Upload folder using huggingface_hub
Browse files- README.md +28 -0
- checkpoint-4916/config.json +32 -0
- checkpoint-4916/model.safetensors +3 -0
- checkpoint-4916/optimizer.pt +3 -0
- checkpoint-4916/rng_state_0.pth +3 -0
- checkpoint-4916/rng_state_1.pth +3 -0
- checkpoint-4916/rng_state_2.pth +3 -0
- checkpoint-4916/rng_state_3.pth +3 -0
- checkpoint-4916/scheduler.pt +3 -0
- checkpoint-4916/trainer_state.json +1440 -0
- checkpoint-4916/training_args.bin +3 -0
- config.json +32 -0
- model.safetensors +3 -0
- runs/Feb22_02-13-45_r-samtuckervegan-autotrain-4xl4-41bfog4i-7c075-ypwih/events.out.tfevents.1740190427.r-samtuckervegan-autotrain-4xl4-41bfog4i-7c075-ypwih.208.0 +2 -2
- runs/Feb22_02-13-45_r-samtuckervegan-autotrain-4xl4-41bfog4i-7c075-ypwih/events.out.tfevents.1740191139.r-samtuckervegan-autotrain-4xl4-41bfog4i-7c075-ypwih.208.1 +3 -0
- special_tokens_map.json +7 -0
- tokenizer.json +0 -0
- tokenizer_config.json +56 -0
- training_args.bin +3 -0
- training_params.json +30 -0
- vocab.txt +0 -0
README.md
ADDED
|
@@ -0,0 +1,28 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
|
| 2 |
+
---
|
| 3 |
+
tags:
|
| 4 |
+
- autotrain
|
| 5 |
+
- text-regression
|
| 6 |
+
base_model: distilbert/distilbert-base-uncased
|
| 7 |
+
widget:
|
| 8 |
+
- text: "I love AutoTrain"
|
| 9 |
+
datasets:
|
| 10 |
+
- samtuckervegan/text_performance
|
| 11 |
+
---
|
| 12 |
+
|
| 13 |
+
# Model Trained Using AutoTrain
|
| 14 |
+
|
| 15 |
+
- Problem type: Text Regression
|
| 16 |
+
|
| 17 |
+
## Validation Metrics
|
| 18 |
+
loss: 0.03380444645881653
|
| 19 |
+
|
| 20 |
+
mse: 0.033803146332502365
|
| 21 |
+
|
| 22 |
+
mae: 0.14292894303798676
|
| 23 |
+
|
| 24 |
+
r2: 0.2760580778121948
|
| 25 |
+
|
| 26 |
+
rmse: 0.18385631980571776
|
| 27 |
+
|
| 28 |
+
explained_variance: 0.2763633728027344
|
checkpoint-4916/config.json
ADDED
|
@@ -0,0 +1,32 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "distilbert/distilbert-base-uncased",
|
| 3 |
+
"_num_labels": 1,
|
| 4 |
+
"activation": "gelu",
|
| 5 |
+
"architectures": [
|
| 6 |
+
"DistilBertForSequenceClassification"
|
| 7 |
+
],
|
| 8 |
+
"attention_dropout": 0.1,
|
| 9 |
+
"dim": 768,
|
| 10 |
+
"dropout": 0.1,
|
| 11 |
+
"hidden_dim": 3072,
|
| 12 |
+
"id2label": {
|
| 13 |
+
"0": "target"
|
| 14 |
+
},
|
| 15 |
+
"initializer_range": 0.02,
|
| 16 |
+
"label2id": {
|
| 17 |
+
"target": 0
|
| 18 |
+
},
|
| 19 |
+
"max_position_embeddings": 512,
|
| 20 |
+
"model_type": "distilbert",
|
| 21 |
+
"n_heads": 12,
|
| 22 |
+
"n_layers": 6,
|
| 23 |
+
"pad_token_id": 0,
|
| 24 |
+
"problem_type": "regression",
|
| 25 |
+
"qa_dropout": 0.1,
|
| 26 |
+
"seq_classif_dropout": 0.2,
|
| 27 |
+
"sinusoidal_pos_embds": false,
|
| 28 |
+
"tie_weights_": true,
|
| 29 |
+
"torch_dtype": "float32",
|
| 30 |
+
"transformers_version": "4.48.0",
|
| 31 |
+
"vocab_size": 30522
|
| 32 |
+
}
|
checkpoint-4916/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:52992a6b1bb7b233e4054bc086b088013d7e7613bcffead944d24fd70d64c55e
|
| 3 |
+
size 267829484
|
checkpoint-4916/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:64c1456eb332d082a347d7cad4735bc4250845738e02e04a63fa9f17e819942a
|
| 3 |
+
size 535721146
|
checkpoint-4916/rng_state_0.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c8fac6987c200600e610946b53cf319ed29c1564951f6e89145ce16858af5139
|
| 3 |
+
size 15024
|
checkpoint-4916/rng_state_1.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7b0f8befb7da5855d4149667159792c7f547534617d1616470ae82b4e1c4af49
|
| 3 |
+
size 15024
|
checkpoint-4916/rng_state_2.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:aa0ca0b967f2e548cdf3eee611a80ad48edb340060335df03d47e61c93cbc6aa
|
| 3 |
+
size 15024
|
checkpoint-4916/rng_state_3.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:754aa36cc24cc6e3f8bda45f9c465b2a33ed0e611d1fa8443bdfc2f99b182652
|
| 3 |
+
size 15024
|
checkpoint-4916/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b39d933709e10bf97e1b41f5aee68f3dcc996b1173e11cc1c55ae6d677f967fa
|
| 3 |
+
size 1064
|
checkpoint-4916/trainer_state.json
ADDED
|
@@ -0,0 +1,1440 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.03380444645881653,
|
| 3 |
+
"best_model_checkpoint": "textprediction/checkpoint-4916",
|
| 4 |
+
"epoch": 2.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 4916,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.01017087062652563,
|
| 13 |
+
"grad_norm": 4.562432289123535,
|
| 14 |
+
"learning_rate": 1.6260162601626018e-06,
|
| 15 |
+
"loss": 0.1968,
|
| 16 |
+
"step": 25
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.02034174125305126,
|
| 20 |
+
"grad_norm": 1.4055429697036743,
|
| 21 |
+
"learning_rate": 3.3197831978319785e-06,
|
| 22 |
+
"loss": 0.0977,
|
| 23 |
+
"step": 50
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.030512611879576892,
|
| 27 |
+
"grad_norm": 0.8549578189849854,
|
| 28 |
+
"learning_rate": 5.013550135501355e-06,
|
| 29 |
+
"loss": 0.0498,
|
| 30 |
+
"step": 75
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.04068348250610252,
|
| 34 |
+
"grad_norm": 0.41762399673461914,
|
| 35 |
+
"learning_rate": 6.707317073170733e-06,
|
| 36 |
+
"loss": 0.0469,
|
| 37 |
+
"step": 100
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.050854353132628156,
|
| 41 |
+
"grad_norm": 0.4172782897949219,
|
| 42 |
+
"learning_rate": 8.401084010840109e-06,
|
| 43 |
+
"loss": 0.0487,
|
| 44 |
+
"step": 125
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.061025223759153785,
|
| 48 |
+
"grad_norm": 0.45810267329216003,
|
| 49 |
+
"learning_rate": 1.0094850948509485e-05,
|
| 50 |
+
"loss": 0.0449,
|
| 51 |
+
"step": 150
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.07119609438567942,
|
| 55 |
+
"grad_norm": 0.41910672187805176,
|
| 56 |
+
"learning_rate": 1.1788617886178862e-05,
|
| 57 |
+
"loss": 0.0417,
|
| 58 |
+
"step": 175
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.08136696501220504,
|
| 62 |
+
"grad_norm": 0.36320051550865173,
|
| 63 |
+
"learning_rate": 1.348238482384824e-05,
|
| 64 |
+
"loss": 0.0421,
|
| 65 |
+
"step": 200
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.09153783563873068,
|
| 69 |
+
"grad_norm": 0.434171199798584,
|
| 70 |
+
"learning_rate": 1.5176151761517615e-05,
|
| 71 |
+
"loss": 0.0465,
|
| 72 |
+
"step": 225
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.10170870626525631,
|
| 76 |
+
"grad_norm": 0.3281106948852539,
|
| 77 |
+
"learning_rate": 1.6869918699186994e-05,
|
| 78 |
+
"loss": 0.0433,
|
| 79 |
+
"step": 250
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.11187957689178193,
|
| 83 |
+
"grad_norm": 0.36068961024284363,
|
| 84 |
+
"learning_rate": 1.856368563685637e-05,
|
| 85 |
+
"loss": 0.0439,
|
| 86 |
+
"step": 275
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.12205044751830757,
|
| 90 |
+
"grad_norm": 0.5031670928001404,
|
| 91 |
+
"learning_rate": 2.0257452574525744e-05,
|
| 92 |
+
"loss": 0.0423,
