SentenceTransformer based on google/embeddinggemma-300m

This is a sentence-transformers model finetuned from google/embeddinggemma-300m on the sebenx_triplets dataset. It maps sentences & paragraphs to a 768-dimensional dense vector space and can be used for semantic textual similarity, semantic search, paraphrase mining, text classification, clustering, and more.

Model Details

Model Description

Model Type: Sentence Transformer
Base model: google/embeddinggemma-300m
Maximum Sequence Length: 2048 tokens
Output Dimensionality: 768 dimensions
Similarity Function: Cosine Similarity
Training Dataset:
- sebenx_triplets

Model Sources

Documentation: Sentence Transformers Documentation
Repository: Sentence Transformers on GitHub
Hugging Face: Sentence Transformers on Hugging Face

Full Model Architecture

SentenceTransformer(
  (0): Transformer({'max_seq_length': 2048, 'do_lower_case': False, 'architecture': 'Gemma3TextModel'})
  (1): Pooling({'word_embedding_dimension': 768, 'pooling_mode_cls_token': False, 'pooling_mode_mean_tokens': True, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': False, 'include_prompt': True})
  (2): Dense({'in_features': 768, 'out_features': 3072, 'bias': False, 'activation_function': 'torch.nn.modules.linear.Identity'})
  (3): Dense({'in_features': 3072, 'out_features': 768, 'bias': False, 'activation_function': 'torch.nn.modules.linear.Identity'})
  (4): Normalize()
)

Usage

Direct Usage (Sentence Transformers)

First install the Sentence Transformers library:

pip install -U sentence-transformers

Then you can load this model and run inference.

from sentence_transformers import SentenceTransformer

# Download from the 🤗 Hub
model = SentenceTransformer("s7d11/SEmbedv1-0.3b")
# Run inference
queries = [
    "Lon b\u025b\u025b, a b\u025b feere damina kabini s\u0254g\u0254ma f\u0254 su f\u025b.",
]
documents = [
    'best tak dia?',
    'Sankasojanw camaw lajɛliw cɛkaɲi ani lajɛli kabɔ sankasojanw dɔ sanfɛlala walima kabɔ finɛtiri dɔla min pozisiyɔn kaɲi be seka kɛ fɛn cɛɲuman ye ka lajɛ.',
    'Pua ka wiliwili nanahu ka mano …',
]
query_embeddings = model.encode_query(queries)
document_embeddings = model.encode_document(documents)
print(query_embeddings.shape, document_embeddings.shape)
# [1, 768] [3, 768]

# Get the similarity scores for the embeddings
similarities = model.similarity(query_embeddings, document_embeddings)
print(similarities)
# tensor([[0.6840, 0.3838, 0.8201]])

Training Details

Training Dataset

sebenx_triplets

Dataset: sebenx_triplets at a152d73
Size: 2,703,977 training samples
Columns: anchor, positive, and negative

Approximate statistics based on the first 1000 samples:

	anchor	positive	negative
type	string	string	string
details	min: 4 tokens mean: 40.61 tokens max: 212 tokens	min: 4 tokens mean: 40.27 tokens max: 153 tokens	min: 3 tokens mean: 39.66 tokens max: 155 tokens

Samples:

anchor	positive	negative
`kalata fan wɛrɛw fɛ, kɛmɛ sarada jabi 29 hakilina ye Australie ka kɛ bɛjɛnjamana ye teliyara, waati minna kɛmɛ sarada 31 hakilina ye Australie man kan abada ka kɛ bɛjɛnjamana ye.`	`Tile minna ko nunuw kɛra be wele forobala ko Sanfɛ Cɛmancɛ Waati, Erɔpu tariku waati sankɛmɛsigi 11na, 12na ni 13na (san 1000-1300 Yesu bangeli kɔfɛ).`	`Nɛnɛ damatɛmɛlen be seka kɛ nisongoyala: goniyajakɛ be jigin cogogɛlɛ jali duguma, nka fiyɛn ani sumaya be faara ɲɔgɔnkan ka nɛnɛ tɔ juguya ka tɛmɛ goniyahakɛ sumana be min fɔ kan.`
`a mɛn ka kɛ tile kelen ye, walima fila,.`	`A tigi dalen ka kan ka to tile caman.`	`Nin ye mun ɲεnajε ye ?`
`Dja ko foyi tèkai ni sababou tala.`	`interj NZ a Māori greeting to two.`	`galoda Sinamuso jugu Nis kasara Gawo taama bɛnkan Ngɔnikɔrɔ bama bamukan Dinbal Fakɔkuru Lolɛ Ncininna Erɔp ntonlan.`

Loss: MatryoshkaLoss with these parameters:

{
    "loss": "MultipleNegativesRankingLoss",
    "matryoshka_dims": [
        768,
        512,
        256
    ],
    "matryoshka_weights": [
        1,
        0.3,
        0.1
    ],
    "n_dims_per_step": -1
}

Evaluation Dataset

sebenx_triplets

Dataset: sebenx_triplets at a152d73
Size: 271 evaluation samples
Columns: anchor, positive, and negative

Approximate statistics based on the first 271 samples:

	anchor	positive	negative
type	string	string	string
details	min: 4 tokens mean: 41.12 tokens max: 141 tokens	min: 4 tokens mean: 41.09 tokens max: 131 tokens	min: 4 tokens mean: 41.08 tokens max: 149 tokens

Samples:

anchor	positive	negative
`Kaladen kɔrɔ dɔ ko 'a tun be kumakan langolo fɔ kalanso kɔnɔ, ka jalali dɔnniw kalankɛ nɔbilaw la an'a tun be inafɔ kalandenw teri .'.`	`Kabini o tuma na Sini jamana sɔrɔ yiriwara siyɛn 90.`	`Nɔremu 802.11n be barakƐ fiɲƐsira 2,4 Ghz ni 5 Ghz kan.`
`maure.»`	`banxanxalle Noun. sore; dartre. Category: 2.5. Healthy. sim: fanqalelle.`	`Bambara, dioula, malinké`
`Barabara cia bũrũri itirĩ umithio mũnene angĩkoro irateithia mĩtoka mĩnini, kwa ũguo no nginya mĩtaratara ya gũthondeka na kũnyihia thogora ya cio ĩthondeko.`	`Sɔrɔboso yɛlɛmabolo fɔlɔ in kɛra Deng Xiaoping ka fanga kɔnɔ.`	`jàmanakuntigi ka Irisilataga Krowasiya Shipuru 2025-05 Mùneyisa 2025-04 Bilgariya Bɔsni 2025-03 2025 Selincinin Ne bɛ n.`

Loss: MatryoshkaLoss with these parameters:

{
    "loss": "MultipleNegativesRankingLoss",
    "matryoshka_dims": [
        768,
        512,
        256
    ],
    "matryoshka_weights": [
        1,
        0.3,
        0.1
    ],
    "n_dims_per_step": -1
}

Training Hyperparameters

Non-Default Hyperparameters

eval_strategy: steps
per_device_train_batch_size: 1
gradient_accumulation_steps: 8
learning_rate: 5e-06
weight_decay: 0.01
max_steps: 5000
lr_scheduler_type: cosine
warmup_ratio: 0.1
fp16: True
prompts: task: sentence similarity | query:

All Hyperparameters

Click to expand

overwrite_output_dir: False
do_predict: False
eval_strategy: steps
prediction_loss_only: True
per_device_train_batch_size: 1
per_device_eval_batch_size: 8
per_gpu_train_batch_size: None
per_gpu_eval_batch_size: None
gradient_accumulation_steps: 8
eval_accumulation_steps: None
torch_empty_cache_steps: None
learning_rate: 5e-06
weight_decay: 0.01
adam_beta1: 0.9
adam_beta2: 0.999
adam_epsilon: 1e-08
max_grad_norm: 1.0
num_train_epochs: 3
max_steps: 5000
lr_scheduler_type: cosine
lr_scheduler_kwargs: {}
warmup_ratio: 0.1
warmup_steps: 0
log_level: passive
log_level_replica: warning
log_on_each_node: True
logging_nan_inf_filter: True
save_safetensors: True
save_on_each_node: False
save_only_model: False
restore_callback_states_from_checkpoint: False
no_cuda: False
use_cpu: False
use_mps_device: False
seed: 42
data_seed: None
jit_mode_eval: False
bf16: False
fp16: True
fp16_opt_level: O1
half_precision_backend: auto
bf16_full_eval: False
fp16_full_eval: False
tf32: None
local_rank: 0
ddp_backend: None
tpu_num_cores: None
tpu_metrics_debug: False
debug: []
dataloader_drop_last: False
dataloader_num_workers: 0
dataloader_prefetch_factor: None
past_index: -1
disable_tqdm: False
remove_unused_columns: True
label_names: None
load_best_model_at_end: False
ignore_data_skip: False
fsdp: []
fsdp_min_num_params: 0
fsdp_config: {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}
fsdp_transformer_layer_cls_to_wrap: None
accelerator_config: {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}
parallelism_config: None
deepspeed: None
label_smoothing_factor: 0.0
optim: adamw_torch_fused
optim_args: None
adafactor: False
group_by_length: False
length_column_name: length
project: huggingface
trackio_space_id: trackio
ddp_find_unused_parameters: None
ddp_bucket_cap_mb: None
ddp_broadcast_buffers: False
dataloader_pin_memory: True
dataloader_persistent_workers: False
skip_memory_metrics: True
use_legacy_prediction_loop: False
push_to_hub: False
resume_from_checkpoint: None
hub_model_id: None
hub_strategy: every_save
hub_private_repo: None
hub_always_push: False
hub_revision: None
gradient_checkpointing: False
gradient_checkpointing_kwargs: None
include_inputs_for_metrics: False
include_for_metrics: []
eval_do_concat_batches: True
fp16_backend: auto
push_to_hub_model_id: None
push_to_hub_organization: None
mp_parameters:
auto_find_batch_size: False
full_determinism: False
torchdynamo: None
ray_scope: last
ddp_timeout: 1800
torch_compile: False
torch_compile_backend: None
torch_compile_mode: None
include_tokens_per_second: False
include_num_input_tokens_seen: no
neftune_noise_alpha: None
optim_target_modules: None
batch_eval_metrics: False
eval_on_start: False
use_liger_kernel: False
liger_kernel_config: None
eval_use_gather_object: False
average_tokens_across_devices: True
prompts: task: sentence similarity | query:
batch_sampler: batch_sampler
multi_dataset_batch_sampler: proportional
router_mapping: {}
learning_rate_mapping: {}

