed",
|
| 1325 |
+
"pick_run": "card_swallowed",
|
| 1326 |
+
"same": true,
|
| 1327 |
+
"max_abs_dp": 0.00024314561891555786
|
| 1328 |
+
},
|
| 1329 |
+
{
|
| 1330 |
+
"set": "jevals-banking77",
|
| 1331 |
+
"id": "jevals-banking77:banking77-704",
|
| 1332 |
+
"qid": "decision",
|
| 1333 |
+
"type": "choice",
|
| 1334 |
+
"pick": "cancel_transfer",
|
| 1335 |
+
"pick_run": "cancel_transfer",
|
| 1336 |
+
"same": true,
|
| 1337 |
+
"max_abs_dp": 0.01998433731365204
|
| 1338 |
+
},
|
| 1339 |
+
{
|
| 1340 |
+
"set": "jevals-banking77",
|
| 1341 |
+
"id": "jevals-banking77:banking77-2288",
|
| 1342 |
+
"qid": "decision",
|
| 1343 |
+
"type": "choice",
|
| 1344 |
+
"pick": "failed_transfer",
|
| 1345 |
+
"pick_run": "failed_transfer",
|
| 1346 |
+
"same": true,
|
| 1347 |
+
"max_abs_dp": 0.0016355493559837356
|
| 1348 |
+
},
|
| 1349 |
+
{
|
| 1350 |
+
"set": "jevals-banking77",
|
| 1351 |
+
"id": "jevals-banking77:banking77-2119",
|
| 1352 |
+
"qid": "decision",
|
| 1353 |
+
"type": "choice",
|
| 1354 |
+
"pick": "change_pin",
|
| 1355 |
+
"pick_run": "change_pin",
|
| 1356 |
+
"same": true,
|
| 1357 |
+
"max_abs_dp": 0.0011953313136100763
|
| 1358 |
+
},
|
| 1359 |
+
{
|
| 1360 |
+
"set": "jevals-banking77",
|
| 1361 |
+
"id": "jevals-banking77:banking77-2886",
|
| 1362 |
+
"qid": "decision",
|
| 1363 |
+
"type": "choice",
|
| 1364 |
+
"pick": "extra_charge_on_statement",
|
| 1365 |
+
"pick_run": "extra_charge_on_statement",
|
| 1366 |
+
"same": true,
|
| 1367 |
+
"max_abs_dp": 0.010501723793029827
|
| 1368 |
+
},
|
| 1369 |
+
{
|
| 1370 |
+
"set": "jevals-banking77",
|
| 1371 |
+
"id": "jevals-banking77:banking77-1931",
|
| 1372 |
+
"qid": "decision",
|
| 1373 |
+
"type": "choice",
|
| 1374 |
+
"pick": "card_swallowed",
|
| 1375 |
+
"pick_run": "card_swallowed",
|
| 1376 |
+
"same": true,
|
| 1377 |
+
"max_abs_dp": 0.0027530267868042246
|
| 1378 |
+
},
|
| 1379 |
+
{
|
| 1380 |
+
"set": "jevals-banking77",
|
| 1381 |
+
"id": "jevals-banking77:banking77-2558",
|
| 1382 |
+
"qid": "decision",
|
| 1383 |
+
"type": "choice",
|
| 1384 |
+
"pick": "cash_withdrawal_charge",
|
| 1385 |
+
"pick_run": "cash_withdrawal_charge",
|
| 1386 |
+
"same": true,
|
| 1387 |
+
"max_abs_dp": 0.005529832390785215
|
| 1388 |
+
},
|
| 1389 |
+
{
|
| 1390 |
+
"set": "jevals-banking77",
|
| 1391 |
+
"id": "jevals-banking77:banking77-2801",
|
| 1392 |
+
"qid": "decision",
|
| 1393 |
+
"type": "choice",
|
| 1394 |
+
"pick": "top_up_by_card_charge",
|
| 1395 |
+
"pick_run": "top_up_by_card_charge",
|
| 1396 |
+
"same": true,
|
| 1397 |
+
"max_abs_dp": 0.006675042510032647
|
| 1398 |
+
},
|
| 1399 |
+
{
|
| 1400 |
+
"set": "jevals-banking77",
|
| 1401 |
+
"id": "jevals-banking77:banking77-2842",
|
| 1402 |
+
"qid": "decision",
|
| 1403 |
+
"type": "choice",
|
| 1404 |
+
"pick": "activate_my_card",
|
| 1405 |
+
"pick_run": "activate_my_card",
|
| 1406 |
+
"same": true,
|
| 1407 |
+
"max_abs_dp": 0.00021903362655639746
|
| 1408 |
+
},
|
| 1409 |
+
{
|
| 1410 |
+
"set": "jevals-banking77",
|
| 1411 |
+
"id": "jevals-banking77:banking77-258",
|
| 1412 |
+
"qid": "decision",
|
| 1413 |
+
"type": "choice",
|
| 1414 |
+
"pick": "fiat_currency_support",
|
| 1415 |
+
"pick_run": "fiat_currency_support",
|
| 1416 |
+
"same": true,
|
| 1417 |
+
"max_abs_dp": 0.005284831920623789
|
| 1418 |
+
},
|
| 1419 |
+
{
|
| 1420 |
+
"set": "jevals-banking77",
|
| 1421 |
+
"id": "jevals-banking77:banking77-180",
|
| 1422 |
+
"qid": "decision",
|
| 1423 |
+
"type": "choice",
|
| 1424 |
+
"pick": "extra_charge_on_statement",
|
| 1425 |
+
"pick_run": "extra_charge_on_statement",
|
| 1426 |
+
"same": true,
|
| 1427 |
+
"max_abs_dp": 0.005434870723724394
|
| 1428 |
+
},
|
| 1429 |
+
{
|
| 1430 |
+
"set": "jevals-banking77",
|
| 1431 |
+
"id": "jevals-banking77:banking77-2691",
|
| 1432 |
+
"qid": "decision",
|
| 1433 |
+
"type": "choice",
|
| 1434 |
+
"pick": "balance_not_updated_after_bank_transfer",
|
| 1435 |
+
"pick_run": "balance_not_updated_after_bank_transfer",
|
| 1436 |
+
"same": true,
|
| 1437 |
+
"max_abs_dp": 0.0010594485855102143
|
| 1438 |
+
},
|
| 1439 |
+
{
|
| 1440 |
+
"set": "jevals-banking77",
|
| 1441 |
+
"id": "jevals-banking77:banking77-634",
|
| 1442 |
+
"qid": "decision",
|
| 1443 |
+
"type": "choice",
|
| 1444 |
+
"pick": "top_up_by_bank_transfer_charge",
|
| 1445 |
+
"pick_run": "top_up_by_bank_transfer_charge",
|
| 1446 |
+
"same": true,
|
| 1447 |
+
"max_abs_dp": 0.0021456788978576613
|
| 1448 |
+
},
|
| 1449 |
+
{
|
| 1450 |
+
"set": "jevals-banking77",
|
| 1451 |
+
"id": "jevals-banking77:banking77-2958",
|
| 1452 |
+
"qid": "decision",
|
| 1453 |
+
"type": "choice",
|
| 1454 |
+
"pick": "top_up_failed",
|
| 1455 |
+
"pick_run": "top_up_failed",
|
| 1456 |
+
"same": true,
|
| 1457 |
+
"max_abs_dp": 0.0007090187530517378
|
| 1458 |
+
},
|
| 1459 |
+
{
|
| 1460 |
+
"set": "jevals-banking77",
|
| 1461 |
+
"id": "jevals-banking77:banking77-882",
|
| 1462 |
+
"qid": "decision",
|
| 1463 |
+
"type": "choice",
|
| 1464 |
+
"pick": "supported_cards_and_currencies",
|
| 1465 |
+
"pick_run": "fiat_currency_support",
|
| 1466 |
+
"same": false,
|
| 1467 |
+
"max_abs_dp": 0.01041107138633729
|
| 1468 |
+
},
|
| 1469 |
+
{
|
| 1470 |
+
"set": "jevals-banking77",
|
| 1471 |
+
"id": "jevals-banking77:banking77-218",
|
| 1472 |
+
"qid": "decision",
|
| 1473 |
+
"type": "choice",
|
| 1474 |
+
"pick": "pending_cash_withdrawal",
|
| 1475 |
+
"pick_run": "pending_cash_withdrawal",
|
| 1476 |
+
"same": true,
|
| 1477 |
+
"max_abs_dp": 0.004912816371917761
|
| 1478 |
+
},
|
| 1479 |
+
{
|
| 1480 |
+
"set": "jevals-banking77",
|
| 1481 |
+
"id": "jevals-banking77:banking77-1871",
|
| 1482 |
+
"qid": "decision",
|
| 1483 |
+
"type": "choice",
|
| 1484 |
+
"pick": "transfer_fee_charged",
|
| 1485 |
+
"pick_run": "transfer_fee_charged",
|
| 1486 |
+
"same": true,
|
| 1487 |
+
"max_abs_dp": 0.020876409551620456
|