|
| 93 |
+
"step": 300
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.1322213181448332,
|
| 97 |
+
"grad_norm": 1.1825827360153198,
|
| 98 |
+
"learning_rate": 2.1951219512195124e-05,
|
| 99 |
+
"loss": 0.0427,
|
| 100 |
+
"step": 325
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.14239218877135884,
|
| 104 |
+
"grad_norm": 0.3233006000518799,
|
| 105 |
+
"learning_rate": 2.36449864498645e-05,
|
| 106 |
+
"loss": 0.0434,
|
| 107 |
+
"step": 350
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.15256305939788445,
|
| 111 |
+
"grad_norm": 0.4184306561946869,
|
| 112 |
+
"learning_rate": 2.5338753387533877e-05,
|
| 113 |
+
"loss": 0.0423,
|
| 114 |
+
"step": 375
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.16273393002441008,
|
| 118 |
+
"grad_norm": 0.38505345582962036,
|
| 119 |
+
"learning_rate": 2.7032520325203254e-05,
|
| 120 |
+
"loss": 0.0415,
|
| 121 |
+
"step": 400
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.17290480065093572,
|
| 125 |
+
"grad_norm": 0.3058652877807617,
|
| 126 |
+
"learning_rate": 2.872628726287263e-05,
|
| 127 |
+
"loss": 0.0431,
|
| 128 |
+
"step": 425
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.18307567127746135,
|
| 132 |
+
"grad_norm": 0.43836578726768494,
|
| 133 |
+
"learning_rate": 3.0420054200542007e-05,
|
| 134 |
+
"loss": 0.0417,
|
| 135 |
+
"step": 450
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.193246541903987,
|
| 139 |
+
"grad_norm": 0.5372226238250732,
|
| 140 |
+
"learning_rate": 3.2113821138211384e-05,
|
| 141 |
+
"loss": 0.0387,
|
| 142 |
+
"step": 475
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.20341741253051263,
|
| 146 |
+
"grad_norm": 0.4316234886646271,
|
| 147 |
+
"learning_rate": 3.380758807588076e-05,
|
| 148 |
+
"loss": 0.0425,
|
| 149 |
+
"step": 500
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.21358828315703823,
|
| 153 |
+
"grad_norm": 0.4142661988735199,
|
| 154 |
+
"learning_rate": 3.550135501355014e-05,
|
| 155 |
+
"loss": 0.0416,
|
| 156 |
+
"step": 525
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.22375915378356387,
|
| 160 |
+
"grad_norm": 0.2900916635990143,
|
| 161 |
+
"learning_rate": 3.7195121951219514e-05,
|
| 162 |
+
"loss": 0.044,
|
| 163 |
+
"step": 550
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.2339300244100895,
|
| 167 |
+
"grad_norm": 0.6529517769813538,
|
| 168 |
+
"learning_rate": 3.888888888888889e-05,
|
| 169 |
+
"loss": 0.0395,
|
| 170 |
+
"step": 575
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.24410089503661514,
|
| 174 |
+
"grad_norm": 0.4228176176548004,
|
| 175 |
+
"learning_rate": 4.058265582655827e-05,
|
| 176 |
+
"loss": 0.0379,
|
| 177 |
+
"step": 600
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.25427176566314075,
|
| 181 |
+
"grad_norm": 0.4153424799442291,
|
| 182 |
+
"learning_rate": 4.2276422764227644e-05,
|
| 183 |
+
"loss": 0.0419,
|
| 184 |
+
"step": 625
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.2644426362896664,
|
| 188 |
+
"grad_norm": 0.5038943290710449,
|
| 189 |
+
"learning_rate": 4.397018970189702e-05,
|
| 190 |
+
"loss": 0.0426,
|
| 191 |
+
"step": 650
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.274613506916192,
|
| 195 |
+
"grad_norm": 0.518610417842865,
|
| 196 |
+
"learning_rate": 4.56639566395664e-05,
|
| 197 |
+
"loss": 0.0415,
|
| 198 |
+
"step": 675
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.2847843775427177,
|
| 202 |
+
"grad_norm": 0.22950297594070435,
|
| 203 |
+
"learning_rate": 4.7357723577235774e-05,
|
| 204 |
+
"loss": 0.0401,
|
| 205 |
+
"step": 700
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.2949552481692433,
|
| 209 |
+
"grad_norm": 0.42790016531944275,
|
| 210 |
+
"learning_rate": 4.905149051490515e-05,
|
| 211 |
+
"loss": 0.0431,
|
| 212 |
+
"step": 725
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.3051261187957689,
|
| 216 |
+
"grad_norm": 0.24033579230308533,
|
| 217 |
+
"learning_rate": 4.991711874623267e-05,
|
| 218 |
+
"loss": 0.0436,
|
| 219 |
+
"step": 750
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.31529698942229456,
|
| 223 |
+
"grad_norm": 0.26290884613990784,
|
| 224 |
+
"learning_rate": 4.972875226039783e-05,
|
| 225 |
+
"loss": 0.0439,
|
| 226 |
+
"step": 775
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.32546786004882017,
|
| 230 |
+
"grad_norm": 0.20243217051029205,
|
| 231 |
+
"learning_rate": 4.9540385774562993e-05,
|
| 232 |
+
"loss": 0.04,
|
| 233 |
+
"step": 800
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.33563873067534583,
|
| 237 |
+
"grad_norm": 0.17254669964313507,
|
| 238 |
+
"learning_rate": 4.935201928872815e-05,
|
| 239 |
+
"loss": 0.0405,
|
| 240 |
+
"step": 825
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.34580960130187144,
|
| 244 |
+
"grad_norm": 0.2651669681072235,
|
| 245 |
+
"learning_rate": 4.916365280289331e-05,
|
| 246 |
+
"loss": 0.0376,
|
| 247 |
+
"step": 850
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.35598047192839705,
|
| 251 |
+
"grad_norm": 0.4401779770851135,
|
| 252 |
+
"learning_rate": 4.897528631705847e-05,
|
| 253 |
+
"loss": 0.038,
|
| 254 |
+
"step": 875
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.3661513425549227,
|
| 258 |
+
"grad_norm": 0.23679669201374054,
|
| 259 |
+
"learning_rate": 4.878691983122363e-05,
|
| 260 |
+
"loss": 0.0417,
|
| 261 |
+
"step": 900
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.3763222131814483,
|
| 265 |
+
"grad_norm": 0.2824667990207672,
|
| 266 |
+
"learning_rate": 4.8598553345388795e-05,
|
| 267 |
+
"loss": 0.0394,
|
| 268 |
+
"step": 925
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.386493083807974,
|
| 272 |
+
"grad_norm": 0.24676434695720673,
|
| 273 |
+
"learning_rate": 4.841018685955395e-05,
|
| 274 |
+
"loss": 0.0375,
|
| 275 |
+
"step": 950
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.3966639544344996,
|
| 279 |
+
"grad_norm": 0.3242221176624298,
|
| 280 |
+
"learning_rate": 4.822182037371911e-05,
|
| 281 |
+
"loss": 0.0384,
|
| 282 |
+
"step": 975
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.40683482506102525,
|
| 286 |
+
"grad_norm": 0.4461381733417511,
|
| 287 |
+
"learning_rate": 4.8033453887884274e-05,
|
| 288 |
+
"loss": 0.0376,
|
| 289 |
+
"step": 1000
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.41700569568755086,
|
| 293 |
+
"grad_norm": 0.3799073398113251,
|
| 294 |
+
"learning_rate": 4.784508740204943e-05,
|
| 295 |
+
"loss": 0.0415,
|
| 296 |
+
"step": 1025
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.42717656631407647,
|
| 300 |
+
"grad_norm": 0.21580813825130463,
|
| 301 |
+
"learning_rate": 4.765672091621459e-05,
|
| 302 |
+
"loss": 0.0389,
|
| 303 |
+
"step": 1050
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.43734743694060213,
|
| 307 |
+
"grad_norm": 0.27040156722068787,
|
| 308 |
+
"learning_rate": 4.7468354430379746e-05,
|
| 309 |
+
"loss": 0.0393,
|
| 310 |
+
"step": 1075
