metadata

base_model: BAAI/bge-large-en-v1.5
datasets: []
language:
  - en
library_name: sentence-transformers
license: other
metrics:
  - cosine_accuracy@1
  - cosine_accuracy@3
  - cosine_accuracy@5
  - cosine_accuracy@10
  - cosine_precision@1
  - cosine_precision@3
  - cosine_precision@5
  - cosine_precision@10
  - cosine_recall@1
  - cosine_recall@3
  - cosine_recall@5
  - cosine_recall@10
  - cosine_ndcg@10
  - cosine_mrr@10
  - cosine_map@100
pipeline_tag: sentence-similarity
tags:
  - sentence-transformers
  - sentence-similarity
  - feature-extraction
  - generated_from_trainer
  - dataset_size:104022
  - loss:MatryoshkaLoss
  - loss:MultipleNegativesRankingLoss
widget:
  - source_sentence: >-
      IZEA's market capitalization is $36 million, indicating potential for
      raising additional funds if needed.
    sentences:
      - >-
        IZEA's market capitalization is $35.65 million, with a P/E ratio of
        -5.19, indicating unprofitability in the last twelve months as of Q3
        2023.
      - >-
        NetApp sells its products and services through a direct sales force and
        an ecosystem of partners.
      - >-
        SAIL's expansion plans have raised concerns among investors, leading to
        underperformance in its stock compared to the Nifty 500 index.
  - source_sentence: >-
      Infinity Mining conducted an eight-hole reverse-circulation (RC) drilling
      campaign at its Tambourah South project in Western Australia, targeting
      lithium-caesium-tantalum (LCT) pegmatites.
    sentences:
      - >-
        The disclosure must be made to a Regulatory Information Service, as
        required by Rule 8 of the Takeover Code.
      - >-
        Infinity Mining plans to expand its exploration efforts at Tambourah
        South, including the use of new technologies and techniques to identify
        and evaluate concealed pegmatite targets.
      - >-
        Russia aims to export over 65 million tons of grain during the season, a
        record volume.
  - source_sentence: >-
      Ukraine expects to receive about $1.5 billion from other international
      financial institutions, including the World Bank, in 2024.
    sentences:
      - >-
        Ukraine has an ongoing cooperation with the International Monetary Fund
        (IMF), with a 48-month lending program worth $15.6 billion, receiving
        $3.6 billion this year and expecting $900 million in December, and $5.4
        billion in 2024 subject to reform targets and economic indicators.
      - >-
        Vodacom Group could be considered a reasonable income stock despite the
        dividend cut, with a solid payout ratio but a less impressive dividend
        track record.
      - >-
        CoStar Group employees, members of the Black Excellence Network and
        Women's Network, worked alongside Feed More volunteers to facilitate the
        giveaway.
  - source_sentence: >-
      WaFd paid out 27% of its profit in dividends last year, indicating a
      comfortable payout ratio.
    sentences:
      - >-
        USP35 knockdown in Hep3B cells inhibits tumor growth and reduces the
        expression of ABHD17C, p-PI3K, and p-AKT in xenograft HCC models.
      - >-
        Nasdaq will suspend trading of CohBar, Inc.'s common stock at the
        opening of business on November 29, 2023, unless the company requests a
        hearing before a Nasdaq Hearings Panel to appeal the determination.
      - >-
        WaFd's earnings per share have grown at a rate of 9.4% per annum over
        the past five years, demonstrating consistent growth.
  - source_sentence: >-
      Scope Control provides a digital ledger of inspected lines, creating a
      credible line history that underscores Custom Truck One Source's
      commitment to operational safety.
    sentences:
      - >-
        China has implemented measures to address hidden debt, including
        extending debt maturities, selling assets to repay debts, and replacing
        short-term local government financial vehicle debts with longer-term,
        lower-cost refinancing bonds.
      - >-
        Scope Control utilizes advanced Computer Vision and Deep Learning
        technologies to accurately assess line health and categorize it as new,
        used, or bad based on safety standards and residual break strength.
      - >-
        The current management regulations for the national social security fund
        were approved in December 2001 and have been implemented for over 20
        years. The MOF stated that parts of the content no longer address the
        current needs of the Chinese financial market and the investment trend
        for the national social security fund, necessitating a systematic and
        thorough revision.
model-index:
  - name: VANTIGE_NEWS_v3_EDGE_DETECTION
    results:
      - task:
          type: information-retrieval
          name: Information Retrieval
