nguyenhuucongzz01/stella_en_1.5B_fine_tuned

SentenceTransformer based on NovaSearch/stella_en_1.5B_v5

This is a sentence-transformers model finetuned from NovaSearch/stella_en_1.5B_v5. It maps sentences & paragraphs to a 1024-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: NovaSearch/stella_en_1.5B_v5
• Maximum Sequence Length: 512 tokens
• Output Dimensionality: 1024 dimensions
• Similarity Function: Cosine Similarity

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': 512, 'do_lower_case': False}) with Transformer model: PeftModelForFeatureExtraction 
  (1): Pooling({'word_embedding_dimension': 1536, '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': 1536, 'out_features': 1024, 'bias': True, 'activation_function': 'torch.nn.modules.linear.Identity'})
)

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("sentence_transformers_model_id")
# Run inference
sentences = [
    "Subject: Adding and Subtracting with Decimals\nConstruct: Subtract decimals where the numbers involved have a different number of decimal places\nQuestion: $0.55-0.2=0.55-0.2=$\nCorrectAnswer: $0.350.35$\nIncorrectAnswer: $0.330.33$\nIncorrectReason: The correct answer to the problem $0.55-0.20.55\; -\; 0.2$ is $0.350.35$. This is correct because when you subtract $0.20.2$ from $0.550.55$, you are essentially performing the operation $0.55-0.200.55\; -\; 0.20$. This can be visualized as $5555$ hundredths minus $2020$ hundredths, which equals $3535$ hundredths, or $0.350.35$.\n\nThe incorrect answer $0.330.33$ likely stems from a common misconception or a calculation error. One possible reason for this mistake is a misunderstanding of decimal subtraction or a misinterpretation of the place values. For example, someone might incorrectly think that $0.55-0.20.55\; -\; 0.2$ is the same as $0.55-0.220.55\; -\; 0.22$, leading to $0.330.33$. Alternatively, the error could be due to a simple arithmetic mistake, such as not properly aligning the decimal points during the subtraction process. It's important to ensure that the decimal points are aligned correctly and to understand the value of each digit in the decimal places.",
    'When subtracting decimals with a different number of decimals, subtracts one digit from more than one column',
    'Does not know that 7 and -7 are different',
]
embeddings = model.encode(sentences)
print(embeddings.shape)
# [3, 1024]

# Get the similarity scores for the embeddings
similarities = model.similarity(embeddings, embeddings)
print(similarities.shape)
# [3, 3]

Evaluation

Metrics

Information Retrieval

• Dataset: val
• Evaluated with InformationRetrievalEvaluator

Metric	Value
cosine_accuracy@25	0.7408
cosine_precision@50	0.0168
cosine_precision@100	0.009
cosine_precision@150	0.0063
cosine_precision@200	0.0048
cosine_recall@50	0.8372
cosine_recall@100	0.9002
cosine_recall@150	0.9381
cosine_recall@200	0.9587
cosine_ndcg@25	0.3986
cosine_mrr@25	0.3017
cosine_map@25	0.3017

Training Details

Training Dataset

Unnamed Dataset

• Size: 8,562 training samples
• Columns: anchor, positive, and negative
• Approximate statistics based on the first 1000 samples:

  	anchor	positive	negative
  type	string	string	string
  details	min: 190 tokens mean: 331.9 tokens max: 512 tokens	min: 4 tokens mean: 13.79 tokens max: 42 tokens	min: 4 tokens mean: 15.33 tokens max: 41 tokens