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.44751830756712774,
|
| 314 |
+
"grad_norm": 0.7307707071304321,
|
| 315 |
+
"learning_rate": 4.7279987944544904e-05,
|
| 316 |
+
"loss": 0.0411,
|
| 317 |
+
"step": 1100
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.4576891781936534,
|
| 321 |
+
"grad_norm": 0.29991018772125244,
|
| 322 |
+
"learning_rate": 4.709162145871007e-05,
|
| 323 |
+
"loss": 0.0396,
|
| 324 |
+
"step": 1125
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.467860048820179,
|
| 328 |
+
"grad_norm": 0.16599082946777344,
|
| 329 |
+
"learning_rate": 4.6903254972875226e-05,
|
| 330 |
+
"loss": 0.0385,
|
| 331 |
+
"step": 1150
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.4780309194467046,
|
| 335 |
+
"grad_norm": 0.23768579959869385,
|
| 336 |
+
"learning_rate": 4.671488848704039e-05,
|
| 337 |
+
"loss": 0.0391,
|
| 338 |
+
"step": 1175
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.4882017900732303,
|
| 342 |
+
"grad_norm": 0.43703967332839966,
|
| 343 |
+
"learning_rate": 4.652652200120555e-05,
|
| 344 |
+
"loss": 0.0371,
|
| 345 |
+
"step": 1200
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.4983726606997559,
|
| 349 |
+
"grad_norm": 0.17367880046367645,
|
| 350 |
+
"learning_rate": 4.6338155515370705e-05,
|
| 351 |
+
"loss": 0.0393,
|
| 352 |
+
"step": 1225
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.5085435313262815,
|
| 356 |
+
"grad_norm": 0.2584339678287506,
|
| 357 |
+
"learning_rate": 4.614978902953587e-05,
|
| 358 |
+
"loss": 0.0389,
|
| 359 |
+
"step": 1250
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.5187144019528072,
|
| 363 |
+
"grad_norm": 0.3418385088443756,
|
| 364 |
+
"learning_rate": 4.596142254370103e-05,
|
| 365 |
+
"loss": 0.0372,
|
| 366 |
+
"step": 1275
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.5288852725793328,
|
| 370 |
+
"grad_norm": 0.41003596782684326,
|
| 371 |
+
"learning_rate": 4.5773056057866184e-05,
|
| 372 |
+
"loss": 0.0358,
|
| 373 |
+
"step": 1300
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.5390561432058584,
|
| 377 |
+
"grad_norm": 0.5690125823020935,
|
| 378 |
+
"learning_rate": 4.558468957203135e-05,
|
| 379 |
+
"loss": 0.0424,
|
| 380 |
+
"step": 1325
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.549227013832384,
|
| 384 |
+
"grad_norm": 0.25894445180892944,
|
| 385 |
+
"learning_rate": 4.5396323086196506e-05,
|
| 386 |
+
"loss": 0.0404,
|
| 387 |
+
"step": 1350
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.5593978844589097,
|
| 391 |
+
"grad_norm": 0.3278766870498657,
|
| 392 |
+
"learning_rate": 4.520795660036167e-05,
|
| 393 |
+
"loss": 0.0394,
|
| 394 |
+
"step": 1375
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.5695687550854354,
|
| 398 |
+
"grad_norm": 0.2761504352092743,
|
| 399 |
+
"learning_rate": 4.501959011452683e-05,
|
| 400 |
+
"loss": 0.0376,
|
| 401 |
+
"step": 1400
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.5797396257119609,
|
| 405 |
+
"grad_norm": 0.1649962216615677,
|
| 406 |
+
"learning_rate": 4.4831223628691985e-05,
|
| 407 |
+
"loss": 0.0379,
|
| 408 |
+
"step": 1425
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.5899104963384866,
|
| 412 |
+
"grad_norm": 0.19504110515117645,
|
| 413 |
+
"learning_rate": 4.464285714285715e-05,
|
| 414 |
+
"loss": 0.0371,
|
| 415 |
+
"step": 1450
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.6000813669650122,
|
| 419 |
+
"grad_norm": 0.17408576607704163,
|
| 420 |
+
"learning_rate": 4.445449065702231e-05,
|
| 421 |
+
"loss": 0.0394,
|
| 422 |
+
"step": 1475
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.6102522375915378,
|
| 426 |
+
"grad_norm": 0.5323575139045715,
|
| 427 |
+
"learning_rate": 4.4266124171187465e-05,
|
| 428 |
+
"loss": 0.0397,
|
| 429 |
+
"step": 1500
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.6204231082180635,
|
| 433 |
+
"grad_norm": 0.6240959167480469,
|
| 434 |
+
"learning_rate": 4.407775768535262e-05,
|
| 435 |
+
"loss": 0.0382,
|
| 436 |
+
"step": 1525
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.6305939788445891,
|
| 440 |
+
"grad_norm": 0.34519892930984497,
|
| 441 |
+
"learning_rate": 4.388939119951778e-05,
|
| 442 |
+
"loss": 0.0412,
|
| 443 |
+
"step": 1550
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.6407648494711147,
|
| 447 |
+
"grad_norm": 0.6749323606491089,
|
| 448 |
+
"learning_rate": 4.3701024713682944e-05,
|
| 449 |
+
"loss": 0.039,
|
| 450 |
+
"step": 1575
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.6509357200976403,
|
| 454 |
+
"grad_norm": 0.3768059313297272,
|
| 455 |
+
"learning_rate": 4.35126582278481e-05,
|
| 456 |
+
"loss": 0.0386,
|
| 457 |
+
"step": 1600
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.661106590724166,
|
| 461 |
+
"grad_norm": 0.22068104147911072,
|
| 462 |
+
"learning_rate": 4.332429174201326e-05,
|
| 463 |
+
"loss": 0.0383,
|
| 464 |
+
"step": 1625
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.6712774613506917,
|
| 468 |
+
"grad_norm": 0.14016976952552795,
|
| 469 |
+
"learning_rate": 4.313592525617842e-05,
|
| 470 |
+
"loss": 0.039,
|
| 471 |
+
"step": 1650
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.6814483319772172,
|
| 475 |
+
"grad_norm": 0.17742925882339478,
|
| 476 |
+
"learning_rate": 4.294755877034358e-05,
|
| 477 |
+
"loss": 0.0369,
|
| 478 |
+
"step": 1675
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.6916192026037429,
|
| 482 |
+
"grad_norm": 0.1665901243686676,
|
| 483 |
+
"learning_rate": 4.2759192284508745e-05,
|
| 484 |
+
"loss": 0.0361,
|
| 485 |
+
"step": 1700
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.7017900732302685,
|
| 489 |
+
"grad_norm": 0.20996783673763275,
|
| 490 |
+
"learning_rate": 4.25708257986739e-05,
|
| 491 |
+
"loss": 0.0372,
|
| 492 |
+
"step": 1725
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 0.7119609438567941,
|
| 496 |
+
"grad_norm": 0.5258492827415466,
|
| 497 |
+
"learning_rate": 4.238245931283906e-05,
|
| 498 |
+
"loss": 0.0377,
|
| 499 |
+
"step": 1750
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.7221318144833198,
|
| 503 |
+
"grad_norm": 0.2202778309583664,
|
| 504 |
+
"learning_rate": 4.2194092827004224e-05,
|
| 505 |
+
"loss": 0.0384,
|
| 506 |
+
"step": 1775
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.7323026851098454,
|
| 510 |
+
"grad_norm": 0.6594241857528687,
|
| 511 |
+
"learning_rate": 4.200572634116938e-05,
|
| 512 |
+
"loss": 0.0329,
|
| 513 |
+
"step": 1800
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.7424735557363711,
|
| 517 |
+
"grad_norm": 0.19332900643348694,
|
| 518 |
+
"learning_rate": 4.181735985533454e-05,
|
| 519 |
+
"loss": 0.0386,
|
| 520 |
+
"step": 1825
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.7526444263628966,
|
| 524 |
+
"grad_norm": 0.22850601375102997,
|
| 525 |
+
"learning_rate": 4.1628993369499704e-05,
|
| 526 |
+
"loss": 0.0367,
|
| 527 |
+
"step": 1850
