Views
No views yet
SentenceTransformer(
(0): Transformer({'max_seq_length': 8192, 'do_lower_case': False, 'architecture': 'ModernBertModel'})
(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): Normalize()
)pip install -U sentence-transformers1from sentence_transformers import SentenceTransformer
2
3# Download from the 🤗 Hub
4model = SentenceTransformer("aaa961/modernbert-embed-base-legal-no_MRL_symmetricMNRL")
5# Run inference
6sentences = [
7 'Did the CIA agree with the relief sought by the plaintiff?',
8 'redacted versions of the documents to him. See id. Since plaintiff’s counsel wished to submit \nthe non-classified portions of the two documents to the Court, the plaintiff filed a motion on \nAugust 3, 2012 to compel the CIA to “provid[e] [plaintiff’s counsel] with redacted copies” of the \ntwo documents in question. See id. at 4. The CIA opposed the relief sought by the plaintiff, \n23',
9 'would be an example of something featuring something.”). At Oral Argument, Defendant’s \ncounsel further suggested that “predominant” means a majority. Oral Arg. Tr. at 78:16–17 (“I \nwould say predominant means at least the majority.”). \nThis Court considers the parties’ interpretations of “feature” to be reasonable, given the',
10]
11embeddings = model.encode(sentences)
12print(embeddings.shape)
13# [3, 768]
14
15# Get the similarity scores for the embeddings
16similarities = model.similarity(embeddings, embeddings)
17print(similarities)
18# tensor([[1.0000, 0.4220, 0.0226],
19# [0.4220, 1.0000, 0.0205],
20# [0.0226, 0.0205, 1.0000]])ir_evalInformationRetrievalEvaluator| Metric | Value |
|---|---|
| cosine_accuracy@1 | 0.592 |
| cosine_accuracy@3 | 0.7264 |
| cosine_accuracy@5 | 0.8068 |
| cosine_accuracy@10 | 0.8671 |
| cosine_precision@1 | 0.592 |
| cosine_precision@3 | 0.2421 |
| cosine_precision@5 | 0.1614 |
| cosine_precision@10 | 0.0867 |
| cosine_recall@1 | 0.592 |
| cosine_recall@3 | 0.7264 |
| cosine_recall@5 | 0.8068 |
| cosine_recall@10 | 0.8671 |
| cosine_ndcg@10 | 0.7241 |
| cosine_mrr@10 | 0.6789 |
| cosine_map@100 | 0.6842 |
anchor and positive| anchor | positive | |
|---|---|---|
| type | string | string |
| details |
|
|
| anchor | positive |
|---|---|
What statute exempts the numbers of personnel employed by the Agency from disclosure? | numbers of personnel employed by the Agency,” 50 U.S.C. § 403g, while the NSA Act more [object Object]broadly exempts from disclosure “the organization or any function of the [NSA],” 73 Stat. at 64. [object Object]The CIA contends that the CIA Act exempts from disclosure, inter alia, “the ‘functions’ of the [object Object]CIA,” Second Lutz Decl. ¶ 36, but the NSA Act demonstrates that when Congress intends for a |
What would be violated if a heightened requirement was imposed on the protégés according to the Defendant’s Counsel? | VCH MJAR at 29 (same); Oral Arg. Tr. at 72:5–12 (Court: “[W]ould you agree that it would not [object Object]be okay to impose a heightened requirement on the protégés?” Defendant’s Counsel: “[I]f that [object Object]were the case here, then, right, we’d be violating 125.8(e).”). [object Object] [object Object]36 [object Object] [object Object]submit an individually completed Relevant Experience Project, Defendant contends that GSA has |
Where is the direct quotation from FACA § 10(b) found? | other documents which were made available to or prepared for or by” the Commission, a direct [object Object]quotation from section 10(b) of FACA. Pl.’s Mot. Exs. at 21. EPIC agrees that its FOIA request [object Object]“exactly track[s] the language of FACA § 10(b)”—i.e., that its FOIA request is meant to be [object Object]coterminous with FACA’s parameters. Pl.’s Mem. at 24; Pl.’s Reply at 9. [object Object]25 |
CachedMultipleNegativesSymmetricRankingLoss with these parameters:
1{
2 "scale": 20.0,
3 "similarity_fct": "cos_sim",
4 "mini_batch_size": 32,
5 "gather_across_devices": false,
6 "directions": [
7 "query_to_doc",
8 "doc_to_query"
9 ],
10 "partition_mode": "per_direction",
11 "hardness_mode": null,
12 "hardness_strength": 0.0
13}per_device_train_batch_size: 32num_train_epochs: 4learning_rate: 2e-05lr_scheduler_type: cosinewarmup_steps: 0.1optim: adamw_torch_fusedgradient_accumulation_steps: 16bf16: Truetf32: Trueeval_strategy: epochper_device_eval_batch_size: 16load_best_model_at_end: Trueper_device_train_batch_size: 32num_train_epochs: 4max_steps: -1learning_rate: 2e-05lr_scheduler_type: cosinelr_scheduler_kwargs: Nonewarmup_steps: 0.1optim: adamw_torch_fusedoptim_args: Noneweight_decay: 0.0adam_beta1: 0.9adam_beta2: 0.999adam_epsilon: 1e-08optim_target_modules: Nonegradient_accumulation_steps: 16average_tokens_across_devices: Truemax_grad_norm: 1.0label_smoothing_factor: 0.0bf16: Truefp16: Falsebf16_full_eval: Falsefp16_full_eval: Falsetf32: Truegradient_checkpointing: Falsegradient_checkpointing_kwargs: Nonetorch_compile: Falsetorch_compile_backend: Nonetorch_compile_mode: Noneuse_liger_kernel: Falseliger_kernel_config: Noneuse_cache: Falseneftune_noise_alpha: Nonetorch_empty_cache_steps: Noneauto_find_batch_size: Falselog_on_each_node: Truelogging_nan_inf_filter: Trueinclude_num_input_tokens_seen: nolog_level: passivelog_level_replica: warningdisable_tqdm: Falseproject: huggingfacetrackio_space_id: trackioeval_strategy: epochper_device_eval_batch_size: 16prediction_loss_only: Trueeval_on_start: Falseeval_do_concat_batches: Trueeval_use_gather_object: Falseeval_accumulation_steps: Noneinclude_for_metrics: []batch_eval_metrics: Falsesave_only_model: Falsesave_on_each_node: Falseenable_jit_checkpoint: Falsepush_to_hub: Falsehub_private_repo: Nonehub_model_id: Nonehub_strategy: every_savehub_always_push: Falsehub_revision: Noneload_best_model_at_end: Trueignore_data_skip: Falserestore_callback_states_from_checkpoint: Falsefull_determinism: Falseseed: 42data_seed: Noneuse_cpu: Falseaccelerator_config: {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}parallelism_config: Nonedataloader_drop_last: Falsedataloader_num_workers: 0dataloader_pin_memory: Truedataloader_persistent_workers: Falsedataloader_prefetch_factor: Noneremove_unused_columns: Truelabel_names: Nonetrain_sampling_strategy: randomlength_column_name: lengthddp_find_unused_parameters: Noneddp_bucket_cap_mb: Noneddp_broadcast_buffers: Falseddp_backend: Noneddp_timeout: 1800fsdp: []fsdp_config: {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}deepspeed: Nonedebug: []skip_memory_metrics: Truedo_predict: Falseresume_from_checkpoint: Nonewarmup_ratio: Nonelocal_rank: -1prompts: Nonebatch_sampler: batch_samplermulti_dataset_batch_sampler: proportionalrouter_mapping: {}learning_rate_mapping: {}| Epoch | Step | Training Loss | ir_eval_cosine_ndcg@10 |
|---|---|---|---|
| -1 | -1 | - | 0.5028 |
| 0.8791 | 10 | 0.9173 | - |
| 1.0 | 12 | - | 0.6791 |
| 1.7033 | 20 | 0.3895 | - |
| 2.0 | 24 | - | 0.7103 |
| 2.5275 | 30 | 0.2989 | - |
| 3.0 | 36 | - | 0.7209 |
| 3.3516 | 40 | 0.2743 | - |
| 4.0 | 48 | - | 0.7241 |
1@inproceedings{reimers-2019-sentence-bert,
2 title = "Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks",
3 author = "Reimers, Nils and Gurevych, Iryna",
4 booktitle = "Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing",
5 month = "11",
6 year = "2019",
7 publisher = "Association for Computational Linguistics",
8 url = "https://arxiv.org/abs/1908.10084",
9}1@misc{gao2021scaling,
2 title={Scaling Deep Contrastive Learning Batch Size under Memory Limited Setup},
3 author={Luyu Gao and Yunyi Zhang and Jiawei Han and Jamie Callan},
4 year={2021},
5 eprint={2101.06983},
6 archivePrefix={arXiv},
7 primaryClass={cs.LG}
8}