• Samples:

  anchor	positive	negative
  Subject: Function Machines Construct: Calculate the square of a number Question: What is the input of this function machine? CorrectAnswer: ( 256 ) IncorrectAnswer: ( 8 ) IncorrectReason: The correct answer is ( 256 ) because the function machine involves taking the square root of the input to produce the output. Given that the output is ( 16 ), we need to find a number whose square root is ( 16 ). Mathematically, this means solving the equation ( \sqrt{x} = 16 ). Squaring both sides, we get ( x = 16^2 = 256 ). Therefore, the input must be ( 256 ). The incorrect answer ( 8 ) likely stems from a common misconception. Someone might have seen the output ( 16 ) and thou...	Mixes up squaring and multiplying by 2 or doubling	Confuses written 'teen' numbers with their corresponding single-digit number
  Subject: Function Machines Construct: Calculate the square of a number Question: What is the input of this function machine? CorrectAnswer: ( 256 ) IncorrectAnswer: ( 8 ) IncorrectReason: The correct answer is ( 256 ) because the function machine involves taking the square root of the input to produce the output. Given that the output is ( 16 ), we need to find a number whose square root is ( 16 ). Mathematically, this means solving the equation ( \sqrt{x} = 16 ). Squaring both sides, we get ( x = 16^2 = 256 ). Therefore, the input must be ( 256 ). The incorrect answer ( 8 ) likely stems from a common misconception. Someone might have seen the output ( 16 ) and thou...	Mixes up squaring and multiplying by 2 or doubling	When multiplying multiples of ten and the answer requires an extra digit, leaves off that extra digit
  Subject: Ratio and Proportion Construct: Convert between currencies given an exchange rate Question: Convert 350 Thai baht to Australian Dollars.\n1 Australian dollar = 25 Thai baht CorrectAnswer: 14 IncorrectAnswer: 350 IncorrectReason: The correct answer is 14 Australian dollars. This is because the conversion rate given is 1 Australian dollar (AUD) equals 25 Thai baht (THB). To convert 350 THB to AUD, you divide 350 by 25, which equals 14 AUD. This calculation correctly reflects the exchange rate and the amount of money being converted. The incorrect answer of 350 is likely due to a misunderstanding of the conversion process. Someone might have mistakenly thought that the amount in Thai baht is the same in Australian dollars, not taking into account the exchange rate. This error occurs when the person fails to apply the conversion factor, instead assuming that the currency values are equivalent without adjustment. This misconception can lead to significant errors in financial trans...	Assumes a 1:1 conversion ratio between different currencies	Believes that the larger the divisor, the larger the answer.

• Loss: CachedMultipleNegativesRankingLoss with these parameters:

  {
      "scale": 20.0,
      "similarity_fct": "cos_sim"
  }
  

Training Hyperparameters

Non-Default Hyperparameters

• eval_strategy: steps
• per_device_train_batch_size: 16
• per_device_eval_batch_size: 4
• learning_rate: 0.001
• num_train_epochs: 1.0
• lr_scheduler_type: cosine
• save_only_model: True
• bf16: True
• load_best_model_at_end: True
• batch_sampler: no_duplicates

All Hyperparameters

Click to expand

• overwrite_output_dir: False
• do_predict: False
• eval_strategy: steps
• prediction_loss_only: True
• per_device_train_batch_size: 16
• per_device_eval_batch_size: 4
• per_gpu_train_batch_size: None
• per_gpu_eval_batch_size: None
• gradient_accumulation_steps: 1
• eval_accumulation_steps: None
• torch_empty_cache_steps: None
• learning_rate: 0.001
• weight_decay: 0.0
• adam_beta1: 0.9
• adam_beta2: 0.999
• adam_epsilon: 1e-08
• max_grad_norm: 1.0
• num_train_epochs: 1.0
• max_steps: -1
• lr_scheduler_type: cosine
• lr_scheduler_kwargs: {}
• warmup_ratio: 0.0
• 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: True
• 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
• use_ipex: False
• bf16: True
• fp16: False
• 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: True
• 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}
• tp_size: 0
• 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}
• deepspeed: None
• label_smoothing_factor: 0.0
• optim: adamw_torch
• optim_args: None
• adafactor: False
• group_by_length: False
• length_column_name: length
• 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
• 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: False
• neftune_noise_alpha: None
• optim_target_modules: None
• batch_eval_metrics: False
• eval_on_start: False
• use_liger_kernel: False
• eval_use_gather_object: False
• average_tokens_across_devices: False
• prompts: None
• batch_sampler: no_duplicates
• multi_dataset_batch_sampler: proportional