| 1488 |
+
},
|
| 1489 |
+
{
|
| 1490 |
+
"set": "jevals-banking77",
|
| 1491 |
+
"id": "jevals-banking77:banking77-1660",
|
| 1492 |
+
"qid": "decision",
|
| 1493 |
+
"type": "choice",
|
| 1494 |
+
"pick": "lost_or_stolen_card",
|
| 1495 |
+
"pick_run": "lost_or_stolen_card",
|
| 1496 |
+
"same": true,
|
| 1497 |
+
"max_abs_dp": 0.00979334916305541
|
| 1498 |
+
},
|
| 1499 |
+
{
|
| 1500 |
+
"set": "jevals-banking77",
|
| 1501 |
+
"id": "jevals-banking77:banking77-1126",
|
| 1502 |
+
"qid": "decision",
|
| 1503 |
+
"type": "choice",
|
| 1504 |
+
"pick": "edit_personal_details",
|
| 1505 |
+
"pick_run": "edit_personal_details",
|
| 1506 |
+
"same": true,
|
| 1507 |
+
"max_abs_dp": 0.00035575588607783537
|
| 1508 |
+
},
|
| 1509 |
+
{
|
| 1510 |
+
"set": "jevals-banking77",
|
| 1511 |
+
"id": "jevals-banking77:banking77-2321",
|
| 1512 |
+
"qid": "decision",
|
| 1513 |
+
"type": "choice",
|
| 1514 |
+
"pick": "transfer_into_account",
|
| 1515 |
+
"pick_run": "transfer_into_account",
|
| 1516 |
+
"same": true,
|
| 1517 |
+
"max_abs_dp": 0.00010597671341151002
|
| 1518 |
+
},
|
| 1519 |
+
{
|
| 1520 |
+
"set": "jevals-banking77",
|
| 1521 |
+
"id": "jevals-banking77:banking77-2667",
|
| 1522 |
+
"qid": "decision",
|
| 1523 |
+
"type": "choice",
|
| 1524 |
+
"pick": "top_up_failed",
|
| 1525 |
+
"pick_run": "top_up_failed",
|
| 1526 |
+
"same": true,
|
| 1527 |
+
"max_abs_dp": 0.0010350030363202103
|
| 1528 |
+
},
|
| 1529 |
+
{
|
| 1530 |
+
"set": "jevals-banking77",
|
| 1531 |
+
"id": "jevals-banking77:banking77-1537",
|
| 1532 |
+
"qid": "decision",
|
| 1533 |
+
"type": "choice",
|
| 1534 |
+
"pick": "passcode_forgotten",
|
| 1535 |
+
"pick_run": "passcode_forgotten",
|
| 1536 |
+
"same": true,
|
| 1537 |
+
"max_abs_dp": 0.012686089736938522
|
| 1538 |
+
},
|
| 1539 |
+
{
|
| 1540 |
+
"set": "jevals-banking77",
|
| 1541 |
+
"id": "jevals-banking77:banking77-1232",
|
| 1542 |
+
"qid": "decision",
|
| 1543 |
+
"type": "choice",
|
| 1544 |
+
"pick": "why_verify_identity",
|
| 1545 |
+
"pick_run": "why_verify_identity",
|
| 1546 |
+
"same": true,
|
| 1547 |
+
"max_abs_dp": 0.01133909769821162
|
| 1548 |
+
},
|
| 1549 |
+
{
|
| 1550 |
+
"set": "jevals-banking77",
|
| 1551 |
+
"id": "jevals-banking77:banking77-2490",
|
| 1552 |
+
"qid": "decision",
|
| 1553 |
+
"type": "choice",
|
| 1554 |
+
"pick": "get_physical_card",
|
| 1555 |
+
"pick_run": "get_physical_card",
|
| 1556 |
+
"same": true,
|
| 1557 |
+
"max_abs_dp": 0.0013011589164734216
|
| 1558 |
+
},
|
| 1559 |
+
{
|
| 1560 |
+
"set": "jevals-banking77",
|
| 1561 |
+
"id": "jevals-banking77:banking77-1156",
|
| 1562 |
+
"qid": "decision",
|
| 1563 |
+
"type": "choice",
|
| 1564 |
+
"pick": "edit_personal_details",
|
| 1565 |
+
"pick_run": "edit_personal_details",
|
| 1566 |
+
"same": true,
|
| 1567 |
+
"max_abs_dp": 0.0003208802604675043
|
| 1568 |
+
},
|
| 1569 |
+
{
|
| 1570 |
+
"set": "jevals-banking77",
|
| 1571 |
+
"id": "jevals-banking77:banking77-2020",
|
| 1572 |
+
"qid": "decision",
|
| 1573 |
+
"type": "choice",
|
| 1574 |
+
"pick": "verify_source_of_funds",
|
| 1575 |
+
"pick_run": "verify_source_of_funds",
|
| 1576 |
+
"same": true,
|
| 1577 |
+
"max_abs_dp": 0.009149723342895544
|
| 1578 |
+
},
|
| 1579 |
+
{
|
| 1580 |
+
"set": "jevals-banking77",
|
| 1581 |
+
"id": "jevals-banking77:banking77-1671",
|
| 1582 |
+
"qid": "decision",
|
| 1583 |
+
"type": "choice",
|
| 1584 |
+
"pick": "lost_or_stolen_phone",
|
| 1585 |
+
"pick_run": "lost_or_stolen_phone",
|
| 1586 |
+
"same": true,
|
| 1587 |
+
"max_abs_dp": 0.009801792629241968
|
| 1588 |
+
},
|
| 1589 |
+
{
|
| 1590 |
+
"set": "jevals-banking77",
|
| 1591 |
+
"id": "jevals-banking77:banking77-2933",
|
| 1592 |
+
"qid": "decision",
|
| 1593 |
+
"type": "choice",
|
| 1594 |
+
"pick": "order_physical_card",
|
| 1595 |
+
"pick_run": "order_physical_card",
|
| 1596 |
+
"same": true,
|
| 1597 |
+
"max_abs_dp": 0.0070786974754333265
|
| 1598 |
+
},
|
| 1599 |
+
{
|
| 1600 |
+
"set": "jevals-helpsteer2",
|
| 1601 |
+
"id": "jevals-helpsteer2:helpsteer2-599",
|
| 1602 |
+
"qid": "decision",
|
| 1603 |
+
"type": "score",
|
| 1604 |
+
"pick": "3",
|
| 1605 |
+
"pick_run": "3",
|
| 1606 |
+
"same": true,
|
| 1607 |
+
"max_abs_dp": 0.0017070311212539768
|
| 1608 |
+
},
|
| 1609 |
+
{
|
| 1610 |
+
"set": "jevals-helpsteer2",
|
| 1611 |
+
"id": "jevals-helpsteer2:helpsteer2-269",
|
| 1612 |
+
"qid": "decision",
|
| 1613 |
+
"type": "score",
|
| 1614 |
+
"pick": "4",
|
| 1615 |
+
"pick_run": "4",
|
| 1616 |
+
"same": true,
|
| 1617 |
+
"max_abs_dp": 0.004396505262374895
|
| 1618 |
+
},
|
| 1619 |
+
{
|
| 1620 |
+
"set": "jevals-helpsteer2",
|
| 1621 |
+
"id": "jevals-helpsteer2:helpsteer2-910",
|
| 1622 |
+
"qid": "decision",
|
| 1623 |
+
"type": "score",
|
| 1624 |
+
"pick": "2",
|
| 1625 |
+
"pick_run": "2",
|
| 1626 |
+
"same": true,
|
| 1627 |
+
"max_abs_dp": 0.001993250860214235
|
| 1628 |
+
},
|
| 1629 |
+
{
|
| 1630 |
+
"set": "jevals-helpsteer2",
|
| 1631 |
+
"id": "jevals-helpsteer2:helpsteer2-243",
|
| 1632 |
+
"qid": "decision",
|
| 1633 |
+
"type": "score",
|
| 1634 |
+
"pick": "4",
|
| 1635 |
+
"pick_run": "4",
|
| 1636 |
+
"same": true,
|
| 1637 |
+
"max_abs_dp": 0.0022569894447326266
|
| 1638 |
+
},
|
| 1639 |
+
{
|
| 1640 |
+
"set": "jevals-helpsteer2",
|
| 1641 |
+
"id": "jevals-helpsteer2:helpsteer2-789",
|
| 1642 |
+
"qid": "decision",
|
| 1643 |
+
"type": "score",
|
| 1644 |
+
"pick": "1",
|
| 1645 |
+
"pick_run": "1",
|
| 1646 |
+
"same": true,
|
| 1647 |
+
"max_abs_dp": 0.0014194283542633135
|
| 1648 |
+
},
|
| 1649 |
+
{
|
| 1650 |
+
"set": "jevals-helpsteer2",
|
| 1651 |
+
"id": "jevals-helpsteer2:helpsteer2-1001",
|
| 1652 |
+
"qid": "decision",
|
| 1653 |
+
"type": "score",
|
| 1654 |
+
"pick": "0",
|
| 1655 |
+
"pick_run": "0",
|
| 1656 |
+
"same": true,
|
| 1657 |
+
"max_abs_dp": 0.0025998412761688017
|
| 1658 |
+
},
|
| 1659 |
+
{