Training Logs

Click to expand

Epoch	Step	Training Loss	Validation Loss
0.0000	10	1.3857	-
0.0001	20	1.3352	-
0.0001	30	1.1998	-
0.0001	40	1.1487	-
0.0001	50	1.0881	-
0.0002	60	1.0239	-
0.0002	70	1.0615	-
0.0002	80	0.9406	-
0.0003	90	1.0496	-
0.0003	100	0.9214	-
0.0003	110	0.953	-
0.0004	120	0.8999	-
0.0004	130	1.0409	-
0.0004	140	0.9998	-
0.0004	150	0.883	-
0.0005	160	0.9923	-
0.0005	170	0.8498	-
0.0005	180	0.7824	-
0.0006	190	0.8451	-
0.0006	200	0.846	-
0.0006	210	0.8094	-
0.0007	220	0.9448	-
0.0007	230	0.9072	-
0.0007	240	0.7502	-
0.0007	250	0.7043	-
0.0008	260	0.777	-
0.0008	270	0.9626	-
0.0008	280	1.0501	-
0.0009	290	0.7489	-
0.0009	300	0.8305	-
0.0009	310	0.6411	-
0.0009	320	0.8251	-
0.0010	330	0.6632	-
0.0010	340	0.8039	-
0.0010	350	0.6465	-
0.0011	360	0.9541	-
0.0011	370	0.8584	-
0.0011	380	0.8907	-
0.0012	390	0.7243	-
0.0012	400	0.7592	-
0.0012	410	0.8199	-
0.0012	420	0.7394	-
0.0013	430	0.7053	-
0.0013	440	0.7955	-
0.0013	450	0.8382	-
0.0014	460	0.7608	-
0.0014	470	0.7308	-
0.0014	480	0.718	-
0.0014	490	0.6998	-
0.0015	500	0.8602	-
0.0015	510	0.9128	-
0.0015	520	0.6842	-
0.0016	530	0.8519	-
0.0016	540	0.9339	-
0.0016	550	0.8003	-
0.0017	560	0.7795	-
0.0017	570	0.8377	-
0.0017	580	0.5871	-
0.0017	590	0.7091	-
0.0018	600	0.79	-
0.0018	610	0.9473	-
0.0018	620	0.8671	-
0.0019	630	0.7165	-
0.0019	640	0.8825	-
0.0019	650	0.8347	-
0.0020	660	0.6016	-
0.0020	670	0.8572	-
0.0020	680	0.772	-
0.0020	690	0.7865	-
0.0021	700	0.9205	-
0.0021	710	0.8145	-
0.0021	720	0.783	-
0.0022	730	0.7304	-
0.0022	740	0.7809	-
0.0022	750	0.8504	-
0.0022	760	0.6971	-
0.0023	770	0.8535	-
0.0023	780	0.7312	-
0.0023	790	0.6701	-
0.0024	800	0.7899	-
0.0024	810	0.7688	-
0.0024	820	0.7493	-
0.0025	830	0.6789	-
0.0025	840	0.8506	-
0.0025	850	0.7875	-
0.0025	860	0.653	-
0.0026	870	0.7847	-
0.0026	880	0.7561	-
0.0026	890	0.6186	-
0.0027	900	0.6631	-
0.0027	910	0.7191	-
0.0027	920	0.666	-
0.0028	930	0.7304	-
0.0028	940	0.9292	-
0.0028	950	0.5899	-
0.0028	960	0.8594	-
0.0029	970	0.6226	-
0.0029	980	0.6514	-
0.0029	990	0.5353	-
0.0030	1000	0.5977	-
0.0030	1010	0.7201	-
0.0030	1020	0.8047	-
0.0030	1030	0.5933	-
0.0031	1040	0.7361	-
0.0031	1050	0.6687	-
0.0031	1060	0.8877	-
0.0032	1070	0.722	-
0.0032	1080	0.6555	-
0.0032	1090	0.7812	-
0.0033	1100	0.6015	-
0.0033	1110	0.7915	-
0.0033	1120	0.6999	-
0.0033	1130	0.6956	-
0.0034	1140	0.5105	-
0.0034	1150	0.7155	-
0.0034	1160	0.6233	-
0.0035	1170	0.9316	-
0.0035	1180	0.6544	-
0.0035	1190	0.6487	-
0.0036	1200	0.6459	-
0.0036	1210	0.8283	-
0.0036	1220	0.7653	-
0.0036	1230	0.7429	-
0.0037	1240	0.6253	-
0.0037	1250	0.7739	3.2216
0.0037	1260	0.5848	-