Training Logs

Click to expand

Epoch	Step	Training Loss	val_cosine_ndcg@25
0.0019	1	0.7341	-
0.0037	2	1.1246	-
0.0056	3	1.1668	-
0.0075	4	1.2752	-
0.0093	5	1.2428	-
0.0112	6	0.8722	-
0.0131	7	0.9877	-
0.0149	8	0.3914	-
0.0168	9	1.6333	-
0.0187	10	0.3793	0.3873
0.0205	11	0.3277	-
0.0224	12	0.689	-
0.0243	13	1.4066	-
0.0261	14	0.7874	-
0.0280	15	0.7898	-
0.0299	16	1.0844	-
0.0317	17	1.0972	-
0.0336	18	1.7414	-
0.0354	19	0.9649	-
0.0373	20	0.9025	0.3383
0.0392	21	1.0195	-
0.0410	22	1.5774	-
0.0429	23	2.6835	-
0.0448	24	1.9685	-
0.0466	25	1.5736	-
0.0485	26	0.4385	-
0.0504	27	1.5777	-
0.0522	28	0.5438	-
0.0541	29	1.1351	-
0.0560	30	0.4636	0.3349
0.0578	31	1.749	-
0.0597	32	0.6608	-
0.0616	33	1.48	-
0.0634	34	0.6442	-
0.0653	35	1.2882	-
0.0672	36	1.5927	-
0.0690	37	0.819	-
0.0709	38	0.5842	-
0.0728	39	0.4818	-
0.0746	40	0.5143	0.3079
0.0765	41	1.4064	-
0.0784	42	0.924	-
0.0802	43	0.9097	-
0.0821	44	0.4214	-
0.0840	45	1.2579	-
0.0858	46	3.0192	-
0.0877	47	0.9019	-
0.0896	48	0.8331	-
0.0914	49	2.1336	-
0.0933	50	0.3793	0.3332
0.0951	51	0.6568	-
0.0970	52	0.7644	-
0.0989	53	1.1422	-
0.1007	54	1.1733	-
0.1026	55	1.1297	-
0.1045	56	0.7746	-
0.1063	57	1.2374	-
0.1082	58	1.0382	-
0.1101	59	0.8722	-
0.1119	60	1.6862	0.2076
0.1138	61	0.9489	-
0.1157	62	1.6074	-
0.1175	63	2.3639	-
0.1194	64	1.2994	-
0.1213	65	1.3806	-
0.1231	66	1.6077	-
0.125	67	1.2359	-
0.1269	68	1.2202	-
0.1287	69	0.8442	-
0.1306	70	0.8537	0.2768
0.1325	71	2.2377	-
0.1343	72	1.0657	-
0.1362	73	0.6213	-
0.1381	74	1.2029	-
0.1399	75	1.4392	-
0.1418	76	0.7116	-
0.1437	77	1.228	-
0.1455	78	0.9498	-
0.1474	79	1.1289	-
0.1493	80	1.6371	0.2504
0.1511	81	0.438	-
0.1530	82	1.0909	-
0.1549	83	0.8301	-
0.1567	84	0.9003	-
0.1586	85	1.8428	-
0.1604	86	2.7758	-
0.1623	87	3.7156	-
0.1642	88	2.6085	-
0.1660	89	2.2705	-
0.1679	90	1.4518	0.2239
0.1698	91	1.3423	-
0.1716	92	1.4066	-
0.1735	93	2.3138	-
0.1754	94	2.256	-
0.1772	95	1.2564	-
0.1791	96	1.477	-
0.1810	97	2.8484	-
0.1828	98	1.3257	-
0.1847	99	1.1516	-
0.1866	100	1.2892	0.2142
0.1884	101	1.7179	-
0.1903	102	2.2282	-
0.1922	103	0.9497	-
0.1940	104	0.9663	-
0.1959	105	1.2476	-
0.1978	106	1.0585	-
0.1996	107	1.565	-
0.2015	108	1.4498	-
0.2034	109	1.237	-
0.2052	110	1.9519	0.1239
0.2071	111	2.4816	-
0.2090	112	2.3602	-
0.2108	113	0.5189	-
0.2127	114	2.1441	-