|
| 1660 |
+
"set": "jevals-helpsteer2",
|
| 1661 |
+
"id": "jevals-helpsteer2:helpsteer2-448",
|
| 1662 |
+
"qid": "decision",
|
| 1663 |
+
"type": "score",
|
| 1664 |
+
"pick": "3",
|
| 1665 |
+
"pick_run": "3",
|
| 1666 |
+
"same": true,
|
| 1667 |
+
"max_abs_dp": 0.0021863538465499988
|
| 1668 |
+
},
|
| 1669 |
+
{
|
| 1670 |
+
"set": "jevals-helpsteer2",
|
| 1671 |
+
"id": "jevals-helpsteer2:helpsteer2-1008",
|
| 1672 |
+
"qid": "decision",
|
| 1673 |
+
"type": "score",
|
| 1674 |
+
"pick": "4",
|
| 1675 |
+
"pick_run": "4",
|
| 1676 |
+
"same": true,
|
| 1677 |
+
"max_abs_dp": 0.0024829814634323155
|
| 1678 |
+
},
|
| 1679 |
+
{
|
| 1680 |
+
"set": "jevals-helpsteer2",
|
| 1681 |
+
"id": "jevals-helpsteer2:helpsteer2-873",
|
| 1682 |
+
"qid": "decision",
|
| 1683 |
+
"type": "score",
|
| 1684 |
+
"pick": "4",
|
| 1685 |
+
"pick_run": "4",
|
| 1686 |
+
"same": true,
|
| 1687 |
+
"max_abs_dp": 0.001164060227394098
|
| 1688 |
+
},
|
| 1689 |
+
{
|
| 1690 |
+
"set": "jevals-helpsteer2",
|
| 1691 |
+
"id": "jevals-helpsteer2:helpsteer2-705",
|
| 1692 |
+
"qid": "decision",
|
| 1693 |
+
"type": "score",
|
| 1694 |
+
"pick": "1",
|
| 1695 |
+
"pick_run": "1",
|
| 1696 |
+
"same": true,
|
| 1697 |
+
"max_abs_dp": 0.002851796442985527
|
| 1698 |
+
},
|
| 1699 |
+
{
|
| 1700 |
+
"set": "jevals-helpsteer2",
|
| 1701 |
+
"id": "jevals-helpsteer2:helpsteer2-151",
|
| 1702 |
+
"qid": "decision",
|
| 1703 |
+
"type": "score",
|
| 1704 |
+
"pick": "2",
|
| 1705 |
+
"pick_run": "2",
|
| 1706 |
+
"same": true,
|
| 1707 |
+
"max_abs_dp": 0.0020692041740417444
|
| 1708 |
+
},
|
| 1709 |
+
{
|
| 1710 |
+
"set": "jevals-helpsteer2",
|
| 1711 |
+
"id": "jevals-helpsteer2:helpsteer2-920",
|
| 1712 |
+
"qid": "decision",
|
| 1713 |
+
"type": "score",
|
| 1714 |
+
"pick": "0",
|
| 1715 |
+
"pick_run": "0",
|
| 1716 |
+
"same": true,
|
| 1717 |
+
"max_abs_dp": 0.004536616085052514
|
| 1718 |
+
},
|
| 1719 |
+
{
|
| 1720 |
+
"set": "jevals-helpsteer2",
|
| 1721 |
+
"id": "jevals-helpsteer2:helpsteer2-496",
|
| 1722 |
+
"qid": "decision",
|
| 1723 |
+
"type": "score",
|
| 1724 |
+
"pick": "4",
|
| 1725 |
+
"pick_run": "4",
|
| 1726 |
+
"same": true,
|
| 1727 |
+
"max_abs_dp": 0.0011169719676971668
|
| 1728 |
+
},
|
| 1729 |
+
{
|
| 1730 |
+
"set": "jevals-helpsteer2",
|
| 1731 |
+
"id": "jevals-helpsteer2:helpsteer2-21",
|
| 1732 |
+
"qid": "decision",
|
| 1733 |
+
"type": "score",
|
| 1734 |
+
"pick": "0",
|
| 1735 |
+
"pick_run": "0",
|
| 1736 |
+
"same": true,
|
| 1737 |
+
"max_abs_dp": 0.0010305535545349631
|
| 1738 |
+
},
|
| 1739 |
+
{
|
| 1740 |
+
"set": "jevals-helpsteer2",
|
| 1741 |
+
"id": "jevals-helpsteer2:helpsteer2-553",
|
| 1742 |
+
"qid": "decision",
|
| 1743 |
+
"type": "score",
|
| 1744 |
+
"pick": "4",
|
| 1745 |
+
"pick_run": "4",
|
| 1746 |
+
"same": true,
|
| 1747 |
+
"max_abs_dp": 0.0019170843048095798
|
| 1748 |
+
},
|
| 1749 |
+
{
|
| 1750 |
+
"set": "jevals-helpsteer2",
|
| 1751 |
+
"id": "jevals-helpsteer2:helpsteer2-618",
|
| 1752 |
+
"qid": "decision",
|
| 1753 |
+
"type": "score",
|
| 1754 |
+
"pick": "4",
|
| 1755 |
+
"pick_run": "4",
|
| 1756 |
+
"same": true,
|
| 1757 |
+
"max_abs_dp": 0.0021986666736602545
|
| 1758 |
+
},
|
| 1759 |
+
{
|
| 1760 |
+
"set": "jevals-helpsteer2",
|
| 1761 |
+
"id": "jevals-helpsteer2:helpsteer2-452",
|
| 1762 |
+
"qid": "decision",
|
| 1763 |
+
"type": "score",
|
| 1764 |
+
"pick": "0",
|
| 1765 |
+
"pick_run": "0",
|
| 1766 |
+
"same": true,
|
| 1767 |
+
"max_abs_dp": 0.0028643911151886026
|
| 1768 |
+
},
|
| 1769 |
+
{
|
| 1770 |
+
"set": "jevals-helpsteer2",
|
| 1771 |
+
"id": "jevals-helpsteer2:helpsteer2-422",
|
| 1772 |
+
"qid": "decision",
|
| 1773 |
+
"type": "score",
|
| 1774 |
+
"pick": "3",
|
| 1775 |
+
"pick_run": "3",
|
| 1776 |
+
"same": true,
|
| 1777 |
+
"max_abs_dp": 0.0068150336418151825
|
| 1778 |
+
},
|
| 1779 |
+
{
|
| 1780 |
+
"set": "jevals-helpsteer2",
|
| 1781 |
+
"id": "jevals-helpsteer2:helpsteer2-820",
|
| 1782 |
+
"qid": "decision",
|
| 1783 |
+
"type": "score",
|
| 1784 |
+
"pick": "4",
|
| 1785 |
+
"pick_run": "4",
|
| 1786 |
+
"same": true,
|
| 1787 |
+
"max_abs_dp": 0.003520079170227064
|
| 1788 |
+
},
|
| 1789 |
+
{
|
| 1790 |
+
"set": "jevals-helpsteer2",
|
| 1791 |
+
"id": "jevals-helpsteer2:helpsteer2-1034",
|
| 1792 |
+
"qid": "decision",
|
| 1793 |
+
"type": "score",
|
| 1794 |
+
"pick": "3",
|
| 1795 |
+
"pick_run": "3",
|
| 1796 |
+
"same": true,
|
| 1797 |
+
"max_abs_dp": 0.003428257497787468
|
| 1798 |
+
},
|
| 1799 |
+
{
|
| 1800 |
+
"set": "jevals-helpsteer2",
|
| 1801 |
+
"id": "jevals-helpsteer2:helpsteer2-14",
|
| 1802 |
+
"qid": "decision",
|
| 1803 |
+
"type": "score",
|
| 1804 |
+
"pick": "4",
|
| 1805 |
+
"pick_run": "4",
|
| 1806 |
+
"same": true,
|
| 1807 |
+
"max_abs_dp": 0.0030382548294067835
|
| 1808 |
+
},
|
| 1809 |
+
{
|
| 1810 |
+
"set": "jevals-helpsteer2",
|
| 1811 |
+
"id": "jevals-helpsteer2:helpsteer2-221",
|
| 1812 |
+
"qid": "decision",
|
| 1813 |
+
"type": "score",
|
| 1814 |
+
"pick": "3",
|
| 1815 |
+
"pick_run": "3",
|
| 1816 |
+
"same": true,
|
| 1817 |
+
"max_abs_dp": 0.002335368223190304
|
| 1818 |
+
},
|
| 1819 |
+
{
|
| 1820 |
+
"set": "jevals-helpsteer2",
|
| 1821 |
+
"id": "jevals-helpsteer2:helpsteer2-367",
|
| 1822 |
+
"qid": "decision",
|
| 1823 |
+
"type": "score",
|
| 1824 |
+
"pick": "3",
|
| 1825 |
+
"pick_run": "3",
|
| 1826 |
+
"same": true,
|
| 1827 |
+
"max_abs_dp": 0.005168520936965931
|
| 1828 |
+
},
|
| 1829 |
+
{
|
| 1830 |
+
"set": "jevals-helpsteer2",
|
| 1831 |
+
"id": "jevals-helpsteer2:helpsteer2-898",
|
| 1832 |
+
"qid": "decision",
|
| 1833 |
+
"type": "score",
|
| 1834 |
+
"pick": "1",
|
| 1835 |
+
"pick_run": "1",
|
| 1836 |
+
"same": true,
|
| 1837 |
+
"max_abs_dp": 0.003371750524520878
|
| 1838 |
+
},
|
| 1839 |
+
{
|
| 1840 |
+
"set": "jevals-helpsteer2",
|
| 1841 |
+
"id": "jevals-helpsteer2:helpsteer2-475",
|