        dataset:
          name: dim 1024
          type: dim_1024
        metrics:
          - type: cosine_accuracy@1
            value: 0.828
            name: Cosine Accuracy@1
          - type: cosine_accuracy@3
            value: 0.978
            name: Cosine Accuracy@3
          - type: cosine_accuracy@5
            value: 0.986
            name: Cosine Accuracy@5
          - type: cosine_accuracy@10
            value: 0.992
            name: Cosine Accuracy@10
          - type: cosine_precision@1
            value: 0.828
            name: Cosine Precision@1
          - type: cosine_precision@3
            value: 0.32599999999999996
            name: Cosine Precision@3
          - type: cosine_precision@5
            value: 0.19720000000000001
            name: Cosine Precision@5
          - type: cosine_precision@10
            value: 0.0992
            name: Cosine Precision@10
          - type: cosine_recall@1
            value: 0.828
            name: Cosine Recall@1
          - type: cosine_recall@3
            value: 0.978
            name: Cosine Recall@3
          - type: cosine_recall@5
            value: 0.986
            name: Cosine Recall@5
          - type: cosine_recall@10
            value: 0.992
            name: Cosine Recall@10
          - type: cosine_ndcg@10
            value: 0.9261911001883877
            name: Cosine Ndcg@10
          - type: cosine_mrr@10
            value: 0.9034555555555557
            name: Cosine Mrr@10
          - type: cosine_map@100
            value: 0.9038902618135377
            name: Cosine Map@100
      - task:
          type: information-retrieval
          name: Information Retrieval
        dataset:
          name: dim 768
          type: dim_768
        metrics:
          - type: cosine_accuracy@1
            value: 0.83
            name: Cosine Accuracy@1
          - type: cosine_accuracy@3
            value: 0.978
            name: Cosine Accuracy@3
          - type: cosine_accuracy@5
            value: 0.986
            name: Cosine Accuracy@5
          - type: cosine_accuracy@10
            value: 0.99
            name: Cosine Accuracy@10
          - type: cosine_precision@1
            value: 0.83
            name: Cosine Precision@1
          - type: cosine_precision@3
            value: 0.32599999999999996
            name: Cosine Precision@3
          - type: cosine_precision@5
            value: 0.1972
            name: Cosine Precision@5
          - type: cosine_precision@10
            value: 0.099
            name: Cosine Precision@10
          - type: cosine_recall@1
            value: 0.83
            name: Cosine Recall@1
          - type: cosine_recall@3
            value: 0.978
            name: Cosine Recall@3
          - type: cosine_recall@5
            value: 0.986
            name: Cosine Recall@5
          - type: cosine_recall@10
            value: 0.99
            name: Cosine Recall@10
          - type: cosine_ndcg@10
            value: 0.9264556449878328
            name: Cosine Ndcg@10
          - type: cosine_mrr@10
            value: 0.9044190476190478
            name: Cosine Mrr@10
          - type: cosine_map@100
            value: 0.9049635033323674
            name: Cosine Map@100
      - task:
          type: information-retrieval
          name: Information Retrieval
        dataset:
          name: dim 512
          type: dim_512
        metrics:
          - type: cosine_accuracy@1
            value: 0.83
            name: Cosine Accuracy@1
          - type: cosine_accuracy@3
            value: 0.978
            name: Cosine Accuracy@3
          - type: cosine_accuracy@5
            value: 0.988
            name: Cosine Accuracy@5
          - type: cosine_accuracy@10
            value: 0.99
            name: Cosine Accuracy@10
          - type: cosine_precision@1
            value: 0.83
            name: Cosine Precision@1
          - type: cosine_precision@3
            value: 0.32599999999999996
            name: Cosine Precision@3
          - type: cosine_precision@5
            value: 0.19760000000000003
            name: Cosine Precision@5
          - type: cosine_precision@10
            value: 0.099
            name: Cosine Precision@10
          - type: cosine_recall@1
            value: 0.83
            name: Cosine Recall@1
          - type: cosine_recall@3
            value: 0.978
            name: Cosine Recall@3
          - type: cosine_recall@5
            value: 0.988
            name: Cosine Recall@5
          - type: cosine_recall@10
            value: 0.99
            name: Cosine Recall@10
          - type: cosine_ndcg@10
            value: 0.9262131769268145
            name: Cosine Ndcg@10
          - type: cosine_mrr@10
            value: 0.9041
            name: Cosine Mrr@10
          - type: cosine_map@100
            value: 0.9046338347982871
            name: Cosine Map@100
      - task:
          type: information-retrieval