0.0038	1270	0.6655	-
0.0038	1280	0.6969	-
0.0038	1290	0.8055	-
0.0038	1300	0.5927	-
0.0039	1310	0.6252	-
0.0039	1320	0.7125	-
0.0039	1330	0.547	-
0.0040	1340	0.7427	-
0.0040	1350	0.8153	-
0.0040	1360	0.6979	-
0.0041	1370	0.6194	-
0.0041	1380	0.6441	-
0.0041	1390	0.5782	-
0.0041	1400	0.6529	-
0.0042	1410	0.8555	-
0.0042	1420	0.7904	-
0.0042	1430	0.5629	-
0.0043	1440	0.6203	-
0.0043	1450	0.6226	-
0.0043	1460	0.7346	-
0.0043	1470	0.8153	-
0.0044	1480	0.7878	-
0.0044	1490	0.769	-
0.0044	1500	0.6265	-
0.0045	1510	0.5634	-
0.0045	1520	0.5782	-
0.0045	1530	0.6093	-
0.0046	1540	0.6989	-
0.0046	1550	0.7951	-
0.0046	1560	0.6215	-
0.0046	1570	0.7065	-
0.0047	1580	0.6772	-
0.0047	1590	0.5745	-
0.0047	1600	0.6832	-
0.0048	1610	0.6318	-
0.0048	1620	0.7641	-
0.0048	1630	0.8019	-
0.0049	1640	0.7143	-
0.0049	1650	0.6369	-
0.0049	1660	0.6575	-
0.0049	1670	0.6055	-
0.0050	1680	0.675	-
0.0050	1690	0.5365	-
0.0050	1700	0.5092	-
0.0051	1710	0.7284	-
0.0051	1720	0.7647	-
0.0051	1730	0.5493	-
0.0051	1740	0.5061	-
0.0052	1750	0.5138	-
0.0052	1760	0.7677	-
0.0052	1770	0.5683	-
0.0053	1780	0.6337	-
0.0053	1790	0.5645	-
0.0053	1800	0.4971	-
0.0054	1810	0.7195	-
0.0054	1820	0.4615	-
0.0054	1830	0.7374	-
0.0054	1840	0.5524	-
0.0055	1850	0.7127	-
0.0055	1860	0.6545	-
0.0055	1870	0.6168	-
0.0056	1880	0.6194	-
0.0056	1890	0.4979	-
0.0056	1900	0.7268	-
0.0057	1910	0.4508	-
0.0057	1920	0.7093	-
0.0057	1930	0.6059	-
0.0057	1940	0.5363	-
0.0058	1950	0.659	-
0.0058	1960	0.5137	-
0.0058	1970	0.6528	-
0.0059	1980	0.6183	-
0.0059	1990	0.8905	-
0.0059	2000	0.6576	-
0.0059	2010	0.6094	-
0.0060	2020	0.5005	-
0.0060	2030	0.5099	-
0.0060	2040	0.631	-
0.0061	2050	0.4429	-
0.0061	2060	0.5831	-
0.0061	2070	0.6217	-
0.0062	2080	0.5121	-
0.0062	2090	0.5428	-
0.0062	2100	0.62	-
0.0062	2110	0.5721	-
0.0063	2120	0.5665	-
0.0063	2130	0.4057	-
0.0063	2140	0.7022	-
0.0064	2150	0.7608	-
0.0064	2160	0.6097	-
0.0064	2170	0.5711	-
0.0064	2180	0.4813	-
0.0065	2190	0.6525	-
0.0065	2200	0.6782	-
0.0065	2210	0.5661	-
0.0066	2220	0.754	-
0.0066	2230	0.6587	-
0.0066	2240	0.5377	-
0.0067	2250	0.8553	-
0.0067	2260	0.4283	-
0.0067	2270	0.6733	-
0.0067	2280	0.6693	-
0.0068	2290	0.5919	-
0.0068	2300	0.5743	-
0.0068	2310	0.7105	-
0.0069	2320	0.4436	-
0.0069	2330	0.6323	-
0.0069	2340	0.5959	-
0.0070	2350	0.6491	-
0.0070	2360	0.7986	-
0.0070	2370	0.5997	-
0.0070	2380	0.4897	-
0.0071	2390	0.5401	-
0.0071	2400	0.7304	-
0.0071	2410	0.5874	-
0.0072	2420	0.5637	-
0.0072	2430	0.5432	-
0.0072	2440	0.5799	-
0.0072	2450	0.5674	-
0.0073	2460	0.846	-
0.0073	2470	0.6006	-
0.0073	2480	0.5279	-
0.0074	2490	0.706	-
0.0074	2500	0.5741	3.0077