| 1842 |
+
"qid": "decision",
|
| 1843 |
+
"type": "score",
|
| 1844 |
+
"pick": "3",
|
| 1845 |
+
"pick_run": "3",
|
| 1846 |
+
"same": true,
|
| 1847 |
+
"max_abs_dp": 0.0018364327745437614
|
| 1848 |
+
},
|
| 1849 |
+
{
|
| 1850 |
+
"set": "jevals-helpsteer2",
|
| 1851 |
+
"id": "jevals-helpsteer2:helpsteer2-1022",
|
| 1852 |
+
"qid": "decision",
|
| 1853 |
+
"type": "score",
|
| 1854 |
+
"pick": "3",
|
| 1855 |
+
"pick_run": "3",
|
| 1856 |
+
"same": true,
|
| 1857 |
+
"max_abs_dp": 0.003297201475143452
|
| 1858 |
+
},
|
| 1859 |
+
{
|
| 1860 |
+
"set": "jevals-helpsteer2",
|
| 1861 |
+
"id": "jevals-helpsteer2:helpsteer2-767",
|
| 1862 |
+
"qid": "decision",
|
| 1863 |
+
"type": "score",
|
| 1864 |
+
"pick": "2",
|
| 1865 |
+
"pick_run": "2",
|
| 1866 |
+
"same": true,
|
| 1867 |
+
"max_abs_dp": 0.00298056771183014
|
| 1868 |
+
},
|
| 1869 |
+
{
|
| 1870 |
+
"set": "jevals-helpsteer2",
|
| 1871 |
+
"id": "jevals-helpsteer2:helpsteer2-588",
|
| 1872 |
+
"qid": "decision",
|
| 1873 |
+
"type": "score",
|
| 1874 |
+
"pick": "4",
|
| 1875 |
+
"pick_run": "4",
|
| 1876 |
+
"same": true,
|
| 1877 |
+
"max_abs_dp": 0.0030666067352295157
|
| 1878 |
+
},
|
| 1879 |
+
{
|
| 1880 |
+
"set": "jevals-helpsteer2",
|
| 1881 |
+
"id": "jevals-helpsteer2:helpsteer2-380",
|
| 1882 |
+
"qid": "decision",
|
| 1883 |
+
"type": "score",
|
| 1884 |
+
"pick": "4",
|
| 1885 |
+
"pick_run": "4",
|
| 1886 |
+
"same": true,
|
| 1887 |
+
"max_abs_dp": 0.0031253642139434934
|
| 1888 |
+
},
|
| 1889 |
+
{
|
| 1890 |
+
"set": "jevals-helpsteer2",
|
| 1891 |
+
"id": "jevals-helpsteer2:helpsteer2-444",
|
| 1892 |
+
"qid": "decision",
|
| 1893 |
+
"type": "score",
|
| 1894 |
+
"pick": "2",
|
| 1895 |
+
"pick_run": "2",
|
| 1896 |
+
"same": true,
|
| 1897 |
+
"max_abs_dp": 0.0029246465024947987
|
| 1898 |
+
},
|
| 1899 |
+
{
|
| 1900 |
+
"set": "jevals-helpsteer2",
|
| 1901 |
+
"id": "jevals-helpsteer2:helpsteer2-786",
|
| 1902 |
+
"qid": "decision",
|
| 1903 |
+
"type": "score",
|
| 1904 |
+
"pick": "1",
|
| 1905 |
+
"pick_run": "1",
|
| 1906 |
+
"same": true,
|
| 1907 |
+
"max_abs_dp": 0.002068277301788324
|
| 1908 |
+
},
|
| 1909 |
+
{
|
| 1910 |
+
"set": "jevals-helpsteer2",
|
| 1911 |
+
"id": "jevals-helpsteer2:helpsteer2-891",
|
| 1912 |
+
"qid": "decision",
|
| 1913 |
+
"type": "score",
|
| 1914 |
+
"pick": "0",
|
| 1915 |
+
"pick_run": "0",
|
| 1916 |
+
"same": true,
|
| 1917 |
+
"max_abs_dp": 0.0015666385917663117
|
| 1918 |
+
},
|
| 1919 |
+
{
|
| 1920 |
+
"set": "jevals-helpsteer2",
|
| 1921 |
+
"id": "jevals-helpsteer2:helpsteer2-805",
|
| 1922 |
+
"qid": "decision",
|
| 1923 |
+
"type": "score",
|
| 1924 |
+
"pick": "1",
|
| 1925 |
+
"pick_run": "1",
|
| 1926 |
+
"same": true,
|
| 1927 |
+
"max_abs_dp": 0.0020603434896468986
|
| 1928 |
+
},
|
| 1929 |
+
{
|
| 1930 |
+
"set": "jevals-helpsteer2",
|
| 1931 |
+
"id": "jevals-helpsteer2:helpsteer2-54",
|
| 1932 |
+
"qid": "decision",
|
| 1933 |
+
"type": "score",
|
| 1934 |
+
"pick": "4",
|
| 1935 |
+
"pick_run": "4",
|
| 1936 |
+
"same": true,
|
| 1937 |
+
"max_abs_dp": 0.005909478620529179
|
| 1938 |
+
},
|
| 1939 |
+
{
|
| 1940 |
+
"set": "jevals-helpsteer2",
|
| 1941 |
+
"id": "jevals-helpsteer2:helpsteer2-142",
|
| 1942 |
+
"qid": "decision",
|
| 1943 |
+
"type": "score",
|
| 1944 |
+
"pick": "3",
|
| 1945 |
+
"pick_run": "3",
|
| 1946 |
+
"same": true,
|
| 1947 |
+
"max_abs_dp": 0.004755552507400507
|
| 1948 |
+
},
|
| 1949 |
+
{
|
| 1950 |
+
"set": "jevals-helpsteer2",
|
| 1951 |
+
"id": "jevals-helpsteer2:helpsteer2-178",
|
| 1952 |
+
"qid": "decision",
|
| 1953 |
+
"type": "score",
|
| 1954 |
+
"pick": "1",
|
| 1955 |
+
"pick_run": "1",
|
| 1956 |
+
"same": true,
|
| 1957 |
+
"max_abs_dp": 0.0042735126762389886
|
| 1958 |
+
},
|
| 1959 |
+
{
|
| 1960 |
+
"set": "jevals-helpsteer2",
|
| 1961 |
+
"id": "jevals-helpsteer2:helpsteer2-531",
|
| 1962 |
+
"qid": "decision",
|
| 1963 |
+
"type": "score",
|
| 1964 |
+
"pick": "4",
|
| 1965 |
+
"pick_run": "4",
|
| 1966 |
+
"same": true,
|
| 1967 |
+
"max_abs_dp": 0.001332084762573238
|
| 1968 |
+
},
|
| 1969 |
+
{
|
| 1970 |
+
"set": "jevals-helpsteer2",
|
| 1971 |
+
"id": "jevals-helpsteer2:helpsteer2-823",
|
| 1972 |
+
"qid": "decision",
|
| 1973 |
+
"type": "score",
|
| 1974 |
+
"pick": "4",
|
| 1975 |
+
"pick_run": "4",
|
| 1976 |
+
"same": true,
|
| 1977 |
+
"max_abs_dp": 0.004615197769165025
|
| 1978 |
+
},
|
| 1979 |
+
{
|
| 1980 |
+
"set": "jevals-helpsteer2",
|
| 1981 |
+
"id": "jevals-helpsteer2:helpsteer2-305",
|
| 1982 |
+
"qid": "decision",
|
| 1983 |
+
"type": "score",
|
| 1984 |
+
"pick": "4",
|
| 1985 |
+
"pick_run": "4",
|
| 1986 |
+
"same": true,
|
| 1987 |
+
"max_abs_dp": 0.0016935211257934624
|
| 1988 |
+
},
|
| 1989 |
+
{
|
| 1990 |
+
"set": "jevals-helpsteer2",
|
| 1991 |
+
"id": "jevals-helpsteer2:helpsteer2-526",
|
| 1992 |
+
"qid": "decision",
|
| 1993 |
+
"type": "score",
|
| 1994 |
+
"pick": "4",
|
| 1995 |
+
"pick_run": "4",
|
| 1996 |
+
"same": true,
|
| 1997 |
+
"max_abs_dp": 0.002539206399917615
|
| 1998 |
+
},
|
| 1999 |
+
{
|
| 2000 |
+
"set": "jevals-helpsteer2",
|
| 2001 |
+
"id": "jevals-helpsteer2:helpsteer2-359",
|
| 2002 |
+
"qid": "decision",
|
| 2003 |
+
"type": "score",
|
| 2004 |
+
"pick": "4",
|
| 2005 |
+
"pick_run": "4",
|
| 2006 |
+
"same": true,
|
| 2007 |
+
"max_abs_dp": 0.0010118418216705227
|
| 2008 |
+
},
|
| 2009 |
+
{
|
| 2010 |
+
"set": "jevals-helpsteer2",
|
| 2011 |
+
"id": "jevals-helpsteer2:helpsteer2-783",
|
| 2012 |
+
"qid": "decision",
|
| 2013 |
+
"type": "score",
|
| 2014 |
+
"pick": "1",
|
| 2015 |
+
"pick_run": "1",
|
| 2016 |
+
"same": true,
|
| 2017 |
+
"max_abs_dp": 0.0010982502269744898
|
| 2018 |
+
},
|
| 2019 |
+
{
|
| 2020 |
+
"set": "jevals-helpsteer2",
|
| 2021 |
+
"id": "jevals-helpsteer2:helpsteer2-644",
|
| 2022 |
+
"qid": "decision",
|
| 2023 |
+
"type": "score",
|
| 2024 |
+
"pick": "0",
|
| 2025 |
+