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.7628152969894223,
|
| 531 |
+
"grad_norm": 0.5552480220794678,
|
| 532 |
+
"learning_rate": 4.144062688366486e-05,
|
| 533 |
+
"loss": 0.0355,
|
| 534 |
+
"step": 1875
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.772986167615948,
|
| 538 |
+
"grad_norm": 0.5165444612503052,
|
| 539 |
+
"learning_rate": 4.1252260397830025e-05,
|
| 540 |
+
"loss": 0.0373,
|
| 541 |
+
"step": 1900
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.7831570382424735,
|
| 545 |
+
"grad_norm": 0.1546785682439804,
|
| 546 |
+
"learning_rate": 4.1063893911995176e-05,
|
| 547 |
+
"loss": 0.0328,
|
| 548 |
+
"step": 1925
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.7933279088689992,
|
| 552 |
+
"grad_norm": 0.45292505621910095,
|
| 553 |
+
"learning_rate": 4.087552742616034e-05,
|
| 554 |
+
"loss": 0.0358,
|
| 555 |
+
"step": 1950
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.8034987794955248,
|
| 559 |
+
"grad_norm": 0.31234049797058105,
|
| 560 |
+
"learning_rate": 4.06871609403255e-05,
|
| 561 |
+
"loss": 0.0375,
|
| 562 |
+
"step": 1975
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 0.8136696501220505,
|
| 566 |
+
"grad_norm": 0.26489096879959106,
|
| 567 |
+
"learning_rate": 4.0498794454490655e-05,
|
| 568 |
+
"loss": 0.0391,
|
| 569 |
+
"step": 2000
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.823840520748576,
|
| 573 |
+
"grad_norm": 0.30596041679382324,
|
| 574 |
+
"learning_rate": 4.031042796865582e-05,
|
| 575 |
+
"loss": 0.0356,
|
| 576 |
+
"step": 2025
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 0.8340113913751017,
|
| 580 |
+
"grad_norm": 0.22221428155899048,
|
| 581 |
+
"learning_rate": 4.012206148282098e-05,
|
| 582 |
+
"loss": 0.0349,
|
| 583 |
+
"step": 2050
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 0.8441822620016274,
|
| 587 |
+
"grad_norm": 0.27404096722602844,
|
| 588 |
+
"learning_rate": 3.9933694996986135e-05,
|
| 589 |
+
"loss": 0.0391,
|
| 590 |
+
"step": 2075
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.8543531326281529,
|
| 594 |
+
"grad_norm": 0.21221423149108887,
|
| 595 |
+
"learning_rate": 3.97453285111513e-05,
|
| 596 |
+
"loss": 0.0364,
|
| 597 |
+
"step": 2100
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 0.8645240032546786,
|
| 601 |
+
"grad_norm": 0.5386448502540588,
|
| 602 |
+
"learning_rate": 3.9556962025316456e-05,
|
| 603 |
+
"loss": 0.0369,
|
| 604 |
+
"step": 2125
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 0.8746948738812043,
|
| 608 |
+
"grad_norm": 0.2866438627243042,
|
| 609 |
+
"learning_rate": 3.936859553948162e-05,
|
| 610 |
+
"loss": 0.0363,
|
| 611 |
+
"step": 2150
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 0.8848657445077298,
|
| 615 |
+
"grad_norm": 0.300436407327652,
|
| 616 |
+
"learning_rate": 3.918022905364678e-05,
|
| 617 |
+
"loss": 0.0353,
|
| 618 |
+
"step": 2175
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 0.8950366151342555,
|
| 622 |
+
"grad_norm": 0.1889338344335556,
|
| 623 |
+
"learning_rate": 3.8991862567811936e-05,
|
| 624 |
+
"loss": 0.0362,
|
| 625 |
+
"step": 2200
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.9052074857607811,
|
| 629 |
+
"grad_norm": 0.47652769088745117,
|
| 630 |
+
"learning_rate": 3.88034960819771e-05,
|
| 631 |
+
"loss": 0.0394,
|
| 632 |
+
"step": 2225
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.9153783563873068,
|
| 636 |
+
"grad_norm": 0.43233829736709595,
|
| 637 |
+
"learning_rate": 3.861512959614226e-05,
|
| 638 |
+
"loss": 0.0377,
|
| 639 |
+
"step": 2250
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.9255492270138324,
|
| 643 |
+
"grad_norm": 0.3206855058670044,
|
| 644 |
+
"learning_rate": 3.8426763110307415e-05,
|
| 645 |
+
"loss": 0.0341,
|
| 646 |
+
"step": 2275
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 0.935720097640358,
|
| 650 |
+
"grad_norm": 0.2778429090976715,
|
| 651 |
+
"learning_rate": 3.823839662447258e-05,
|
| 652 |
+
"loss": 0.035,
|
| 653 |
+
"step": 2300
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 0.9458909682668837,
|
| 657 |
+
"grad_norm": 0.22730959951877594,
|
| 658 |
+
"learning_rate": 3.805003013863774e-05,
|
| 659 |
+
"loss": 0.0346,
|
| 660 |
+
"step": 2325
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 0.9560618388934092,
|
| 664 |
+
"grad_norm": 0.29823628067970276,
|
| 665 |
+
"learning_rate": 3.78616636528029e-05,
|
| 666 |
+
"loss": 0.0363,
|
| 667 |
+
"step": 2350
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 0.9662327095199349,
|
| 671 |
+
"grad_norm": 0.5740765333175659,
|
| 672 |
+
"learning_rate": 3.767329716696805e-05,
|
| 673 |
+
"loss": 0.0348,
|
| 674 |
+
"step": 2375
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 0.9764035801464606,
|
| 678 |
+
"grad_norm": 0.4176069498062134,
|
| 679 |
+
"learning_rate": 3.748493068113321e-05,
|
| 680 |
+
"loss": 0.0387,
|
| 681 |
+
"step": 2400
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 0.9865744507729862,
|
| 685 |
+
"grad_norm": 0.14277301728725433,
|
| 686 |
+
"learning_rate": 3.7296564195298374e-05,
|
| 687 |
+
"loss": 0.0362,
|
| 688 |
+
"step": 2425
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 0.9967453213995118,
|
| 692 |
+
"grad_norm": 0.20792150497436523,
|
| 693 |
+
"learning_rate": 3.710819770946353e-05,
|
| 694 |
+
"loss": 0.0351,
|
| 695 |
+
"step": 2450
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 1.0,
|
| 699 |
+
"eval_explained_variance": 0.24840307235717773,
|
| 700 |
+
"eval_loss": 0.03510129451751709,
|
| 701 |
+
"eval_mae": 0.1494196206331253,
|
| 702 |
+
"eval_mse": 0.035100266337394714,
|
| 703 |
+
"eval_r2": 0.2482784390449524,
|
| 704 |
+
"eval_rmse": 0.18735065075252533,
|
| 705 |
+
"eval_runtime": 4.0725,
|
| 706 |
+
"eval_samples_per_second": 4826.955,
|
| 707 |
+
"eval_steps_per_second": 75.628,
|
| 708 |
+
"step": 2458
|
| 709 |
+
},
|
| 710 |
+
{
|
| 711 |
+
"epoch": 1.0069161920260374,
|
| 712 |
+
"grad_norm": 0.4044972360134125,
|
| 713 |
+
"learning_rate": 3.6919831223628695e-05,
|
| 714 |
+
"loss": 0.0344,
|
| 715 |
+
"step": 2475
|
| 716 |
+
},
|
| 717 |
+
{
|
| 718 |
+
"epoch": 1.017087062652563,
|
| 719 |
+
"grad_norm": 0.1920347958803177,
|
| 720 |
+
"learning_rate": 3.673146473779385e-05,
|
| 721 |
+
"loss": 0.0333,
|
| 722 |
+
"step": 2500
|
| 723 |
+
},
|
| 724 |
+
{
|
| 725 |
+
"epoch": 1.0272579332790888,
|
| 726 |
+
"grad_norm": 0.2734430432319641,
|
| 727 |
+
"learning_rate": 3.654309825195901e-05,
|
| 728 |
+
"loss": 0.0329,
|
| 729 |
+
"step": 2525
|
| 730 |
+
},
|
| 731 |
+
{
|
| 732 |
+
"epoch": 1.0374288039056143,
|
| 733 |
+
"grad_norm": 0.2458937019109726,
|
| 734 |
+
"learning_rate": 3.6354731766124175e-05,
|
| 735 |
+
"loss": 0.0341,
|
| 736 |
+
"step": 2550
|
| 737 |
+
},
|
| 738 |
+
{
|
| 739 |
+
"epoch": 1.0475996745321399,
|
| 740 |
+
"grad_norm": 0.302209734916687,
|
| 741 |
+
"learning_rate": 3.616636528028933e-05,