          name: Information Retrieval
        dataset:
          name: dim 256
          type: dim_256
        metrics:
          - type: cosine_accuracy@1
            value: 0.828
            name: Cosine Accuracy@1
          - type: cosine_accuracy@3
            value: 0.978
            name: Cosine Accuracy@3
          - type: cosine_accuracy@5
            value: 0.984
            name: Cosine Accuracy@5
          - type: cosine_accuracy@10
            value: 0.99
            name: Cosine Accuracy@10
          - type: cosine_precision@1
            value: 0.828
            name: Cosine Precision@1
          - type: cosine_precision@3
            value: 0.32599999999999996
            name: Cosine Precision@3
          - type: cosine_precision@5
            value: 0.1968
            name: Cosine Precision@5
          - type: cosine_precision@10
            value: 0.099
            name: Cosine Precision@10
          - type: cosine_recall@1
            value: 0.828
            name: Cosine Recall@1
          - type: cosine_recall@3
            value: 0.978
            name: Cosine Recall@3
          - type: cosine_recall@5
            value: 0.984
            name: Cosine Recall@5
          - type: cosine_recall@10
            value: 0.99
            name: Cosine Recall@10
          - type: cosine_ndcg@10
            value: 0.9250967573273415
            name: Cosine Ndcg@10
          - type: cosine_mrr@10
            value: 0.90265
            name: Cosine Mrr@10
          - type: cosine_map@100
            value: 0.9031974089635855
            name: Cosine Map@100
      - task:
          type: information-retrieval
          name: Information Retrieval
        dataset:
          name: dim 128
          type: dim_128
        metrics:
          - type: cosine_accuracy@1
            value: 0.832
            name: Cosine Accuracy@1
          - type: cosine_accuracy@3
            value: 0.978
            name: Cosine Accuracy@3
          - type: cosine_accuracy@5
            value: 0.986
            name: Cosine Accuracy@5
          - type: cosine_accuracy@10
            value: 0.992
            name: Cosine Accuracy@10
          - type: cosine_precision@1
            value: 0.832
            name: Cosine Precision@1
          - type: cosine_precision@3
            value: 0.32599999999999996
            name: Cosine Precision@3
          - type: cosine_precision@5
            value: 0.19720000000000001
            name: Cosine Precision@5
          - type: cosine_precision@10
            value: 0.0992
            name: Cosine Precision@10
          - type: cosine_recall@1
            value: 0.832
            name: Cosine Recall@1
          - type: cosine_recall@3
            value: 0.978
            name: Cosine Recall@3
          - type: cosine_recall@5
            value: 0.986
            name: Cosine Recall@5
          - type: cosine_recall@10
            value: 0.992
            name: Cosine Recall@10
          - type: cosine_ndcg@10
            value: 0.9276434508354098
            name: Cosine Ndcg@10
          - type: cosine_mrr@10
            value: 0.9054333333333333
            name: Cosine Mrr@10
          - type: cosine_map@100
            value: 0.9058527890466532
            name: Cosine Map@100
      - task:
          type: information-retrieval
          name: Information Retrieval
        dataset:
          name: dim 64
          type: dim_64
        metrics:
          - type: cosine_accuracy@1
            value: 0.822
            name: Cosine Accuracy@1
          - type: cosine_accuracy@3
            value: 0.978
            name: Cosine Accuracy@3
          - type: cosine_accuracy@5
            value: 0.986
            name: Cosine Accuracy@5
          - type: cosine_accuracy@10
            value: 0.99
            name: Cosine Accuracy@10
          - type: cosine_precision@1
            value: 0.822
            name: Cosine Precision@1
          - type: cosine_precision@3
            value: 0.32599999999999996
            name: Cosine Precision@3
          - type: cosine_precision@5
            value: 0.19720000000000001
            name: Cosine Precision@5
          - type: cosine_precision@10
            value: 0.099
            name: Cosine Precision@10
          - type: cosine_recall@1
            value: 0.822
            name: Cosine Recall@1
          - type: cosine_recall@3
            value: 0.978
            name: Cosine Recall@3
          - type: cosine_recall@5
            value: 0.986
            name: Cosine Recall@5
          - type: cosine_recall@10
            value: 0.99
            name: Cosine Recall@10
          - type: cosine_ndcg@10
            value: 0.9224148281915946
            name: Cosine Ndcg@10
          - type: cosine_mrr@10
            value: 0.8989999999999999
            name: Cosine Mrr@10
          - type: cosine_map@100
            value: 0.8995256769374417
            name: Cosine Map@100