0.0074	2510	0.5416	-
0.0075	2520	0.448	-
0.0075	2530	0.5437	-
0.0075	2540	0.662	-
0.0075	2550	0.6424	-
0.0076	2560	0.682	-
0.0076	2570	0.6211	-
0.0076	2580	0.5738	-
0.0077	2590	0.5747	-
0.0077	2600	0.959	-
0.0077	2610	0.56	-
0.0078	2620	0.6612	-
0.0078	2630	0.5008	-
0.0078	2640	0.4839	-
0.0078	2650	0.6241	-
0.0079	2660	0.6323	-
0.0079	2670	0.6601	-
0.0079	2680	0.517	-
0.0080	2690	0.6023	-
0.0080	2700	0.5601	-
0.0080	2710	0.611	-
0.0080	2720	0.7261	-
0.0081	2730	0.515	-
0.0081	2740	0.5517	-
0.0081	2750	0.5843	-
0.0082	2760	0.4607	-
0.0082	2770	0.5416	-
0.0082	2780	0.6806	-
0.0083	2790	0.6127	-
0.0083	2800	0.6366	-
0.0083	2810	0.6962	-
0.0083	2820	0.4876	-
0.0084	2830	0.7263	-
0.0084	2840	0.5974	-
0.0084	2850	0.4835	-
0.0085	2860	0.4579	-
0.0085	2870	0.429	-
0.0085	2880	0.4439	-
0.0086	2890	0.5631	-
0.0086	2900	0.6307	-
0.0086	2910	0.5138	-
0.0086	2920	0.617	-
0.0087	2930	0.5033	-
0.0087	2940	0.6152	-
0.0087	2950	0.5089	-
0.0088	2960	0.4937	-
0.0088	2970	0.5528	-
0.0088	2980	0.5194	-
0.0088	2990	0.772	-
0.0089	3000	0.5303	-
0.0089	3010	0.565	-
0.0089	3020	0.5464	-
0.0090	3030	0.6153	-
0.0090	3040	0.5965	-
0.0090	3050	0.712	-
0.0091	3060	0.4347	-
0.0091	3070	0.4398	-
0.0091	3080	0.6925	-
0.0091	3090	0.8619	-
0.0092	3100	0.7581	-
0.0092	3110	0.8109	-
0.0092	3120	0.4329	-
0.0093	3130	0.4853	-
0.0093	3140	0.5674	-
0.0093	3150	0.6655	-
0.0093	3160	0.48	-
0.0094	3170	0.3521	-
0.0094	3180	0.5814	-
0.0094	3190	0.4354	-
0.0095	3200	0.6543	-
0.0095	3210	0.5167	-
0.0095	3220	0.8639	-
0.0096	3230	0.48	-
0.0096	3240	0.6677	-
0.0096	3250	0.6518	-
0.0096	3260	0.5602	-
0.0097	3270	0.589	-
0.0097	3280	0.6361	-
0.0097	3290	0.6589	-
0.0098	3300	0.5138	-
0.0098	3310	0.5356	-
0.0098	3320	0.533	-
0.0099	3330	0.6241	-
0.0099	3340	0.6112	-
0.0099	3350	0.5351	-
0.0099	3360	0.4903	-
0.0100	3370	0.4544	-
0.0100	3380	0.4495	-
0.0100	3390	0.4382	-
0.0101	3400	0.5671	-
0.0101	3410	0.4735	-
0.0101	3420	0.638	-
0.0101	3430	0.5626	-
0.0102	3440	0.4754	-
0.0102	3450	0.4749	-
0.0102	3460	0.4778	-
0.0103	3470	0.3425	-
0.0103	3480	0.5415	-
0.0103	3490	0.5165	-
0.0104	3500	0.6016	-
0.0104	3510	0.5639	-
0.0104	3520	0.8738	-
0.0104	3530	0.5062	-
0.0105	3540	0.4332	-
0.0105	3550	0.8084	-
0.0105	3560	0.7191	-
0.0106	3570	0.5944	-
0.0106	3580	0.6997	-
0.0106	3590	0.63	-
0.0107	3600	0.4186	-
0.0107	3610	0.5776	-
0.0107	3620	0.4875	-
0.0107	3630	0.5769	-
0.0108	3640	0.509	-
0.0108	3650	0.5627	-
0.0108	3660	0.5159	-
0.0109	3670	0.6378	-
0.0109	3680	0.4965	-
0.0109	3690	0.5775	-
0.0109	3700	0.657	-
0.0110	3710	0.7192	-
0.0110	3720	0.3836	-
0.0110	3730	0.6142	-
0.0111	3740	0.4774	-
0.0111	3750	0.5099	2.9023
0.0111	3760	0.6325	-