0.2146	115	1.9018	-
0.2164	116	1.1875	-
0.2183	117	1.033	-
0.2201	118	1.7925	-
0.2220	119	1.1472	-
0.2239	120	1.0008	0.2699
0.2257	121	1.4836	-
0.2276	122	0.9753	-
0.2295	123	0.7691	-
0.2313	124	0.9119	-
0.2332	125	0.7913	-
0.2351	126	1.4574	-
0.2369	127	1.3908	-
0.2388	128	1.2722	-
0.2407	129	0.3513	-
0.2425	130	1.2904	0.2267
0.2444	131	1.1935	-
0.2463	132	2.024	-
0.2481	133	1.2138	-
0.25	134	1.909	-
0.2519	135	1.4939	-
0.2537	136	2.5559	-
0.2556	137	1.1896	-
0.2575	138	1.5372	-
0.2593	139	1.3159	-
0.2612	140	2.8622	0.1801
0.2631	141	2.2284	-
0.2649	142	1.1668	-
0.2668	143	1.5383	-
0.2687	144	1.6872	-
0.2705	145	1.3499	-
0.2724	146	1.7111	-
0.2743	147	0.8461	-
0.2761	148	1.0737	-
0.2780	149	1.2229	-
0.2799	150	1.4991	0.2705
0.2817	151	1.2098	-
0.2836	152	0.8411	-
0.2854	153	0.7454	-
0.2873	154	0.5295	-
0.2892	155	1.2309	-
0.2910	156	1.1437	-
0.2929	157	1.3461	-
0.2948	158	1.1028	-
0.2966	159	1.6687	-
0.2985	160	1.1048	0.2228
0.3004	161	1.4661	-
0.3022	162	2.3891	-
0.3041	163	2.0019	-
0.3060	164	1.9604	-
0.3078	165	2.1173	-
0.3097	166	1.2352	-
0.3116	167	1.0883	-
0.3134	168	1.0343	-
0.3153	169	0.6048	-
0.3172	170	1.2634	0.2747
0.3190	171	0.724	-
0.3209	172	0.5937	-
0.3228	173	0.9735	-
0.3246	174	1.1059	-
0.3265	175	0.5561	-
0.3284	176	0.9019	-
0.3302	177	0.6012	-
0.3321	178	0.6203	-
0.3340	179	0.4729	-
0.3358	180	0.488	0.2880
0.3377	181	0.5171	-
0.3396	182	1.2202	-
0.3414	183	0.4338	-
0.3433	184	0.2286	-
0.3451	185	1.5921	-
0.3470	186	0.9065	-
0.3489	187	0.7728	-
0.3507	188	0.6743	-
0.3526	189	0.6354	-
0.3545	190	1.0883	0.3092
0.3563	191	0.7866	-
0.3582	192	0.4465	-
0.3601	193	0.9169	-
0.3619	194	1.2751	-
0.3638	195	0.6479	-
0.3657	196	1.0898	-
0.3675	197	0.4064	-
0.3694	198	1.216	-
0.3713	199	0.5892	-
0.3731	200	0.9736	0.2627
0.375	201	1.8989	-
0.3769	202	1.4159	-
0.3787	203	1.4947	-
0.3806	204	1.6758	-
0.3825	205	1.1081	-
0.3843	206	1.1187	-
0.3862	207	1.7538	-
0.3881	208	2.3149	-
0.3899	209	0.7799	-
0.3918	210	0.7268	0.2772
0.3937	211	0.6603	-
0.3955	212	1.034	-
0.3974	213	0.765	-
0.3993	214	1.8519	-
0.4011	215	1.6521	-
0.4030	216	1.7584	-
0.4049	217	2.2637	-
0.4067	218	1.1289	-
0.4086	219	1.9741	-
0.4104	220	1.8754	0.1599
0.4123	221	1.8528	-
0.4142	222	2.1507	-
0.4160	223	2.1293	-
0.4179	224	0.9261	-
0.4198	225	1.2636	-
0.4216	226	1.7696	-
0.4235	227	1.0828	-