VANTIGE_NEWS_v3_EDGE_DETECTION

This is a sentence-transformers model finetuned from BAAI/bge-large-en-v1.5. 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: BAAI/bge-large-en-v1.5
Maximum Sequence Length: 512 tokens
Output Dimensionality: 1024 tokens
Similarity Function: Cosine Similarity
Language: en
License: other

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': True}) with Transformer model: BertModel 
  (1): Pooling({'word_embedding_dimension': 1024, 'pooling_mode_cls_token': True, 'pooling_mode_mean_tokens': False, '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()
)

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("dustyatx/news_v3_graph_edges_embeddings_setence_paragraph")
# Run inference
sentences = [
    "Scope Control provides a digital ledger of inspected lines, creating a credible line history that underscores Custom Truck One Source's commitment to operational safety.",
    'Scope Control utilizes advanced Computer Vision and Deep Learning technologies to accurately assess line health and categorize it as new, used, or bad based on safety standards and residual break strength.',
    'China has implemented measures to address hidden debt, including extending debt maturities, selling assets to repay debts, and replacing short-term local government financial vehicle debts with longer-term, lower-cost refinancing bonds.',
]
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: dim_1024
Evaluated with InformationRetrievalEvaluator

Metric	Value
cosine_accuracy@1	0.828
cosine_accuracy@3	0.978
cosine_accuracy@5	0.986
cosine_accuracy@10	0.992
cosine_precision@1	0.828
cosine_precision@3	0.326
cosine_precision@5	0.1972
cosine_precision@10	0.0992
cosine_recall@1	0.828
cosine_recall@3	0.978
cosine_recall@5	0.986
cosine_recall@10	0.992
cosine_ndcg@10	0.9262
cosine_mrr@10	0.9035
cosine_map@100	0.9039