0.0112	3770	0.6974	-
0.0112	3780	0.5958	-
0.0112	3790	0.5643	-
0.0112	3800	0.5476	-
0.0113	3810	0.3828	-
0.0113	3820	0.7134	-
0.0113	3830	0.5593	-
0.0114	3840	0.4622	-
0.0114	3850	0.4911	-
0.0114	3860	0.7652	-
0.0114	3870	0.4124	-
0.0115	3880	0.7257	-
0.0115	3890	0.459	-
0.0115	3900	0.4988	-
0.0116	3910	0.5146	-
0.0116	3920	0.5613	-
0.0116	3930	0.6893	-
0.0117	3940	0.4245	-
0.0117	3950	0.4426	-
0.0117	3960	0.8301	-
0.0117	3970	0.3732	-
0.0118	3980	0.516	-
0.0118	3990	0.445	-
0.0118	4000	0.838	-
0.0119	4010	0.6627	-
0.0119	4020	0.3563	-
0.0119	4030	0.532	-
0.0120	4040	0.7707	-
0.0120	4050	0.5832	-
0.0120	4060	0.5266	-
0.0120	4070	0.5309	-
0.0121	4080	0.6722	-
0.0121	4090	0.5141	-
0.0121	4100	0.4724	-
0.0122	4110	0.7266	-
0.0122	4120	0.4685	-
0.0122	4130	0.4988	-
0.0122	4140	0.4194	-
0.0123	4150	0.4976	-
0.0123	4160	0.5164	-
0.0123	4170	0.6077	-
0.0124	4180	0.6547	-
0.0124	4190	0.6342	-
0.0124	4200	0.5514	-
0.0125	4210	0.4814	-
0.0125	4220	0.4895	-
0.0125	4230	0.7219	-
0.0125	4240	0.5481	-
0.0126	4250	0.4702	-
0.0126	4260	0.7058	-
0.0126	4270	0.3936	-
0.0127	4280	0.6489	-
0.0127	4290	0.5032	-
0.0127	4300	0.5088	-
0.0128	4310	0.523	-
0.0128	4320	0.4418	-
0.0128	4330	0.583	-
0.0128	4340	0.564	-
0.0129	4350	0.6308	-
0.0129	4360	0.5444	-
0.0129	4370	0.5474	-
0.0130	4380	0.4261	-
0.0130	4390	0.5347	-
0.0130	4400	0.6137	-
0.0130	4410	0.4739	-
0.0131	4420	0.5185	-
0.0131	4430	0.4315	-
0.0131	4440	0.5913	-
0.0132	4450	0.5222	-
0.0132	4460	0.4818	-
0.0132	4470	0.5603	-
0.0133	4480	0.6157	-
0.0133	4490	0.6436	-
0.0133	4500	0.6227	-
0.0133	4510	0.4639	-
0.0134	4520	0.6379	-
0.0134	4530	0.5369	-
0.0134	4540	0.4951	-
0.0135	4550	0.5235	-
0.0135	4560	0.5048	-
0.0135	4570	0.4953	-
0.0136	4580	0.6981	-
0.0136	4590	0.5543	-
0.0136	4600	0.5432	-
0.0136	4610	0.4719	-
0.0137	4620	0.5418	-
0.0137	4630	0.7021	-
0.0137	4640	0.5176	-
0.0138	4650	0.459	-
0.0138	4660	0.6334	-
0.0138	4670	0.4691	-
0.0138	4680	0.4473	-
0.0139	4690	0.474	-
0.0139	4700	0.5297	-
0.0139	4710	0.6543	-
0.0140	4720	0.5651	-
0.0140	4730	0.5072	-
0.0140	4740	0.5961	-
0.0141	4750	0.5262	-
0.0141	4760	0.6235	-
0.0141	4770	0.4718	-
0.0141	4780	0.497	-
0.0142	4790	0.5046	-
0.0142	4800	0.5694	-
0.0142	4810	0.4202	-
0.0143	4820	0.6833	-
0.0143	4830	0.6341	-
0.0143	4840	0.5327	-
0.0143	4850	0.4914	-
0.0144	4860	0.6098	-
0.0144	4870	0.4093	-
0.0144	4880	0.5317	-
0.0145	4890	0.5809	-
0.0145	4900	0.3474	-
0.0145	4910	0.4408	-
0.0146	4920	0.4957	-
0.0146	4930	0.5085	-
0.0146	4940	0.5089	-
0.0146	4950	0.6008	-
0.0147	4960	0.3984	-
0.0147	4970	0.489	-
0.0147	4980	0.3918	-
0.0148	4990	0.6235	-
0.0148	5000	0.5644	2.8565