0.4254	228	1.533	-
0.4272	229	1.438	-
0.4291	230	0.9375	0.2517
0.4310	231	0.8709	-
0.4328	232	1.0026	-
0.4347	233	1.0076	-
0.4366	234	0.8922	-
0.4384	235	0.828	-
0.4403	236	1.111	-
0.4422	237	1.5364	-
0.4440	238	0.9463	-
0.4459	239	1.059	-
0.4478	240	1.4188	0.1832
0.4496	241	1.7641	-
0.4515	242	1.4712	-
0.4534	243	1.2123	-
0.4552	244	0.9881	-
0.4571	245	2.1159	-
0.4590	246	1.073	-
0.4608	247	0.3211	-
0.4627	248	1.7917	-
0.4646	249	0.6342	-
0.4664	250	1.3472	0.2687
0.4683	251	0.492	-
0.4701	252	1.0642	-
0.4720	253	0.6704	-
0.4739	254	0.6744	-
0.4757	255	1.7866	-
0.4776	256	1.2805	-
0.4795	257	1.0666	-
0.4813	258	2.4739	-
0.4832	259	2.7657	-
0.4851	260	2.4601	0.1183
0.4869	261	2.5174	-
0.4888	262	2.7207	-
0.4907	263	2.7801	-
0.4925	264	1.2408	-
0.4944	265	2.3538	-
0.4963	266	2.2384	-
0.4981	267	1.4689	-
0.5	268	1.6905	-
0.5019	269	1.4729	-
0.5037	270	1.2211	0.2667
0.5056	271	0.6759	-
0.5075	272	0.8592	-
0.5093	273	0.4822	-
0.5112	274	1.2476	-
0.5131	275	0.6806	-
0.5149	276	1.3813	-
0.5168	277	0.7919	-
0.5187	278	0.7511	-
0.5205	279	0.6702	-
0.5224	280	0.8166	0.3069
0.5243	281	0.3796	-
0.5261	282	0.7048	-
0.5280	283	1.2978	-
0.5299	284	0.7682	-
0.5317	285	0.554	-
0.5336	286	1.0344	-
0.5354	287	0.8375	-
0.5373	288	0.361	-
0.5392	289	0.3193	-
0.5410	290	0.7264	0.2902
0.5429	291	1.2829	-
0.5448	292	1.6457	-
0.5466	293	0.9561	-
0.5485	294	1.2187	-
0.5504	295	1.5597	-
0.5522	296	1.6294	-
0.5541	297	0.9754	-
0.5560	298	1.121	-
0.5578	299	1.0038	-
0.5597	300	1.472	0.2603
0.5616	301	1.1317	-
0.5634	302	0.678	-
0.5653	303	1.2261	-
0.5672	304	1.4552	-
0.5690	305	0.7346	-
0.5709	306	1.2259	-
0.5728	307	0.5651	-
0.5746	308	0.5246	-
0.5765	309	0.5817	-
0.5784	310	1.0662	0.2983
0.5802	311	1.2422	-
0.5821	312	0.9479	-
0.5840	313	0.8528	-
0.5858	314	0.9502	-
0.5877	315	1.0885	-
0.5896	316	1.4663	-
0.5914	317	0.6274	-
0.5933	318	1.0567	-
0.5951	319	1.4394	-
0.5970	320	0.455	0.2463
0.5989	321	0.5577	-
0.6007	322	0.7305	-
0.6026	323	1.3569	-
0.6045	324	1.9528	-
0.6063	325	0.7332	-
0.6082	326	1.6955	-
0.6101	327	1.5237	-
0.6119	328	2.0396	-
0.6138	329	1.913	-
0.6157	330	1.8478	0.0902
0.6175	331	2.7965	-
0.6194	332	2.4383	-
0.6213	333	3.3085	-
0.6231	334	2.4657	-
0.625	335	2.3933	-
0.6269	336	2.3603	-
0.6287	337	1.3248	-
0.6306	338	1.568	-
0.6325	339	1.6271	-
0.6343	340	1.3838	0.1664