"pick_run": "0",
|
| 2026 |
+
"same": true,
|
| 2027 |
+
"max_abs_dp": 0.0013904987869262642
|
| 2028 |
+
},
|
| 2029 |
+
{
|
| 2030 |
+
"set": "jevals-helpsteer2",
|
| 2031 |
+
"id": "jevals-helpsteer2:helpsteer2-278",
|
| 2032 |
+
"qid": "decision",
|
| 2033 |
+
"type": "score",
|
| 2034 |
+
"pick": "1",
|
| 2035 |
+
"pick_run": "1",
|
| 2036 |
+
"same": true,
|
| 2037 |
+
"max_abs_dp": 0.0006149902935028151
|
| 2038 |
+
},
|
| 2039 |
+
{
|
| 2040 |
+
"set": "jevals-helpsteer2",
|
| 2041 |
+
"id": "jevals-helpsteer2:helpsteer2-790",
|
| 2042 |
+
"qid": "decision",
|
| 2043 |
+
"type": "score",
|
| 2044 |
+
"pick": "4",
|
| 2045 |
+
"pick_run": "4",
|
| 2046 |
+
"same": true,
|
| 2047 |
+
"max_abs_dp": 0.0033819403591155983
|
| 2048 |
+
},
|
| 2049 |
+
{
|
| 2050 |
+
"set": "jevals-helpsteer2",
|
| 2051 |
+
"id": "jevals-helpsteer2:helpsteer2-723",
|
| 2052 |
+
"qid": "decision",
|
| 2053 |
+
"type": "score",
|
| 2054 |
+
"pick": "4",
|
| 2055 |
+
"pick_run": "4",
|
| 2056 |
+
"same": true,
|
| 2057 |
+
"max_abs_dp": 0.000605950281143186
|
| 2058 |
+
},
|
| 2059 |
+
{
|
| 2060 |
+
"set": "jevals-helpsteer2",
|
| 2061 |
+
"id": "jevals-helpsteer2:helpsteer2-972",
|
| 2062 |
+
"qid": "decision",
|
| 2063 |
+
"type": "score",
|
| 2064 |
+
"pick": "3",
|
| 2065 |
+
"pick_run": "2",
|
| 2066 |
+
"same": false,
|
| 2067 |
+
"max_abs_dp": 0.007170431488037116
|
| 2068 |
+
},
|
| 2069 |
+
{
|
| 2070 |
+
"set": "jevals-helpsteer2",
|
| 2071 |
+
"id": "jevals-helpsteer2:helpsteer2-646",
|
| 2072 |
+
"qid": "decision",
|
| 2073 |
+
"type": "score",
|
| 2074 |
+
"pick": "3",
|
| 2075 |
+
"pick_run": "3",
|
| 2076 |
+
"same": true,
|
| 2077 |
+
"max_abs_dp": 0.006493632644653302
|
| 2078 |
+
},
|
| 2079 |
+
{
|
| 2080 |
+
"set": "jevals-helpsteer2",
|
| 2081 |
+
"id": "jevals-helpsteer2:helpsteer2-878",
|
| 2082 |
+
"qid": "decision",
|
| 2083 |
+
"type": "score",
|
| 2084 |
+
"pick": "3",
|
| 2085 |
+
"pick_run": "3",
|
| 2086 |
+
"same": true,
|
| 2087 |
+
"max_abs_dp": 0.0025942115430832025
|
| 2088 |
+
},
|
| 2089 |
+
{
|
| 2090 |
+
"set": "jevals-helpsteer2",
|
| 2091 |
+
"id": "jevals-helpsteer2:helpsteer2-945",
|
| 2092 |
+
"qid": "decision",
|
| 2093 |
+
"type": "score",
|
| 2094 |
+
"pick": "2",
|
| 2095 |
+
"pick_run": "3",
|
| 2096 |
+
"same": false,
|
| 2097 |
+
"max_abs_dp": 0.003689604001998914
|
| 2098 |
+
},
|
| 2099 |
+
{
|
| 2100 |
+
"set": "jevals-helpsteer2",
|
| 2101 |
+
"id": "jevals-helpsteer2:helpsteer2-963",
|
| 2102 |
+
"qid": "decision",
|
| 2103 |
+
"type": "score",
|
| 2104 |
+
"pick": "4",
|
| 2105 |
+
"pick_run": "4",
|
| 2106 |
+
"same": true,
|
| 2107 |
+
"max_abs_dp": 0.0027546647338866914
|
| 2108 |
+
},
|
| 2109 |
+
{
|
| 2110 |
+
"set": "jevals-helpsteer2",
|
| 2111 |
+
"id": "jevals-helpsteer2:helpsteer2-83",
|
| 2112 |
+
"qid": "decision",
|
| 2113 |
+
"type": "score",
|
| 2114 |
+
"pick": "3",
|
| 2115 |
+
"pick_run": "3",
|
| 2116 |
+
"same": true,
|
| 2117 |
+
"max_abs_dp": 0.003922615794181822
|
| 2118 |
+
},
|
| 2119 |
+
{
|
| 2120 |
+
"set": "jevals-helpsteer2",
|
| 2121 |
+
"id": "jevals-helpsteer2:helpsteer2-66",
|
| 2122 |
+
"qid": "decision",
|
| 2123 |
+
"type": "score",
|
| 2124 |
+
"pick": "3",
|
| 2125 |
+
"pick_run": "3",
|
| 2126 |
+
"same": true,
|
| 2127 |
+
"max_abs_dp": 0.0036037472667694304
|
| 2128 |
+
},
|
| 2129 |
+
{
|
| 2130 |
+
"set": "jevals-helpsteer2",
|
| 2131 |
+
"id": "jevals-helpsteer2:helpsteer2-914",
|
| 2132 |
+
"qid": "decision",
|
| 2133 |
+
"type": "score",
|
| 2134 |
+
"pick": "3",
|
| 2135 |
+
"pick_run": "3",
|
| 2136 |
+
"same": true,
|
| 2137 |
+
"max_abs_dp": 0.0057809354381561096
|
| 2138 |
+
},
|
| 2139 |
+
{
|
| 2140 |
+
"set": "jevals-helpsteer2",
|
| 2141 |
+
"id": "jevals-helpsteer2:helpsteer2-258",
|
| 2142 |
+
"qid": "decision",
|
| 2143 |
+
"type": "score",
|
| 2144 |
+
"pick": "4",
|
| 2145 |
+
"pick_run": "4",
|
| 2146 |
+
"same": true,
|
| 2147 |
+
"max_abs_dp": 0.0025659412574768337
|
| 2148 |
+
},
|
| 2149 |
+
{
|
| 2150 |
+
"set": "jevals-helpsteer2",
|
| 2151 |
+
"id": "jevals-helpsteer2:helpsteer2-998",
|
| 2152 |
+
"qid": "decision",
|
| 2153 |
+
"type": "score",
|
| 2154 |
+
"pick": "1",
|
| 2155 |
+
"pick_run": "1",
|
| 2156 |
+
"same": true,
|
| 2157 |
+
"max_abs_dp": 0.0053724314994811895
|
| 2158 |
+
},
|
| 2159 |
+
{
|
| 2160 |
+
"set": "jevals-helpsteer2",
|
| 2161 |
+
"id": "jevals-helpsteer2:helpsteer2-326",
|
| 2162 |
+
"qid": "decision",
|
| 2163 |
+
"type": "score",
|
| 2164 |
+
"pick": "4",
|
| 2165 |
+
"pick_run": "4",
|
| 2166 |
+
"same": true,
|
| 2167 |
+
"max_abs_dp": 0.0018118249874115033
|
| 2168 |
+
},
|
| 2169 |
+
{
|
| 2170 |
+
"set": "jevals-helpsteer2",
|
| 2171 |
+
"id": "jevals-helpsteer2:helpsteer2-69",
|
| 2172 |
+
"qid": "decision",
|
| 2173 |
+
"type": "score",
|
| 2174 |
+
"pick": "3",
|
| 2175 |
+
"pick_run": "3",
|
| 2176 |
+
"same": true,
|
| 2177 |
+
"max_abs_dp": 0.005150423919677738
|
| 2178 |
+
},
|
| 2179 |
+
{
|
| 2180 |
+
"set": "jevals-helpsteer2",
|
| 2181 |
+
"id": "jevals-helpsteer2:helpsteer2-630",
|
| 2182 |
+
"qid": "decision",
|
| 2183 |
+
"type": "score",
|
| 2184 |
+
"pick": "4",
|
| 2185 |
+
"pick_run": "4",
|
| 2186 |
+
"same": true,
|
| 2187 |
+
"max_abs_dp": 0.0009986610774994076
|
| 2188 |
+
},
|
| 2189 |
+
{
|
| 2190 |
+
"set": "jevals-helpsteer2",
|
| 2191 |
+
"id": "jevals-helpsteer2:helpsteer2-566",
|
| 2192 |
+
"qid": "decision",
|
| 2193 |
+
"type": "score",
|
| 2194 |
+
"pick": "3",
|
| 2195 |
+
"pick_run": "3",
|
| 2196 |
+
"same": true,
|
| 2197 |
+
"max_abs_dp": 0.0033846534290313812
|
| 2198 |
+
},
|
| 2199 |
+
{
|
| 2200 |
+
"set": "jevals-helpsteer2",
|
| 2201 |
+
"id": "jevals-helpsteer2:helpsteer2-404",
|
| 2202 |
+
"qid": "decision",
|
| 2203 |
+
"type": "score",
|
| 2204 |
+
"pick": "4",
|
| 2205 |
+
"pick_run": "4",
|
| 2206 |
+
"same": true,
|
| 2207 |
+
"max_abs_dp": 0.0016019409198760881
|