|
| 742 |
+
"loss": 0.0336,
|
| 743 |
+
"step": 2575
|
| 744 |
+
},
|
| 745 |
+
{
|
| 746 |
+
"epoch": 1.0577705451586656,
|
| 747 |
+
"grad_norm": 0.23956753313541412,
|
| 748 |
+
"learning_rate": 3.597799879445449e-05,
|
| 749 |
+
"loss": 0.0351,
|
| 750 |
+
"step": 2600
|
| 751 |
+
},
|
| 752 |
+
{
|
| 753 |
+
"epoch": 1.0679414157851912,
|
| 754 |
+
"grad_norm": 0.4500243663787842,
|
| 755 |
+
"learning_rate": 3.5789632308619654e-05,
|
| 756 |
+
"loss": 0.031,
|
| 757 |
+
"step": 2625
|
| 758 |
+
},
|
| 759 |
+
{
|
| 760 |
+
"epoch": 1.0781122864117167,
|
| 761 |
+
"grad_norm": 0.23080125451087952,
|
| 762 |
+
"learning_rate": 3.560126582278481e-05,
|
| 763 |
+
"loss": 0.0358,
|
| 764 |
+
"step": 2650
|
| 765 |
+
},
|
| 766 |
+
{
|
| 767 |
+
"epoch": 1.0882831570382425,
|
| 768 |
+
"grad_norm": 0.44963157176971436,
|
| 769 |
+
"learning_rate": 3.5412899336949976e-05,
|
| 770 |
+
"loss": 0.0326,
|
| 771 |
+
"step": 2675
|
| 772 |
+
},
|
| 773 |
+
{
|
| 774 |
+
"epoch": 1.098454027664768,
|
| 775 |
+
"grad_norm": 0.721524715423584,
|
| 776 |
+
"learning_rate": 3.522453285111513e-05,
|
| 777 |
+
"loss": 0.0383,
|
| 778 |
+
"step": 2700
|
| 779 |
+
},
|
| 780 |
+
{
|
| 781 |
+
"epoch": 1.1086248982912936,
|
| 782 |
+
"grad_norm": 0.24731744825839996,
|
| 783 |
+
"learning_rate": 3.503616636528029e-05,
|
| 784 |
+
"loss": 0.0354,
|
| 785 |
+
"step": 2725
|
| 786 |
+
},
|
| 787 |
+
{
|
| 788 |
+
"epoch": 1.1187957689178194,
|
| 789 |
+
"grad_norm": 0.19786667823791504,
|
| 790 |
+
"learning_rate": 3.4847799879445455e-05,
|
| 791 |
+
"loss": 0.0335,
|
| 792 |
+
"step": 2750
|
| 793 |
+
},
|
| 794 |
+
{
|
| 795 |
+
"epoch": 1.128966639544345,
|
| 796 |
+
"grad_norm": 0.16557921469211578,
|
| 797 |
+
"learning_rate": 3.465943339361061e-05,
|
| 798 |
+
"loss": 0.0349,
|
| 799 |
+
"step": 2775
|
| 800 |
+
},
|
| 801 |
+
{
|
| 802 |
+
"epoch": 1.1391375101708707,
|
| 803 |
+
"grad_norm": 0.30080434679985046,
|
| 804 |
+
"learning_rate": 3.447106690777577e-05,
|
| 805 |
+
"loss": 0.0335,
|
| 806 |
+
"step": 2800
|
| 807 |
+
},
|
| 808 |
+
{
|
| 809 |
+
"epoch": 1.1493083807973963,
|
| 810 |
+
"grad_norm": 0.2929253876209259,
|
| 811 |
+
"learning_rate": 3.428270042194093e-05,
|
| 812 |
+
"loss": 0.0338,
|
| 813 |
+
"step": 2825
|
| 814 |
+
},
|
| 815 |
+
{
|
| 816 |
+
"epoch": 1.1594792514239218,
|
| 817 |
+
"grad_norm": 0.19096094369888306,
|
| 818 |
+
"learning_rate": 3.4094333936106085e-05,
|
| 819 |
+
"loss": 0.0307,
|
| 820 |
+
"step": 2850
|
| 821 |
+
},
|
| 822 |
+
{
|
| 823 |
+
"epoch": 1.1696501220504476,
|
| 824 |
+
"grad_norm": 0.2764925956726074,
|
| 825 |
+
"learning_rate": 3.390596745027125e-05,
|
| 826 |
+
"loss": 0.034,
|
| 827 |
+
"step": 2875
|
| 828 |
+
},
|
| 829 |
+
{
|
| 830 |
+
"epoch": 1.1798209926769732,
|
| 831 |
+
"grad_norm": 0.36045756936073303,
|
| 832 |
+
"learning_rate": 3.371760096443641e-05,
|
| 833 |
+
"loss": 0.0327,
|
| 834 |
+
"step": 2900
|
| 835 |
+
},
|
| 836 |
+
{
|
| 837 |
+
"epoch": 1.1899918633034987,
|
| 838 |
+
"grad_norm": 0.4153495728969574,
|
| 839 |
+
"learning_rate": 3.352923447860157e-05,
|
| 840 |
+
"loss": 0.0341,
|
| 841 |
+
"step": 2925
|
| 842 |
+
},
|
| 843 |
+
{
|
| 844 |
+
"epoch": 1.2001627339300245,
|
| 845 |
+
"grad_norm": 0.3471417725086212,
|
| 846 |
+
"learning_rate": 3.334086799276673e-05,
|
| 847 |
+
"loss": 0.0329,
|
| 848 |
+
"step": 2950
|
| 849 |
+
},
|
| 850 |
+
{
|
| 851 |
+
"epoch": 1.21033360455655,
|
| 852 |
+
"grad_norm": 0.3038371205329895,
|
| 853 |
+
"learning_rate": 3.3152501506931886e-05,
|
| 854 |
+
"loss": 0.0344,
|
| 855 |
+
"step": 2975
|
| 856 |
+
},
|
| 857 |
+
{
|
| 858 |
+
"epoch": 1.2205044751830756,
|
| 859 |
+
"grad_norm": 0.589869499206543,
|
| 860 |
+
"learning_rate": 3.296413502109705e-05,
|
| 861 |
+
"loss": 0.0369,
|
| 862 |
+
"step": 3000
|
| 863 |
+
},
|
| 864 |
+
{
|
| 865 |
+
"epoch": 1.2306753458096014,
|
| 866 |
+
"grad_norm": 0.19090452790260315,
|
| 867 |
+
"learning_rate": 3.277576853526221e-05,
|
| 868 |
+
"loss": 0.0325,
|
| 869 |
+
"step": 3025
|
| 870 |
+
},
|
| 871 |
+
{
|
| 872 |
+
"epoch": 1.240846216436127,
|
| 873 |
+
"grad_norm": 0.2946130335330963,
|
| 874 |
+
"learning_rate": 3.2587402049427365e-05,
|
| 875 |
+
"loss": 0.033,
|
| 876 |
+
"step": 3050
|
| 877 |
+
},
|
| 878 |
+
{
|
| 879 |
+
"epoch": 1.2510170870626527,
|
| 880 |
+
"grad_norm": 0.33456677198410034,
|
| 881 |
+
"learning_rate": 3.239903556359253e-05,
|
| 882 |
+
"loss": 0.0311,
|
| 883 |
+
"step": 3075
|
| 884 |
+
},
|
| 885 |
+
{
|
| 886 |
+
"epoch": 1.2611879576891782,
|
| 887 |
+
"grad_norm": 0.23250174522399902,
|
| 888 |
+
"learning_rate": 3.221066907775769e-05,
|
| 889 |
+
"loss": 0.034,
|
| 890 |
+
"step": 3100
|
| 891 |
+
},
|
| 892 |
+
{
|
| 893 |
+
"epoch": 1.2713588283157038,
|
| 894 |
+
"grad_norm": 0.2728452682495117,
|
| 895 |
+
"learning_rate": 3.2022302591922845e-05,
|
| 896 |
+
"loss": 0.0328,
|
| 897 |
+
"step": 3125
|
| 898 |
+
},
|
| 899 |
+
{
|
| 900 |
+
"epoch": 1.2815296989422293,
|
| 901 |
+
"grad_norm": 0.22802890837192535,
|
| 902 |
+
"learning_rate": 3.183393610608801e-05,
|
| 903 |
+
"loss": 0.0291,
|
| 904 |
+
"step": 3150
|
| 905 |
+
},
|
| 906 |
+
{
|
| 907 |
+
"epoch": 1.2917005695687551,
|
| 908 |
+
"grad_norm": 0.3694087862968445,
|
| 909 |
+
"learning_rate": 3.1645569620253167e-05,
|
| 910 |
+
"loss": 0.033,
|
| 911 |
+
"step": 3175
|
| 912 |
+
},
|
| 913 |
+
{
|
| 914 |
+
"epoch": 1.3018714401952807,
|
| 915 |
+
"grad_norm": 0.3649369478225708,
|
| 916 |
+
"learning_rate": 3.145720313441833e-05,
|
| 917 |
+
"loss": 0.0353,
|
| 918 |
+
"step": 3200
|
| 919 |
+
},
|
| 920 |
+
{
|
| 921 |
+
"epoch": 1.3120423108218064,
|
| 922 |
+
"grad_norm": 0.45792657136917114,
|
| 923 |
+
"learning_rate": 3.126883664858349e-05,
|
| 924 |
+
"loss": 0.0332,
|
| 925 |
+
"step": 3225
|
| 926 |
+
},
|
| 927 |
+
{
|
| 928 |
+
"epoch": 1.322213181448332,
|
| 929 |
+
"grad_norm": 0.325790137052536,
|
| 930 |
+
"learning_rate": 3.1080470162748646e-05,
|
| 931 |
+
"loss": 0.0314,
|
| 932 |
+
"step": 3250
|
| 933 |
+
},
|
| 934 |
+
{
|
| 935 |
+
"epoch": 1.3323840520748575,
|
| 936 |
+
"grad_norm": 0.25394582748413086,
|
| 937 |
+
"learning_rate": 3.08921036769138e-05,
|
| 938 |
+
"loss": 0.0347,
|
| 939 |
+
"step": 3275
|
| 940 |
+
},
|
| 941 |
+
{
|
| 942 |
+
"epoch": 1.342554922701383,
|
| 943 |
+
"grad_norm": 0.25821924209594727,
|
| 944 |
+
"learning_rate": 3.070373719107896e-05,
|
| 945 |
+
"loss": 0.0326,
|
| 946 |
+
"step": 3300
|
| 947 |
+
},
|
| 948 |
+
{
|
| 949 |
+
"epoch": 1.3527257933279089,
|
| 950 |
+
"grad_norm": 0.24513724446296692,
|
| 951 |
+
"learning_rate": 3.0515370705244122e-05,
|
| 952 |
+
"loss": 0.0323,
|
| 953 |
+
"step": 3325
|
| 954 |
+
},
|
| 955 |
+
{
|
| 956 |
+
"epoch": 1.3628966639544344,
|
| 957 |
+
"grad_norm": 0.5198257565498352,
|
| 958 |
+
"learning_rate": 3.0327004219409283e-05,
|
| 959 |
+
"loss": 0.0337,
|
| 960 |
+
"step": 3350
|
| 961 |
+