0.6362	341	2.0098	-
0.6381	342	1.7105	-
0.6399	343	1.2461	-
0.6418	344	1.293	-
0.6437	345	1.4298	-
0.6455	346	1.7789	-
0.6474	347	1.0361	-
0.6493	348	0.6129	-
0.6511	349	1.5476	-
0.6530	350	0.8251	0.2059
0.6549	351	0.9453	-
0.6567	352	1.1893	-
0.6586	353	0.7976	-
0.6604	354	0.5457	-
0.6623	355	0.6489	-
0.6642	356	1.0474	-
0.6660	357	1.0201	-
0.6679	358	0.5917	-
0.6698	359	1.0068	-
0.6716	360	0.5708	0.2568
0.6735	361	0.6778	-
0.6754	362	0.5382	-
0.6772	363	0.9939	-
0.6791	364	0.7322	-
0.6810	365	1.1926	-
0.6828	366	1.5369	-
0.6847	367	0.9815	-
0.6866	368	0.8891	-
0.6884	369	1.2503	-
0.6903	370	0.9369	0.2584
0.6922	371	0.538	-
0.6940	372	0.7312	-
0.6959	373	1.1477	-
0.6978	374	1.9885	-
0.6996	375	0.9605	-
0.7015	376	0.7769	-
0.7034	377	0.7701	-
0.7052	378	0.7166	-
0.7071	379	0.9712	-
0.7090	380	0.2171	0.3315
0.7108	381	1.1501	-
0.7127	382	0.9079	-
0.7146	383	0.3611	-
0.7164	384	0.1937	-
0.7183	385	0.5164	-
0.7201	386	1.4014	-
0.7220	387	0.5033	-
0.7239	388	0.7722	-
0.7257	389	0.1686	-
0.7276	390	0.5965	0.3521
0.7295	391	0.2465	-
0.7313	392	0.2342	-
0.7332	393	0.6155	-
0.7351	394	0.6689	-
0.7369	395	0.4981	-
0.7388	396	0.4915	-
0.7407	397	0.5064	-
0.7425	398	1.244	-
0.7444	399	0.8528	-
0.7463	400	0.6747	0.3463
0.7481	401	0.3525	-
0.75	402	1.2951	-
0.7519	403	0.6925	-
0.7537	404	0.7087	-
0.7556	405	0.1436	-
0.7575	406	0.6327	-
0.7593	407	0.3393	-
0.7612	408	0.5633	-
0.7631	409	0.6249	-
0.7649	410	1.5898	0.3513
0.7668	411	0.6968	-
0.7687	412	0.9603	-
0.7705	413	0.4476	-
0.7724	414	0.9167	-
0.7743	415	1.2049	-
0.7761	416	0.4518	-
0.7780	417	0.6315	-
0.7799	418	0.2537	-
0.7817	419	0.6812	-
0.7836	420	0.6971	0.3573
0.7854	421	0.6064	-
0.7873	422	0.4359	-
0.7892	423	0.4889	-
0.7910	424	0.7253	-
0.7929	425	0.519	-
0.7948	426	0.2237	-
0.7966	427	0.3144	-
0.7985	428	0.7395	-
0.8004	429	0.5903	-
0.8022	430	1.3353	0.3664
0.8041	431	0.5381	-
0.8060	432	0.5692	-
0.8078	433	0.3789	-
0.8097	434	0.4091	-
0.8116	435	0.4686	-
0.8134	436	0.5685	-
0.8153	437	0.5923	-
0.8172	438	0.2288	-
0.8190	439	0.5233	-
0.8209	440	0.7775	0.3810
0.8228	441	1.1349	-
0.8246	442	0.3454	-
0.8265	443	0.3732	-
0.8284	444	0.2545	-
0.8302	445	0.6133	-
0.8321	446	0.3711	-
0.8340	447	0.2668	-
0.8358	448	0.9298	-
0.8377	449	0.5457	-
0.8396	450	0.5153	0.3762
0.8414	451	0.7944	-
0.8433	452	0.274	-