| 2208 |
+
},
|
| 2209 |
+
{
|
| 2210 |
+
"set": "jevals-helpsteer2",
|
| 2211 |
+
"id": "jevals-helpsteer2:helpsteer2-793",
|
| 2212 |
+
"qid": "decision",
|
| 2213 |
+
"type": "score",
|
| 2214 |
+
"pick": "4",
|
| 2215 |
+
"pick_run": "4",
|
| 2216 |
+
"same": true,
|
| 2217 |
+
"max_abs_dp": 0.0037356444110870357
|
| 2218 |
+
},
|
| 2219 |
+
{
|
| 2220 |
+
"set": "jevals-helpsteer2",
|
| 2221 |
+
"id": "jevals-helpsteer2:helpsteer2-901",
|
| 2222 |
+
"qid": "decision",
|
| 2223 |
+
"type": "score",
|
| 2224 |
+
"pick": "0",
|
| 2225 |
+
"pick_run": "0",
|
| 2226 |
+
"same": true,
|
| 2227 |
+
"max_abs_dp": 0.0011874838867187432
|
| 2228 |
+
},
|
| 2229 |
+
{
|
| 2230 |
+
"set": "jevals-helpsteer2",
|
| 2231 |
+
"id": "jevals-helpsteer2:helpsteer2-532",
|
| 2232 |
+
"qid": "decision",
|
| 2233 |
+
"type": "score",
|
| 2234 |
+
"pick": "3",
|
| 2235 |
+
"pick_run": "3",
|
| 2236 |
+
"same": true,
|
| 2237 |
+
"max_abs_dp": 0.001475205549240105
|
| 2238 |
+
},
|
| 2239 |
+
{
|
| 2240 |
+
"set": "jevals-helpsteer2",
|
| 2241 |
+
"id": "jevals-helpsteer2:helpsteer2-446",
|
| 2242 |
+
"qid": "decision",
|
| 2243 |
+
"type": "score",
|
| 2244 |
+
"pick": "3",
|
| 2245 |
+
"pick_run": "3",
|
| 2246 |
+
"same": true,
|
| 2247 |
+
"max_abs_dp": 0.0035881805105209275
|
| 2248 |
+
},
|
| 2249 |
+
{
|
| 2250 |
+
"set": "jevals-helpsteer2",
|
| 2251 |
+
"id": "jevals-helpsteer2:helpsteer2-844",
|
| 2252 |
+
"qid": "decision",
|
| 2253 |
+
"type": "score",
|
| 2254 |
+
"pick": "3",
|
| 2255 |
+
"pick_run": "3",
|
| 2256 |
+
"same": true,
|
| 2257 |
+
"max_abs_dp": 0.0017071657466888324
|
| 2258 |
+
},
|
| 2259 |
+
{
|
| 2260 |
+
"set": "jevals-helpsteer2",
|
| 2261 |
+
"id": "jevals-helpsteer2:helpsteer2-411",
|
| 2262 |
+
"qid": "decision",
|
| 2263 |
+
"type": "score",
|
| 2264 |
+
"pick": "4",
|
| 2265 |
+
"pick_run": "4",
|
| 2266 |
+
"same": true,
|
| 2267 |
+
"max_abs_dp": 0.0013958457717895278
|
| 2268 |
+
},
|
| 2269 |
+
{
|
| 2270 |
+
"set": "jevals-helpsteer2",
|
| 2271 |
+
"id": "jevals-helpsteer2:helpsteer2-685",
|
| 2272 |
+
"qid": "decision",
|
| 2273 |
+
"type": "score",
|
| 2274 |
+
"pick": "4",
|
| 2275 |
+
"pick_run": "4",
|
| 2276 |
+
"same": true,
|
| 2277 |
+
"max_abs_dp": 0.0019896724929809317
|
| 2278 |
+
},
|
| 2279 |
+
{
|
| 2280 |
+
"set": "jevals-helpsteer2",
|
| 2281 |
+
"id": "jevals-helpsteer2:helpsteer2-569",
|
| 2282 |
+
"qid": "decision",
|
| 2283 |
+
"type": "score",
|
| 2284 |
+
"pick": "1",
|
| 2285 |
+
"pick_run": "1",
|
| 2286 |
+
"same": true,
|
| 2287 |
+
"max_abs_dp": 0.003196729080200189
|
| 2288 |
+
},
|
| 2289 |
+
{
|
| 2290 |
+
"set": "jevals-helpsteer2",
|
| 2291 |
+
"id": "jevals-helpsteer2:helpsteer2-980",
|
| 2292 |
+
"qid": "decision",
|
| 2293 |
+
"type": "score",
|
| 2294 |
+
"pick": "2",
|
| 2295 |
+
"pick_run": "2",
|
| 2296 |
+
"same": true,
|
| 2297 |
+
"max_abs_dp": 0.003319434814453126
|
| 2298 |
+
},
|
| 2299 |
+
{
|
| 2300 |
+
"set": "jevals-helpsteer2",
|
| 2301 |
+
"id": "jevals-helpsteer2:helpsteer2-338",
|
| 2302 |
+
"qid": "decision",
|
| 2303 |
+
"type": "score",
|
| 2304 |
+
"pick": "4",
|
| 2305 |
+
"pick_run": "4",
|
| 2306 |
+
"same": true,
|
| 2307 |
+
"max_abs_dp": 0.0026797406692504744
|
| 2308 |
+
},
|
| 2309 |
+
{
|
| 2310 |
+
"set": "jevals-helpsteer2",
|
| 2311 |
+
"id": "jevals-helpsteer2:helpsteer2-926",
|
| 2312 |
+
"qid": "decision",
|
| 2313 |
+
"type": "score",
|
| 2314 |
+
"pick": "4",
|
| 2315 |
+
"pick_run": "4",
|
| 2316 |
+
"same": true,
|
| 2317 |
+
"max_abs_dp": 0.0032809732131958147
|
| 2318 |
+
},
|
| 2319 |
+
{
|
| 2320 |
+
"set": "jevals-helpsteer2",
|
| 2321 |
+
"id": "jevals-helpsteer2:helpsteer2-369",
|
| 2322 |
+
"qid": "decision",
|
| 2323 |
+
"type": "score",
|
| 2324 |
+
"pick": "4",
|
| 2325 |
+
"pick_run": "4",
|
| 2326 |
+
"same": true,
|
| 2327 |
+
"max_abs_dp": 0.003506187999725352
|
| 2328 |
+
},
|
| 2329 |
+
{
|
| 2330 |
+
"set": "jevals-helpsteer2",
|
| 2331 |
+
"id": "jevals-helpsteer2:helpsteer2-648",
|
| 2332 |
+
"qid": "decision",
|
| 2333 |
+
"type": "score",
|
| 2334 |
+
"pick": "3",
|
| 2335 |
+
"pick_run": "3",
|
| 2336 |
+
"same": true,
|
| 2337 |
+
"max_abs_dp": 0.001363708106040959
|
| 2338 |
+
},
|
| 2339 |
+
{
|
| 2340 |
+
"set": "jevals-helpsteer2",
|
| 2341 |
+
"id": "jevals-helpsteer2:helpsteer2-87",
|
| 2342 |
+
"qid": "decision",
|
| 2343 |
+
"type": "score",
|
| 2344 |
+
"pick": "1",
|
| 2345 |
+
"pick_run": "1",
|
| 2346 |
+
"same": true,
|
| 2347 |
+
"max_abs_dp": 0.002428788396835302
|
| 2348 |
+
},
|
| 2349 |
+
{
|
| 2350 |
+
"set": "jevals-helpsteer2",
|
| 2351 |
+
"id": "jevals-helpsteer2:helpsteer2-727",
|
| 2352 |
+
"qid": "decision",
|
| 2353 |
+
"type": "score",
|
| 2354 |
+
"pick": "3",
|
| 2355 |
+
"pick_run": "3",
|
| 2356 |
+
"same": true,
|
| 2357 |
+
"max_abs_dp": 0.002152754644393934
|
| 2358 |
+
},
|
| 2359 |
+
{
|
| 2360 |
+
"set": "jevals-helpsteer2",
|
| 2361 |
+
"id": "jevals-helpsteer2:helpsteer2-401",
|
| 2362 |
+
"qid": "decision",
|
| 2363 |
+
"type": "score",
|
| 2364 |
+
"pick": "4",
|
| 2365 |
+
"pick_run": "4",
|
| 2366 |
+
"same": true,
|
| 2367 |
+
"max_abs_dp": 0.004214124761581395
|
| 2368 |
+
},
|
| 2369 |
+
{
|
| 2370 |
+
"set": "jevals-helpsteer2",
|
| 2371 |
+
"id": "jevals-helpsteer2:helpsteer2-265",
|
| 2372 |
+
"qid": "decision",
|
| 2373 |
+
"type": "score",
|
| 2374 |
+
"pick": "0",
|
| 2375 |
+
"pick_run": "0",
|
| 2376 |
+
"same": true,
|
| 2377 |
+
"max_abs_dp": 0.0055157948646545885
|
| 2378 |
+
},
|
| 2379 |
+
{
|
| 2380 |
+
"set": "jevals-helpsteer2",
|
| 2381 |
+
"id": "jevals-helpsteer2:helpsteer2-780",
|
| 2382 |
+
"qid": "decision",
|
| 2383 |
+
"type": "score",
|
| 2384 |
+
"pick": "1",
|
| 2385 |
+
"pick_run": "1",
|
| 2386 |
+
"same": true,
|
| 2387 |
+
"max_abs_dp": 0.001469104606628413
|
| 2388 |
+
},
|
| 2389 |
+
{
|
| 2390 |
+
"set": "jevals-helpsteer2",
|
| 2391 |
+
"id": "jevals-helpsteer2:helpsteer2-899",