},
|
| 962 |
+
{
|
| 963 |
+
"epoch": 1.3730675345809602,
|
| 964 |
+
"grad_norm": 0.4150083661079407,
|
| 965 |
+
"learning_rate": 3.0138637733574443e-05,
|
| 966 |
+
"loss": 0.0313,
|
| 967 |
+
"step": 3375
|
| 968 |
+
},
|
| 969 |
+
{
|
| 970 |
+
"epoch": 1.3832384052074858,
|
| 971 |
+
"grad_norm": 0.148391991853714,
|
| 972 |
+
"learning_rate": 2.9950271247739604e-05,
|
| 973 |
+
"loss": 0.0351,
|
| 974 |
+
"step": 3400
|
| 975 |
+
},
|
| 976 |
+
{
|
| 977 |
+
"epoch": 1.3934092758340113,
|
| 978 |
+
"grad_norm": 0.23965270817279816,
|
| 979 |
+
"learning_rate": 2.9761904761904762e-05,
|
| 980 |
+
"loss": 0.0338,
|
| 981 |
+
"step": 3425
|
| 982 |
+
},
|
| 983 |
+
{
|
| 984 |
+
"epoch": 1.403580146460537,
|
| 985 |
+
"grad_norm": 0.21719323098659515,
|
| 986 |
+
"learning_rate": 2.9573538276069923e-05,
|
| 987 |
+
"loss": 0.0305,
|
| 988 |
+
"step": 3450
|
| 989 |
+
},
|
| 990 |
+
{
|
| 991 |
+
"epoch": 1.4137510170870626,
|
| 992 |
+
"grad_norm": 0.25082919001579285,
|
| 993 |
+
"learning_rate": 2.9385171790235084e-05,
|
| 994 |
+
"loss": 0.0328,
|
| 995 |
+
"step": 3475
|
| 996 |
+
},
|
| 997 |
+
{
|
| 998 |
+
"epoch": 1.4239218877135884,
|
| 999 |
+
"grad_norm": 0.3688701093196869,
|
| 1000 |
+
"learning_rate": 2.9196805304400245e-05,
|
| 1001 |
+
"loss": 0.0352,
|
| 1002 |
+
"step": 3500
|
| 1003 |
+
},
|
| 1004 |
+
{
|
| 1005 |
+
"epoch": 1.434092758340114,
|
| 1006 |
+
"grad_norm": 0.3461095094680786,
|
| 1007 |
+
"learning_rate": 2.9008438818565402e-05,
|
| 1008 |
+
"loss": 0.0356,
|
| 1009 |
+
"step": 3525
|
| 1010 |
+
},
|
| 1011 |
+
{
|
| 1012 |
+
"epoch": 1.4442636289666395,
|
| 1013 |
+
"grad_norm": 0.2626365125179291,
|
| 1014 |
+
"learning_rate": 2.8820072332730563e-05,
|
| 1015 |
+
"loss": 0.033,
|
| 1016 |
+
"step": 3550
|
| 1017 |
+
},
|
| 1018 |
+
{
|
| 1019 |
+
"epoch": 1.454434499593165,
|
| 1020 |
+
"grad_norm": 0.27237704396247864,
|
| 1021 |
+
"learning_rate": 2.8631705846895724e-05,
|
| 1022 |
+
"loss": 0.0342,
|
| 1023 |
+
"step": 3575
|
| 1024 |
+
},
|
| 1025 |
+
{
|
| 1026 |
+
"epoch": 1.4646053702196908,
|
| 1027 |
+
"grad_norm": 0.258208304643631,
|
| 1028 |
+
"learning_rate": 2.8443339361060885e-05,
|
| 1029 |
+
"loss": 0.0327,
|
| 1030 |
+
"step": 3600
|
| 1031 |
+
},
|
| 1032 |
+
{
|
| 1033 |
+
"epoch": 1.4747762408462164,
|
| 1034 |
+
"grad_norm": 0.2694801390171051,
|
| 1035 |
+
"learning_rate": 2.8254972875226042e-05,
|
| 1036 |
+
"loss": 0.0353,
|
| 1037 |
+
"step": 3625
|
| 1038 |
+
},
|
| 1039 |
+
{
|
| 1040 |
+
"epoch": 1.4849471114727422,
|
| 1041 |
+
"grad_norm": 0.3417627513408661,
|
| 1042 |
+
"learning_rate": 2.8066606389391203e-05,
|
| 1043 |
+
"loss": 0.0325,
|
| 1044 |
+
"step": 3650
|
| 1045 |
+
},
|
| 1046 |
+
{
|
| 1047 |
+
"epoch": 1.4951179820992677,
|
| 1048 |
+
"grad_norm": 0.4780226945877075,
|
| 1049 |
+
"learning_rate": 2.7878239903556357e-05,
|
| 1050 |
+
"loss": 0.0338,
|
| 1051 |
+
"step": 3675
|
| 1052 |
+
},
|
| 1053 |
+
{
|
| 1054 |
+
"epoch": 1.5052888527257933,
|
| 1055 |
+
"grad_norm": 0.2635546922683716,
|
| 1056 |
+
"learning_rate": 2.7689873417721518e-05,
|
| 1057 |
+
"loss": 0.0321,
|
| 1058 |
+
"step": 3700
|
| 1059 |
+
},
|
| 1060 |
+
{
|
| 1061 |
+
"epoch": 1.5154597233523188,
|
| 1062 |
+
"grad_norm": 0.2416383922100067,
|
| 1063 |
+
"learning_rate": 2.750150693188668e-05,
|
| 1064 |
+
"loss": 0.0338,
|
| 1065 |
+
"step": 3725
|
| 1066 |
+
},
|
| 1067 |
+
{
|
| 1068 |
+
"epoch": 1.5256305939788446,
|
| 1069 |
+
"grad_norm": 0.5338820219039917,
|
| 1070 |
+
"learning_rate": 2.7313140446051837e-05,
|
| 1071 |
+
"loss": 0.0319,
|
| 1072 |
+
"step": 3750
|
| 1073 |
+
},
|
| 1074 |
+
{
|
| 1075 |
+
"epoch": 1.5358014646053704,
|
| 1076 |
+
"grad_norm": 0.35653921961784363,
|
| 1077 |
+
"learning_rate": 2.7124773960216997e-05,
|
| 1078 |
+
"loss": 0.0338,
|
| 1079 |
+
"step": 3775
|
| 1080 |
+
},
|
| 1081 |
+
{
|
| 1082 |
+
"epoch": 1.545972335231896,
|
| 1083 |
+
"grad_norm": 0.23320654034614563,
|
| 1084 |
+
"learning_rate": 2.693640747438216e-05,
|
| 1085 |
+
"loss": 0.0312,
|
| 1086 |
+
"step": 3800
|
| 1087 |
+
},
|
| 1088 |
+
{
|
| 1089 |
+
"epoch": 1.5561432058584215,
|
| 1090 |
+
"grad_norm": 0.31537100672721863,
|
| 1091 |
+
"learning_rate": 2.674804098854732e-05,
|
| 1092 |
+
"loss": 0.037,
|
| 1093 |
+
"step": 3825
|
| 1094 |
+
},
|
| 1095 |
+
{
|
| 1096 |
+
"epoch": 1.566314076484947,
|
| 1097 |
+
"grad_norm": 0.24033057689666748,
|
| 1098 |
+
"learning_rate": 2.6559674502712477e-05,
|
| 1099 |
+
"loss": 0.0315,
|
| 1100 |
+
"step": 3850
|
| 1101 |
+
},
|
| 1102 |
+
{
|
| 1103 |
+
"epoch": 1.5764849471114726,
|
| 1104 |
+
"grad_norm": 0.21600840985774994,
|
| 1105 |
+
"learning_rate": 2.6371308016877638e-05,
|
| 1106 |
+
"loss": 0.0329,
|
| 1107 |
+
"step": 3875
|
| 1108 |
+
},
|
| 1109 |
+
{
|
| 1110 |
+
"epoch": 1.5866558177379984,
|
| 1111 |
+
"grad_norm": 0.4238574206829071,
|
| 1112 |
+
"learning_rate": 2.61829415310428e-05,
|
| 1113 |
+
"loss": 0.0356,
|
| 1114 |
+
"step": 3900
|
| 1115 |
+
},
|
| 1116 |
+
{
|
| 1117 |
+
"epoch": 1.5968266883645241,
|
| 1118 |
+
"grad_norm": 0.3188216984272003,
|
| 1119 |
+
"learning_rate": 2.599457504520796e-05,
|
| 1120 |
+
"loss": 0.0321,
|
| 1121 |
+
"step": 3925
|
| 1122 |
+
},
|
| 1123 |
+
{
|
| 1124 |
+
"epoch": 1.6069975589910497,
|
| 1125 |
+
"grad_norm": 0.29613322019577026,
|
| 1126 |
+
"learning_rate": 2.5806208559373117e-05,
|
| 1127 |
+
"loss": 0.0288,
|
| 1128 |
+
"step": 3950
|
| 1129 |
+
},
|
| 1130 |
+
{
|
| 1131 |
+
"epoch": 1.6171684296175752,
|
| 1132 |
+
"grad_norm": 0.3398037850856781,
|
| 1133 |
+
"learning_rate": 2.5617842073538278e-05,
|
| 1134 |
+
"loss": 0.0356,
|
| 1135 |
+
"step": 3975
|
| 1136 |
+
},
|
| 1137 |
+
{
|
| 1138 |
+
"epoch": 1.6273393002441008,
|
| 1139 |
+
"grad_norm": 0.7744150757789612,
|
| 1140 |
+
"learning_rate": 2.542947558770344e-05,
|
| 1141 |
+
"loss": 0.03,
|
| 1142 |
+
"step": 4000
|
| 1143 |
+
},
|
| 1144 |
+
{
|
| 1145 |
+
"epoch": 1.6375101708706266,
|
| 1146 |
+
"grad_norm": 0.2858869135379791,
|
| 1147 |
+
"learning_rate": 2.52411091018686e-05,
|
| 1148 |
+
"loss": 0.0313,
|
| 1149 |
+
"step": 4025
|
| 1150 |
+
},
|
| 1151 |
+
{
|
| 1152 |
+
"epoch": 1.647681041497152,
|
| 1153 |
+
"grad_norm": 0.5115847587585449,
|
| 1154 |
+
"learning_rate": 2.5052742616033757e-05,
|
| 1155 |
+
"loss": 0.0321,
|
| 1156 |
+
"step": 4050
|
| 1157 |
+
},
|
| 1158 |
+
{
|
| 1159 |
+
"epoch": 1.6578519121236779,
|
| 1160 |
+
"grad_norm": 0.2705928385257721,
|
| 1161 |
+
"learning_rate": 2.4864376130198915e-05,
|
| 1162 |
+
"loss": 0.0299,
|
| 1163 |
+
"step": 4075
|
| 1164 |
+
},
|
| 1165 |
+
{
|
| 1166 |
+
"epoch": 1.6680227827502034,
|
| 1167 |
+
"grad_norm": 0.23888413608074188,
|
| 1168 |
+
"learning_rate": 2.4676009644364075e-05,
|
| 1169 |
+
"loss": 0.0302,
|
| 1170 |
+
"step": 4100
|
| 1171 |
+
},
|
| 1172 |
+
{
|
| 1173 |
+
"epoch": 1.678193653376729,
|
| 1174 |
+
"grad_norm": 0.251142680644989,
|
| 1175 |