0.8451	453	0.1943	-
0.8470	454	0.865	-
0.8489	455	0.577	-
0.8507	456	0.1895	-
0.8526	457	0.284	-
0.8545	458	0.2472	-
0.8563	459	0.3254	-
0.8582	460	0.9113	0.3778
0.8601	461	0.4037	-
0.8619	462	0.2395	-
0.8638	463	0.9176	-
0.8657	464	0.1605	-
0.8675	465	0.2563	-
0.8694	466	0.403	-
0.8713	467	0.6036	-
0.8731	468	0.368	-
0.875	469	0.3447	-
0.8769	470	0.1836	0.3848
0.8787	471	0.4374	-
0.8806	472	0.1704	-
0.8825	473	0.326	-
0.8843	474	0.3527	-
0.8862	475	0.8108	-
0.8881	476	0.7219	-
0.8899	477	0.2727	-
0.8918	478	0.6034	-
0.8937	479	0.8513	-
0.8955	480	0.2772	0.3935
0.8974	481	0.4888	-
0.8993	482	0.6024	-
0.9011	483	1.1502	-
0.9030	484	0.5434	-
0.9049	485	0.2632	-
0.9067	486	0.0767	-
0.9086	487	0.5782	-
0.9104	488	0.6047	-
0.9123	489	0.7541	-
0.9142	490	0.2185	0.3965
0.9160	491	0.1558	-
0.9179	492	0.1106	-
0.9198	493	0.7286	-
0.9216	494	0.1932	-
0.9235	495	0.6639	-
0.9254	496	0.422	-
0.9272	497	0.7506	-
0.9291	498	0.1227	-
0.9310	499	0.8022	-
0.9328	500	0.2475	0.3951
0.9347	501	0.3068	-
0.9366	502	0.9188	-
0.9384	503	0.3704	-
0.9403	504	0.2393	-
0.9422	505	0.7569	-
0.9440	506	0.3823	-
0.9459	507	0.1712	-
0.9478	508	0.3331	-
0.9496	509	0.3538	-
0.9515	510	0.4431	0.3976
0.9534	511	0.422	-
0.9552	512	0.3282	-
0.9571	513	0.5834	-
0.9590	514	1.1424	-
0.9608	515	0.8699	-
0.9627	516	0.2811	-
0.9646	517	0.0964	-
0.9664	518	0.2971	-
0.9683	519	0.2435	-
0.9701	520	1.1154	0.3987
0.9720	521	0.2209	-
0.9739	522	0.1551	-
0.9757	523	0.3366	-
0.9776	524	0.5526	-
0.9795	525	0.3624	-
0.9813	526	0.3311	-
0.9832	527	0.7184	-
0.9851	528	0.893	-
0.9869	529	0.2642	-
0.9888	530	0.4994	0.3986
0.9907	531	0.6881	-
0.9925	532	0.2637	-
0.9944	533	0.6997	-
0.9963	534	0.3827	-
0.9981	535	0.4079	-
1.0	536	0.0003	-

• The bold row denotes the saved checkpoint.

Framework Versions

• Python: 3.11.11
• Sentence Transformers: 3.4.1
• Transformers: 4.51.1
• PyTorch: 2.5.1+cu124
• Accelerate: 1.3.0
• Datasets: 3.5.0
• Tokenizers: 0.21.0

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",
}

CachedMultipleNegativesRankingLoss

@misc{gao2021scaling,
    title={Scaling Deep Contrastive Learning Batch Size under Memory Limited Setup},
    author={Luyu Gao and Yunyi Zhang and Jiawei Han and Jamie Callan},
    year={2021},
    eprint={2101.06983},
    archivePrefix={arXiv},
    primaryClass={cs.LG}
}