|
| 2392 |
+
"qid": "decision",
|
| 2393 |
+
"type": "score",
|
| 2394 |
+
"pick": "4",
|
| 2395 |
+
"pick_run": "4",
|
| 2396 |
+
"same": true,
|
| 2397 |
+
"max_abs_dp": 0.0003629968891143698
|
| 2398 |
+
},
|
| 2399 |
+
{
|
| 2400 |
+
"set": "nimble-eval",
|
| 2401 |
+
"id": "nimble-eval:fast-41-diverse-211-020-base",
|
| 2402 |
+
"qid": "decision",
|
| 2403 |
+
"type": "choice",
|
| 2404 |
+
"pick": "city_no_r3",
|
| 2405 |
+
"pick_run": "city_no_r3",
|
| 2406 |
+
"same": true,
|
| 2407 |
+
"max_abs_dp": 0.0024864998931884985
|
| 2408 |
+
},
|
| 2409 |
+
{
|
| 2410 |
+
"set": "nimble-eval",
|
| 2411 |
+
"id": "nimble-eval:scale-diverse-174-003-base",
|
| 2412 |
+
"qid": "decision",
|
| 2413 |
+
"type": "score",
|
| 2414 |
+
"pick": "3",
|
| 2415 |
+
"pick_run": "3",
|
| 2416 |
+
"same": true,
|
| 2417 |
+
"max_abs_dp": 0.002774182624816879
|
| 2418 |
+
},
|
| 2419 |
+
{
|
| 2420 |
+
"set": "nimble-eval",
|
| 2421 |
+
"id": "nimble-eval:fast-43-diverse-169-065-counterfactual",
|
| 2422 |
+
"qid": "decision",
|
| 2423 |
+
"type": "choice",
|
| 2424 |
+
"pick": "deny_low",
|
| 2425 |
+
"pick_run": "deny_low",
|
| 2426 |
+
"same": true,
|
| 2427 |
+
"max_abs_dp": 0.001843123527526802
|
| 2428 |
+
},
|
| 2429 |
+
{
|
| 2430 |
+
"set": "nimble-eval",
|
| 2431 |
+
"id": "nimble-eval:scale-diverse-173-001-base",
|
| 2432 |
+
"qid": "decision",
|
| 2433 |
+
"type": "score",
|
| 2434 |
+
"pick": "2",
|
| 2435 |
+
"pick_run": "2",
|
| 2436 |
+
"same": true,
|
| 2437 |
+
"max_abs_dp": 0.0012800111370086797
|
| 2438 |
+
},
|
| 2439 |
+
{
|
| 2440 |
+
"set": "nimble-eval",
|
| 2441 |
+
"id": "nimble-eval:fast-43-diverse-079-037-counterfactual",
|
| 2442 |
+
"qid": "decision",
|
| 2443 |
+
"type": "choice",
|
| 2444 |
+
"pick": "Proceed \u2014 Medium; route camera setup to teaching assistant",
|
| 2445 |
+
"pick_run": "Proceed \u2014 Medium; route camera setup to teaching assistant",
|
| 2446 |
+
"same": true,
|
| 2447 |
+
"max_abs_dp": 0.008462921947479218
|
| 2448 |
+
},
|
| 2449 |
+
{
|
| 2450 |
+
"set": "nimble-eval",
|
| 2451 |
+
"id": "nimble-eval:fast-43-diverse-171-008-counterfactual",
|
| 2452 |
+
"qid": "decision",
|
| 2453 |
+
"type": "noul",
|
| 2454 |
+
"pick": "false",
|
| 2455 |
+
"pick_run": "false",
|
| 2456 |
+
"same": true,
|
| 2457 |
+
"max_abs_dp": 0.0006690865592956508
|
| 2458 |
+
},
|
| 2459 |
+
{
|
| 2460 |
+
"set": "nimble-eval",
|
| 2461 |
+
"id": "nimble-eval:fast-43-diverse-174-011-base",
|
| 2462 |
+
"qid": "decision",
|
| 2463 |
+
"type": "score",
|
| 2464 |
+
"pick": "2",
|
| 2465 |
+
"pick_run": "2",
|
| 2466 |
+
"same": true,
|
| 2467 |
+
"max_abs_dp": 0.0037998003120422452
|
| 2468 |
+
},
|
| 2469 |
+
{
|
| 2470 |
+
"set": "nimble-eval",
|
| 2471 |
+
"id": "nimble-eval:fast-41-diverse-009-004-base",
|
| 2472 |
+
"qid": "decision",
|
| 2473 |
+
"type": "noul",
|
| 2474 |
+
"pick": "true",
|
| 2475 |
+
"pick_run": "true",
|
| 2476 |
+
"same": true,
|
| 2477 |
+
"max_abs_dp": 0.00020508449935907613
|
| 2478 |
+
},
|
| 2479 |
+
{
|
| 2480 |
+
"set": "nimble-eval",
|
| 2481 |
+
"id": "nimble-eval:fast-43-diverse-171-010-base",
|
| 2482 |
+
"qid": "decision",
|
| 2483 |
+
"type": "noul",
|
| 2484 |
+
"pick": "true",
|
| 2485 |
+
"pick_run": "true",
|
| 2486 |
+
"same": true,
|
| 2487 |
+
"max_abs_dp": 0.0001036642386466266
|
| 2488 |
+
},
|
| 2489 |
+
{
|
| 2490 |
+
"set": "nimble-eval",
|
| 2491 |
+
"id": "nimble-eval:fast-43-diverse-169-012-counterfactual",
|
| 2492 |
+
"qid": "decision",
|
| 2493 |
+
"type": "choice",
|
| 2494 |
+
"pick": "deny_low",
|
| 2495 |
+
"pick_run": "deny_low",
|
| 2496 |
+
"same": true,
|
| 2497 |
+
"max_abs_dp": 0.001906281406402588
|
| 2498 |
+
},
|
| 2499 |
+
{
|
| 2500 |
+
"set": "nimble-eval",
|
| 2501 |
+
"id": "nimble-eval:fast-41-diverse-275-013-counterfactual",
|
| 2502 |
+
"qid": "decision",
|
| 2503 |
+
"type": "score",
|
| 2504 |
+
"pick": "0",
|
| 2505 |
+
"pick_run": "0",
|
| 2506 |
+
"same": true,
|
| 2507 |
+
"max_abs_dp": 0.0015062385268211353
|
| 2508 |
+
},
|
| 2509 |
+
{
|
| 2510 |
+
"set": "nimble-eval",
|
| 2511 |
+
"id": "nimble-eval:scale-diverse-097-005-base",
|
| 2512 |
+
"qid": "decision",
|
| 2513 |
+
"type": "choice",
|
| 2514 |
+
"pick": "assign_420_as_best_fit",
|
| 2515 |
+
"pick_run": "assign_420_as_best_fit",
|
| 2516 |
+
"same": true,
|
| 2517 |
+
"max_abs_dp": 0.001358479156494119
|
| 2518 |
+
},
|
| 2519 |
+
{
|
| 2520 |
+
"set": "nimble-eval",
|
| 2521 |
+
"id": "nimble-eval:fast-43-diverse-169-068-base",
|
| 2522 |
+
"qid": "decision",
|
| 2523 |
+
"type": "choice",
|
| 2524 |
+
"pick": "release_high",
|
| 2525 |
+
"pick_run": "release_high",
|
| 2526 |
+
"same": true,
|
| 2527 |
+
"max_abs_dp": 0.005447728443145783
|
| 2528 |
+
},
|
| 2529 |
+
{
|
| 2530 |
+
"set": "nimble-eval",
|
| 2531 |
+
"id": "nimble-eval:fast-41-diverse-079-012-counterfactual",
|
| 2532 |
+
"qid": "decision",
|
| 2533 |
+
"type": "choice",
|
| 2534 |
+
"pick": "Proceed \u2014 High; no remaining routing",
|
| 2535 |
+
"pick_run": "Proceed \u2014 High; no remaining routing",
|
| 2536 |
+
"same": true,
|
| 2537 |
+
"max_abs_dp": 0.005806731111526464
|
| 2538 |
+
},
|
| 2539 |
+
{
|
| 2540 |
+
"set": "nimble-eval",
|
| 2541 |
+
"id": "nimble-eval:scale-diverse-007-005-base",
|
| 2542 |
+
"qid": "decision",
|
| 2543 |
+
"type": "choice",
|
| 2544 |
+
"pick": "major_seller_correction_not_ready",
|
| 2545 |
+
"pick_run": "major_seller_correction_not_ready",
|
| 2546 |
+
"same": true,
|
| 2547 |
+
"max_abs_dp": 0.0009021766853332114
|
| 2548 |
+
},
|
| 2549 |
+
{
|
| 2550 |
+
"set": "nimble-eval",
|
| 2551 |
+
"id": "nimble-eval:fast-41-diverse-171-003-counterfactual",
|
| 2552 |
+
"qid": "decision",
|
| 2553 |
+
"type": "noul",
|
| 2554 |
+
"pick": "false",
|
| 2555 |
+
"pick_run": "false",
|
| 2556 |
+
"same": true,
|
| 2557 |
+
"max_abs_dp": 0.00016909534835818363
|
| 2558 |
+
},
|