+
"learning_rate": 2.4487643158529236e-05,
|
| 1176 |
+
"loss": 0.0352,
|
| 1177 |
+
"step": 4125
|
| 1178 |
+
},
|
| 1179 |
+
{
|
| 1180 |
+
"epoch": 1.6883645240032545,
|
| 1181 |
+
"grad_norm": 0.4335423707962036,
|
| 1182 |
+
"learning_rate": 2.4299276672694397e-05,
|
| 1183 |
+
"loss": 0.0342,
|
| 1184 |
+
"step": 4150
|
| 1185 |
+
},
|
| 1186 |
+
{
|
| 1187 |
+
"epoch": 1.6985353946297803,
|
| 1188 |
+
"grad_norm": 0.26633137464523315,
|
| 1189 |
+
"learning_rate": 2.4110910186859555e-05,
|
| 1190 |
+
"loss": 0.0303,
|
| 1191 |
+
"step": 4175
|
| 1192 |
+
},
|
| 1193 |
+
{
|
| 1194 |
+
"epoch": 1.708706265256306,
|
| 1195 |
+
"grad_norm": 0.18798017501831055,
|
| 1196 |
+
"learning_rate": 2.3922543701024716e-05,
|
| 1197 |
+
"loss": 0.0318,
|
| 1198 |
+
"step": 4200
|
| 1199 |
+
},
|
| 1200 |
+
{
|
| 1201 |
+
"epoch": 1.7188771358828316,
|
| 1202 |
+
"grad_norm": 0.14591360092163086,
|
| 1203 |
+
"learning_rate": 2.3734177215189873e-05,
|
| 1204 |
+
"loss": 0.0306,
|
| 1205 |
+
"step": 4225
|
| 1206 |
+
},
|
| 1207 |
+
{
|
| 1208 |
+
"epoch": 1.7290480065093572,
|
| 1209 |
+
"grad_norm": 0.22364169359207153,
|
| 1210 |
+
"learning_rate": 2.3545810729355034e-05,
|
| 1211 |
+
"loss": 0.031,
|
| 1212 |
+
"step": 4250
|
| 1213 |
+
},
|
| 1214 |
+
{
|
| 1215 |
+
"epoch": 1.7392188771358827,
|
| 1216 |
+
"grad_norm": 0.28795015811920166,
|
| 1217 |
+
"learning_rate": 2.3357444243520195e-05,
|
| 1218 |
+
"loss": 0.0337,
|
| 1219 |
+
"step": 4275
|
| 1220 |
+
},
|
| 1221 |
+
{
|
| 1222 |
+
"epoch": 1.7493897477624083,
|
| 1223 |
+
"grad_norm": 0.5833514332771301,
|
| 1224 |
+
"learning_rate": 2.3169077757685352e-05,
|
| 1225 |
+
"loss": 0.0356,
|
| 1226 |
+
"step": 4300
|
| 1227 |
+
},
|
| 1228 |
+
{
|
| 1229 |
+
"epoch": 1.759560618388934,
|
| 1230 |
+
"grad_norm": 0.20125386118888855,
|
| 1231 |
+
"learning_rate": 2.2980711271850513e-05,
|
| 1232 |
+
"loss": 0.035,
|
| 1233 |
+
"step": 4325
|
| 1234 |
+
},
|
| 1235 |
+
{
|
| 1236 |
+
"epoch": 1.7697314890154598,
|
| 1237 |
+
"grad_norm": 0.39359915256500244,
|
| 1238 |
+
"learning_rate": 2.2792344786015674e-05,
|
| 1239 |
+
"loss": 0.0325,
|
| 1240 |
+
"step": 4350
|
| 1241 |
+
},
|
| 1242 |
+
{
|
| 1243 |
+
"epoch": 1.7799023596419854,
|
| 1244 |
+
"grad_norm": 0.18621224164962769,
|
| 1245 |
+
"learning_rate": 2.2603978300180835e-05,
|
| 1246 |
+
"loss": 0.0287,
|
| 1247 |
+
"step": 4375
|
| 1248 |
+
},
|
| 1249 |
+
{
|
| 1250 |
+
"epoch": 1.790073230268511,
|
| 1251 |
+
"grad_norm": 0.23905445635318756,
|
| 1252 |
+
"learning_rate": 2.2415611814345993e-05,
|
| 1253 |
+
"loss": 0.0338,
|
| 1254 |
+
"step": 4400
|
| 1255 |
+
},
|
| 1256 |
+
{
|
| 1257 |
+
"epoch": 1.8002441008950365,
|
| 1258 |
+
"grad_norm": 0.5699689984321594,
|
| 1259 |
+
"learning_rate": 2.2227245328511154e-05,
|
| 1260 |
+
"loss": 0.034,
|
| 1261 |
+
"step": 4425
|
| 1262 |
+
},
|
| 1263 |
+
{
|
| 1264 |
+
"epoch": 1.8104149715215623,
|
| 1265 |
+
"grad_norm": 0.3101656138896942,
|
| 1266 |
+
"learning_rate": 2.203887884267631e-05,
|
| 1267 |
+
"loss": 0.0344,
|
| 1268 |
+
"step": 4450
|
| 1269 |
+
},
|
| 1270 |
+
{
|
| 1271 |
+
"epoch": 1.8205858421480878,
|
| 1272 |
+
"grad_norm": 0.19921617209911346,
|
| 1273 |
+
"learning_rate": 2.1850512356841472e-05,
|
| 1274 |
+
"loss": 0.0348,
|
| 1275 |
+
"step": 4475
|
| 1276 |
+
},
|
| 1277 |
+
{
|
| 1278 |
+
"epoch": 1.8307567127746136,
|
| 1279 |
+
"grad_norm": 0.4034786522388458,
|
| 1280 |
+
"learning_rate": 2.166214587100663e-05,
|
| 1281 |
+
"loss": 0.0332,
|
| 1282 |
+
"step": 4500
|
| 1283 |
+
},
|
| 1284 |
+
{
|
| 1285 |
+
"epoch": 1.8409275834011392,
|
| 1286 |
+
"grad_norm": 0.5094945430755615,
|
| 1287 |
+
"learning_rate": 2.147377938517179e-05,
|
| 1288 |
+
"loss": 0.0293,
|
| 1289 |
+
"step": 4525
|
| 1290 |
+
},
|
| 1291 |
+
{
|
| 1292 |
+
"epoch": 1.8510984540276647,
|
| 1293 |
+
"grad_norm": 0.4296955168247223,
|
| 1294 |
+
"learning_rate": 2.128541289933695e-05,
|
| 1295 |
+
"loss": 0.0318,
|
| 1296 |
+
"step": 4550
|
| 1297 |
+
},
|
| 1298 |
+
{
|
| 1299 |
+
"epoch": 1.8612693246541903,
|
| 1300 |
+
"grad_norm": 0.2767828404903412,
|
| 1301 |
+
"learning_rate": 2.1097046413502112e-05,
|
| 1302 |
+
"loss": 0.0337,
|
| 1303 |
+
"step": 4575
|
| 1304 |
+
},
|
| 1305 |
+
{
|
| 1306 |
+
"epoch": 1.871440195280716,
|
| 1307 |
+
"grad_norm": 0.2815410792827606,
|
| 1308 |
+
"learning_rate": 2.090867992766727e-05,
|
| 1309 |
+
"loss": 0.0322,
|
| 1310 |
+
"step": 4600
|
| 1311 |
+
},
|
| 1312 |
+
{
|
| 1313 |
+
"epoch": 1.8816110659072418,
|
| 1314 |
+
"grad_norm": 0.2712397873401642,
|
| 1315 |
+
"learning_rate": 2.072031344183243e-05,
|
| 1316 |
+
"loss": 0.0309,
|
| 1317 |
+
"step": 4625
|
| 1318 |
+
},
|
| 1319 |
+
{
|
| 1320 |
+
"epoch": 1.8917819365337674,
|
| 1321 |
+
"grad_norm": 0.48615217208862305,
|
| 1322 |
+
"learning_rate": 2.0531946955997588e-05,
|
| 1323 |
+
"loss": 0.0335,
|
| 1324 |
+
"step": 4650
|
| 1325 |
+
},
|
| 1326 |
+
{
|
| 1327 |
+
"epoch": 1.901952807160293,
|
| 1328 |
+
"grad_norm": 0.20523346960544586,
|
| 1329 |
+
"learning_rate": 2.034358047016275e-05,
|
| 1330 |
+
"loss": 0.032,
|
| 1331 |
+
"step": 4675
|
| 1332 |
+
},
|
| 1333 |
+
{
|
| 1334 |
+
"epoch": 1.9121236777868185,
|
| 1335 |
+
"grad_norm": 0.21368514001369476,
|
| 1336 |
+
"learning_rate": 2.015521398432791e-05,
|
| 1337 |
+
"loss": 0.0323,
|
| 1338 |
+
"step": 4700
|
| 1339 |
+
},
|
| 1340 |
+
{
|
| 1341 |
+
"epoch": 1.922294548413344,
|
| 1342 |
+
"grad_norm": 0.5739328265190125,
|
| 1343 |
+
"learning_rate": 1.9966847498493067e-05,
|
| 1344 |
+
"loss": 0.0324,
|
| 1345 |
+
"step": 4725
|
| 1346 |
+
},
|
| 1347 |
+
{
|
| 1348 |
+
"epoch": 1.9324654190398698,
|
| 1349 |
+
"grad_norm": 0.34018173813819885,
|
| 1350 |
+
"learning_rate": 1.9778481012658228e-05,
|
| 1351 |
+
"loss": 0.0333,
|
| 1352 |
+
"step": 4750
|
| 1353 |
+
},
|
| 1354 |
+
{
|
| 1355 |
+
"epoch": 1.9426362896663956,
|
| 1356 |
+
"grad_norm": 0.29020246863365173,
|
| 1357 |
+
"learning_rate": 1.959011452682339e-05,
|
| 1358 |
+
"loss": 0.032,
|
| 1359 |
+
"step": 4775
|
| 1360 |
+
},
|
| 1361 |
+
{
|
| 1362 |
+
"epoch": 1.9528071602929211,
|
| 1363 |
+
"grad_norm": 0.2339148074388504,
|
| 1364 |
+
"learning_rate": 1.940174804098855e-05,
|
| 1365 |
+
"loss": 0.0307,
|
| 1366 |
+
"step": 4800
|
| 1367 |
+
},
|
| 1368 |
+
{
|
| 1369 |
+
"epoch": 1.9629780309194467,
|
| 1370 |
+
"grad_norm": 0.23286353051662445,
|
| 1371 |
+
"learning_rate": 1.9213381555153708e-05,
|
| 1372 |
+
"loss": 0.0339,
|
| 1373 |
+
"step": 4825
|
| 1374 |
+
},
|
| 1375 |
+
{
|
| 1376 |
+
"epoch": 1.9731489015459722,
|
| 1377 |
+
"grad_norm": 0.2174939662218094,
|
| 1378 |
+
"learning_rate": 1.902501506931887e-05,
|
| 1379 |
+
"loss": 0.0318,
|
| 1380 |
+
"step": 4850
|
| 1381 |
+
},
|
| 1382 |
+
{
|
| 1383 |
+
"epoch": 1.983319772172498,
|
| 1384 |
+
"grad_norm": 0.2397424727678299,
|
| 1385 |
+
"learning_rate": 1.8836648583484026e-05,
|
| 1386 |
+
"loss": 0.032,