Framework Versions

Python: 3.12.3
Sentence Transformers: 5.2.0
Transformers: 4.57.3
PyTorch: 2.9.1+cu128
Accelerate: 1.12.0
Datasets: 4.4.2
Tokenizers: 0.22.1

Citation

BibTeX

Sentence Transformers

@inproceedings{reimers-2019-sentence-bert,
    title = "Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks",
    author = "Reimers, Nils and Gurevych, Iryna",
    booktitle = "Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing",
    month = "11",
    year = "2019",
    publisher = "Association for Computational Linguistics",
    url = "https://arxiv.org/abs/1908.10084",
}

MatryoshkaLoss

@misc{kusupati2024matryoshka,
    title={Matryoshka Representation Learning},
    author={Aditya Kusupati and Gantavya Bhatt and Aniket Rege and Matthew Wallingford and Aditya Sinha and Vivek Ramanujan and William Howard-Snyder and Kaifeng Chen and Sham Kakade and Prateek Jain and Ali Farhadi},
    year={2024},
    eprint={2205.13147},
    archivePrefix={arXiv},
    primaryClass={cs.LG}
}

MultipleNegativesRankingLoss

@misc{henderson2017efficient,
    title={Efficient Natural Language Response Suggestion for Smart Reply},
    author={Matthew Henderson and Rami Al-Rfou and Brian Strope and Yun-hsuan Sung and Laszlo Lukacs and Ruiqi Guo and Sanjiv Kumar and Balint Miklos and Ray Kurzweil},
    year={2017},
    eprint={1705.00652},
    archivePrefix={arXiv},
    primaryClass={cs.CL}
}

Downloads last month: 32

Safetensors

Model size

0.3B params

Tensor type

F32

Model tree for s7d11/SEmbedv1-0.3b

Base model

google/embeddinggemma-300m

Finetuned

(168)

this model

Dataset used to train s7d11/SEmbedv1-0.3b

Papers for s7d11/SEmbedv1-0.3b