| 2559 |
+
{
|
| 2560 |
+
"set": "nimble-eval",
|
| 2561 |
+
"id": "nimble-eval:fast-41-diverse-212-032-counterfactual",
|
| 2562 |
+
"qid": "decision",
|
| 2563 |
+
"type": "choice",
|
| 2564 |
+
"pick": "travel_advance_3",
|
| 2565 |
+
"pick_run": "travel_advance_3",
|
| 2566 |
+
"same": true,
|
| 2567 |
+
"max_abs_dp": 0.0012396228532791098
|
| 2568 |
+
},
|
| 2569 |
+
{
|
| 2570 |
+
"set": "nimble-eval",
|
| 2571 |
+
"id": "nimble-eval:fast-41-diverse-009-004-counterfactual",
|
| 2572 |
+
"qid": "decision",
|
| 2573 |
+
"type": "noul",
|
| 2574 |
+
"pick": "false",
|
| 2575 |
+
"pick_run": "false",
|
| 2576 |
+
"same": true,
|
| 2577 |
+
"max_abs_dp": 0.0009295430221557544
|
| 2578 |
+
},
|
| 2579 |
+
{
|
| 2580 |
+
"set": "nimble-eval",
|
| 2581 |
+
"id": "nimble-eval:fast-43-diverse-174-010-counterfactual",
|
| 2582 |
+
"qid": "decision",
|
| 2583 |
+
"type": "score",
|
| 2584 |
+
"pick": "0",
|
| 2585 |
+
"pick_run": "0",
|
| 2586 |
+
"same": true,
|
| 2587 |
+
"max_abs_dp": 0.0011149642276763738
|
| 2588 |
+
},
|
| 2589 |
+
{
|
| 2590 |
+
"set": "nimble-eval",
|
| 2591 |
+
"id": "nimble-eval:fast-43-diverse-174-006-counterfactual",
|
| 2592 |
+
"qid": "decision",
|
| 2593 |
+
"type": "score",
|
| 2594 |
+
"pick": "1",
|
| 2595 |
+
"pick_run": "1",
|
| 2596 |
+
"same": true,
|
| 2597 |
+
"max_abs_dp": 0.0016253944072723292
|
| 2598 |
+
},
|
| 2599 |
+
{
|
| 2600 |
+
"set": "nimble-eval",
|
| 2601 |
+
"id": "nimble-eval:fast-41-diverse-009-001-base",
|
| 2602 |
+
"qid": "decision",
|
| 2603 |
+
"type": "noul",
|
| 2604 |
+
"pick": "true",
|
| 2605 |
+
"pick_run": "true",
|
| 2606 |
+
"same": true,
|
| 2607 |
+
"max_abs_dp": 0.00025924222946172026
|
| 2608 |
+
},
|
| 2609 |
+
{
|
| 2610 |
+
"set": "nimble-eval",
|
| 2611 |
+
"id": "nimble-eval:fast-43-diverse-079-058-base",
|
| 2612 |
+
"qid": "decision",
|
| 2613 |
+
"type": "choice",
|
| 2614 |
+
"pick": "Proceed \u2014 High; no remaining routing",
|
| 2615 |
+
"pick_run": "Proceed \u2014 High; no remaining routing",
|
| 2616 |
+
"same": true,
|
| 2617 |
+
"max_abs_dp": 0.0026827917366027965
|
| 2618 |
+
},
|
| 2619 |
+
{
|
| 2620 |
+
"set": "nimble-eval",
|
| 2621 |
+
"id": "nimble-eval:fast-43-diverse-171-016-base",
|
| 2622 |
+
"qid": "decision",
|
| 2623 |
+
"type": "noul",
|
| 2624 |
+
"pick": "true",
|
| 2625 |
+
"pick_run": "true",
|
| 2626 |
+
"same": true,
|
| 2627 |
+
"max_abs_dp": 3.662064181268222e-05
|
| 2628 |
+
},
|
| 2629 |
+
{
|
| 2630 |
+
"set": "nimble-eval",
|
| 2631 |
+
"id": "nimble-eval:scale-diverse-007-004-base",
|
| 2632 |
+
"qid": "decision",
|
| 2633 |
+
"type": "choice",
|
| 2634 |
+
"pick": "major_seller_correction_not_ready",
|
| 2635 |
+
"pick_run": "major_seller_correction_not_ready",
|
| 2636 |
+
"same": true,
|
| 2637 |
+
"max_abs_dp": 0.001442355716705368
|
| 2638 |
+
},
|
| 2639 |
+
{
|
| 2640 |
+
"set": "nimble-eval",
|
| 2641 |
+
"id": "nimble-eval:scale-diverse-171-001-base",
|
| 2642 |
+
"qid": "decision",
|
| 2643 |
+
"type": "noul",
|
| 2644 |
+
"pick": "true",
|
| 2645 |
+
"pick_run": "true",
|
| 2646 |
+
"same": true,
|
| 2647 |
+
"max_abs_dp": 0.0028766922264099604
|
| 2648 |
+
},
|
| 2649 |
+
{
|
| 2650 |
+
"set": "nimble-eval",
|
| 2651 |
+
"id": "nimble-eval:scale-diverse-275-004-counterfactual",
|
| 2652 |
+
"qid": "decision",
|
| 2653 |
+
"type": "score",
|
| 2654 |
+
"pick": "0",
|
| 2655 |
+
"pick_run": "0",
|
| 2656 |
+
"same": true,
|
| 2657 |
+
"max_abs_dp": 0.002890404029846194
|
| 2658 |
+
},
|
| 2659 |
+
{
|
| 2660 |
+
"set": "nimble-eval",
|
| 2661 |
+
"id": "nimble-eval:fast-43-diverse-169-053-counterfactual",
|
| 2662 |
+
"qid": "decision",
|
| 2663 |
+
"type": "choice",
|
| 2664 |
+
"pick": "deny_low",
|
| 2665 |
+
"pick_run": "deny_low",
|
| 2666 |
+
"same": true,
|
| 2667 |
+
"max_abs_dp": 0.004763497699737496
|
| 2668 |
+
},
|
| 2669 |
+
{
|
| 2670 |
+
"set": "nimble-eval",
|
| 2671 |
+
"id": "nimble-eval:fast-41-diverse-079-003-base",
|
| 2672 |
+
"qid": "decision",
|
| 2673 |
+
"type": "choice",
|
| 2674 |
+
"pick": "Proceed \u2014 High; no remaining routing",
|
| 2675 |
+
"pick_run": "Proceed \u2014 High; no remaining routing",
|
| 2676 |
+
"same": true,
|
| 2677 |
+
"max_abs_dp": 0.0015070434989929438
|
| 2678 |
+
},
|
| 2679 |
+
{
|
| 2680 |
+
"set": "nimble-eval",
|
| 2681 |
+
"id": "nimble-eval:fast-43-diverse-171-013-base",
|
| 2682 |
+
"qid": "decision",
|
| 2683 |
+
"type": "noul",
|
| 2684 |
+
"pick": "true",
|
| 2685 |
+
"pick_run": "true",
|
| 2686 |
+
"same": true,
|
| 2687 |
+
"max_abs_dp": 0.00013033839035037076
|
| 2688 |
+
},
|
| 2689 |
+
{
|
| 2690 |
+
"set": "nimble-eval",
|
| 2691 |
+
"id": "nimble-eval:fast-43-diverse-079-015-counterfactual",
|
| 2692 |
+
"qid": "decision",
|
| 2693 |
+
"type": "choice",
|
| 2694 |
+
"pick": "Proceed \u2014 High; no remaining routing",
|
| 2695 |
+
"pick_run": "Proceed \u2014 High; no remaining routing",
|
| 2696 |
+
"same": true,
|
| 2697 |
+
"max_abs_dp": 0.0023361680927276862
|
| 2698 |
+
}
|
| 2699 |
+
]
|
| 2700 |
+
}
|
scripts/decide_standalone.py
CHANGED
|
@@ -26,6 +26,9 @@ def main():
|
|
| 26 |
ap.add_argument("--attn", default="sdpa")
|
| 27 |
a = ap.parse_args()
|
| 28 |
req = json.load(open(a.request))
|
|
|
|
|
|
|
|
|
|
| 29 |
dtype = torch.bfloat16 if a.device.startswith(("cuda", "mps")) else torch.float32
|
| 30 |
model_dir = a.model
|
| 31 |
if not os.path.isdir(model_dir):
|
|
|
|
| 26 |
ap.add_argument("--attn", default="sdpa")
|
| 27 |
a = ap.parse_args()
|
| 28 |
req = json.load(open(a.request))
|
| 29 |
+
if a.device.startswith("mps"):
|
| 30 |
+
# transformers 5.17's threaded weight loader intermittently segfaults moving tensors to MPS
|
| 31 |
+
os.environ.setdefault("HF_DEACTIVATE_ASYNC_LOAD", "1")
|
| 32 |
dtype = torch.bfloat16 if a.device.startswith(("cuda", "mps")) else torch.float32
|
| 33 |
model_dir = a.model
|
| 34 |
if not os.path.isdir(model_dir):
|