|
| 1387 |
+
"step": 4875
|
| 1388 |
+
},
|
| 1389 |
+
{
|
| 1390 |
+
"epoch": 1.9934906427990235,
|
| 1391 |
+
"grad_norm": 0.3197477161884308,
|
| 1392 |
+
"learning_rate": 1.8648282097649187e-05,
|
| 1393 |
+
"loss": 0.0319,
|
| 1394 |
+
"step": 4900
|
| 1395 |
+
},
|
| 1396 |
+
{
|
| 1397 |
+
"epoch": 2.0,
|
| 1398 |
+
"eval_explained_variance": 0.2763633728027344,
|
| 1399 |
+
"eval_loss": 0.03380444645881653,
|
| 1400 |
+
"eval_mae": 0.14292894303798676,
|
| 1401 |
+
"eval_mse": 0.033803146332502365,
|
| 1402 |
+
"eval_r2": 0.2760580778121948,
|
| 1403 |
+
"eval_rmse": 0.18385631980571776,
|
| 1404 |
+
"eval_runtime": 4.0998,
|
| 1405 |
+
"eval_samples_per_second": 4794.858,
|
| 1406 |
+
"eval_steps_per_second": 75.125,
|
| 1407 |
+
"step": 4916
|
| 1408 |
+
}
|
| 1409 |
+
],
|
| 1410 |
+
"logging_steps": 25,
|
| 1411 |
+
"max_steps": 7374,
|
| 1412 |
+
"num_input_tokens_seen": 0,
|
| 1413 |
+
"num_train_epochs": 3,
|
| 1414 |
+
"save_steps": 500,
|
| 1415 |
+
"stateful_callbacks": {
|
| 1416 |
+
"EarlyStoppingCallback": {
|
| 1417 |
+
"args": {
|
| 1418 |
+
"early_stopping_patience": 5,
|
| 1419 |
+
"early_stopping_threshold": 0.01
|
| 1420 |
+
},
|
| 1421 |
+
"attributes": {
|
| 1422 |
+
"early_stopping_patience_counter": 1
|
| 1423 |
+
}
|
| 1424 |
+
},
|
| 1425 |
+
"TrainerControl": {
|
| 1426 |
+
"args": {
|
| 1427 |
+
"should_epoch_stop": false,
|
| 1428 |
+
"should_evaluate": false,
|
| 1429 |
+
"should_log": false,
|
| 1430 |
+
"should_save": true,
|
| 1431 |
+
"should_training_stop": false
|
| 1432 |
+
},
|
| 1433 |
+
"attributes": {}
|
| 1434 |
+
}
|
| 1435 |
+
},
|
| 1436 |
+
"total_flos": 5209584826318848.0,
|
| 1437 |
+
"train_batch_size": 8,
|
| 1438 |
+
"trial_name": null,
|
| 1439 |
+
"trial_params": null
|
| 1440 |
+
}
|
checkpoint-4916/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:51a4264247fbe1fe88d05c68d61b8084763d2f8eb4fde109f383a88bcaed7463
|
| 3 |
+
size 5368
|
config.json
ADDED
|
@@ -0,0 +1,32 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "distilbert/distilbert-base-uncased",
|
| 3 |
+
"_num_labels": 1,
|
| 4 |
+
"activation": "gelu",
|
| 5 |
+
"architectures": [
|
| 6 |
+
"DistilBertForSequenceClassification"
|
| 7 |
+
],
|
| 8 |
+
"attention_dropout": 0.1,
|
| 9 |
+
"dim": 768,
|
| 10 |
+
"dropout": 0.1,
|
| 11 |
+
"hidden_dim": 3072,
|
| 12 |
+
"id2label": {
|
| 13 |
+
"0": "target"
|
| 14 |
+
},
|
| 15 |
+
"initializer_range": 0.02,
|
| 16 |
+
"label2id": {
|
| 17 |
+
"target": 0
|
| 18 |
+
},
|
| 19 |
+
"max_position_embeddings": 512,
|
| 20 |
+
"model_type": "distilbert",
|
| 21 |
+
"n_heads": 12,
|
| 22 |
+
"n_layers": 6,
|
| 23 |
+
"pad_token_id": 0,
|
| 24 |
+
"problem_type": "regression",
|
| 25 |
+
"qa_dropout": 0.1,
|
| 26 |
+
"seq_classif_dropout": 0.2,
|
| 27 |
+
"sinusoidal_pos_embds": false,
|
| 28 |
+
"tie_weights_": true,
|
| 29 |
+
"torch_dtype": "float32",
|
| 30 |
+
"transformers_version": "4.48.0",
|
| 31 |
+
"vocab_size": 30522
|
| 32 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:52992a6b1bb7b233e4054bc086b088013d7e7613bcffead944d24fd70d64c55e
|
| 3 |
+
size 267829484
|
runs/Feb22_02-13-45_r-samtuckervegan-autotrain-4xl4-41bfog4i-7c075-ypwih/events.out.tfevents.1740190427.r-samtuckervegan-autotrain-4xl4-41bfog4i-7c075-ypwih.208.0
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5382562246cf343789c90591b47487305c08fddb4bb31ad9423e1e4afb19f09d
|
| 3 |
+
size 68990
|
runs/Feb22_02-13-45_r-samtuckervegan-autotrain-4xl4-41bfog4i-7c075-ypwih/events.out.tfevents.1740191139.r-samtuckervegan-autotrain-4xl4-41bfog4i-7c075-ypwih.208.1
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9c56ff9864bfc4fb732368db29e9cb5af3bfdc993c409ac957d8db486cf30a2a
|
| 3 |
+
size 609
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"cls_token": "[CLS]",
|
| 3 |
+
"mask_token": "[MASK]",
|
| 4 |
+
"pad_token": "[PAD]",
|
| 5 |
+
"sep_token": "[SEP]",
|
| 6 |
+
"unk_token": "[UNK]"
|
| 7 |
+
}
|
tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,56 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"added_tokens_decoder": {
|
| 3 |
+
"0": {
|
| 4 |
+
"content": "[PAD]",
|
| 5 |
+
"lstrip": false,
|
| 6 |
+
"normalized": false,
|
| 7 |
+
"rstrip": false,
|
| 8 |
+
"single_word": false,
|
| 9 |
+
"special": true
|
| 10 |
+
},
|
| 11 |
+
"100": {
|
| 12 |
+
"content": "[UNK]",
|
| 13 |
+
"lstrip": false,
|
| 14 |
+
"normalized": false,
|
| 15 |
+
"rstrip": false,
|
| 16 |
+
"single_word": false,
|
| 17 |
+
"special": true
|
| 18 |
+
},
|
| 19 |
+
"101": {
|
| 20 |
+
"content": "[CLS]",
|
| 21 |
+
"lstrip": false,
|
| 22 |
+
"normalized": false,
|
| 23 |
+
"rstrip": false,
|
| 24 |
+
"single_word": false,
|
| 25 |
+
"special": true
|
| 26 |
+
},
|
| 27 |
+
"102": {
|
| 28 |
+
"content": "[SEP]",
|
| 29 |
+
"lstrip": false,
|
| 30 |
+
"normalized": false,
|
| 31 |
+
"rstrip": false,
|
| 32 |
+
"single_word": false,
|
| 33 |
+
"special": true
|
| 34 |
+
},
|
| 35 |
+
"103": {
|
| 36 |
+
"content": "[MASK]",
|
| 37 |
+
"lstrip": false,
|
| 38 |
+
"normalized": false,
|
| 39 |
+
"rstrip": false,
|
| 40 |
+
"single_word": false,
|
| 41 |
+
"special": true
|
| 42 |
+
}
|
| 43 |
+
},
|
| 44 |
+
"clean_up_tokenization_spaces": false,
|
| 45 |
+
"cls_token": "[CLS]",
|
| 46 |
+
"do_lower_case": true,
|
| 47 |
+
"extra_special_tokens": {},
|
| 48 |
+
"mask_token": "[MASK]",
|
| 49 |
+
"model_max_length": 512,
|
| 50 |
+
"pad_token": "[PAD]",
|
| 51 |
+
"sep_token": "[SEP]",
|
| 52 |
+
"strip_accents": null,
|
| 53 |
+
"tokenize_chinese_chars": true,
|
| 54 |
+
"tokenizer_class": "DistilBertTokenizer",
|
| 55 |
+
"unk_token": "[UNK]"
|
| 56 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:51a4264247fbe1fe88d05c68d61b8084763d2f8eb4fde109f383a88bcaed7463
|
| 3 |
+
size 5368
|
training_params.json
ADDED
|
@@ -0,0 +1,30 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"data_path": "samtuckervegan/text_performance",
|
| 3 |
+
"model": "distilbert/distilbert-base-uncased",
|
| 4 |
+
"lr": 5e-05,
|
| 5 |
+
"epochs": 3,
|
| 6 |
+
"max_seq_length": 128,
|
| 7 |
+
"batch_size": 8,
|
| 8 |
+
"warmup_ratio": 0.1,
|
| 9 |
+
"gradient_accumulation": 1,
|
| 10 |
+
"optimizer": "adamw_torch",
|
| 11 |
+
"scheduler": "linear",
|
| 12 |
+
"weight_decay": 0.0,
|
| 13 |
+
"max_grad_norm": 1.0,
|
| 14 |
+
"seed": 42,
|
| 15 |
+
"train_split": "train",
|
| 16 |
+
"valid_split": "test",
|
| 17 |
+
"text_column": "text",
|
| 18 |
+
"target_column": "target",
|
| 19 |
+
"logging_steps": -1,
|
| 20 |
+
"project_name": "textprediction",
|
| 21 |
+
"auto_find_batch_size": false,
|
| 22 |
+
"mixed_precision": "fp16",
|
| 23 |
+
"save_total_limit": 1,
|
| 24 |
+
"push_to_hub": true,
|
| 25 |
+
"eval_strategy": "epoch",
|
| 26 |
+
"username": "samtuckervegan",
|
| 27 |
+
"log": "tensorboard",
|
| 28 |
+
"early_stopping_patience": 5,
|
| 29 |
+
"early_stopping_threshold": 0.01
|
| 30 |
+
}
|
vocab.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|