Information Retrieval

Dataset: dim_768
Evaluated with InformationRetrievalEvaluator

Metric	Value
cosine_accuracy@1	0.83
cosine_accuracy@3	0.978
cosine_accuracy@5	0.986
cosine_accuracy@10	0.99
cosine_precision@1	0.83
cosine_precision@3	0.326
cosine_precision@5	0.1972
cosine_precision@10	0.099
cosine_recall@1	0.83
cosine_recall@3	0.978
cosine_recall@5	0.986
cosine_recall@10	0.99
cosine_ndcg@10	0.9265
cosine_mrr@10	0.9044
cosine_map@100	0.905

Information Retrieval

Dataset: dim_512
Evaluated with InformationRetrievalEvaluator

Metric	Value
cosine_accuracy@1	0.83
cosine_accuracy@3	0.978
cosine_accuracy@5	0.988
cosine_accuracy@10	0.99
cosine_precision@1	0.83
cosine_precision@3	0.326
cosine_precision@5	0.1976
cosine_precision@10	0.099
cosine_recall@1	0.83
cosine_recall@3	0.978
cosine_recall@5	0.988
cosine_recall@10	0.99
cosine_ndcg@10	0.9262
cosine_mrr@10	0.9041
cosine_map@100	0.9046

Information Retrieval

Dataset: dim_256
Evaluated with InformationRetrievalEvaluator

Metric	Value
cosine_accuracy@1	0.828
cosine_accuracy@3	0.978
cosine_accuracy@5	0.984
cosine_accuracy@10	0.99
cosine_precision@1	0.828
cosine_precision@3	0.326
cosine_precision@5	0.1968
cosine_precision@10	0.099
cosine_recall@1	0.828
cosine_recall@3	0.978
cosine_recall@5	0.984
cosine_recall@10	0.99
cosine_ndcg@10	0.9251
cosine_mrr@10	0.9026
cosine_map@100	0.9032

Information Retrieval

Dataset: dim_128
Evaluated with InformationRetrievalEvaluator

Metric	Value
cosine_accuracy@1	0.832
cosine_accuracy@3	0.978
cosine_accuracy@5	0.986
cosine_accuracy@10	0.992
cosine_precision@1	0.832
cosine_precision@3	0.326
cosine_precision@5	0.1972
cosine_precision@10	0.0992
cosine_recall@1	0.832
cosine_recall@3	0.978
cosine_recall@5	0.986
cosine_recall@10	0.992
cosine_ndcg@10	0.9276
cosine_mrr@10	0.9054
cosine_map@100	0.9059

Information Retrieval

Dataset: dim_64
Evaluated with InformationRetrievalEvaluator

Metric	Value
cosine_accuracy@1	0.822
cosine_accuracy@3	0.978
cosine_accuracy@5	0.986
cosine_accuracy@10	0.99
cosine_precision@1	0.822
cosine_precision@3	0.326
cosine_precision@5	0.1972
cosine_precision@10	0.099
cosine_recall@1	0.822
cosine_recall@3	0.978
cosine_recall@5	0.986
cosine_recall@10	0.99
cosine_ndcg@10	0.9224
cosine_mrr@10	0.899
cosine_map@100	0.8995

Training Details

Training Dataset

Unnamed Dataset

Size: 104,022 training samples
Columns: anchor and positive
Approximate statistics based on the first 1000 samples:
anchor positive
type string string
details
min: 16 tokens
mean: 36.53 tokens
max: 102 tokens

min: 13 tokens
mean: 35.17 tokens
max: 117 tokens

	anchor	positive
type	string	string
details	min: 16 tokens mean: 36.53 tokens max: 102 tokens	min: 13 tokens mean: 35.17 tokens max: 117 tokens

Samples:

anchor	positive
`The general public, including retail investors, collectively own 11% of FINEOS Corporation Holdings' shares, representing a minority stake in the company.`	`Private companies, with their 50% ownership stake, have substantial influence over FINEOS Corporation Holdings' management and governance decisions.`
`A study by the Insurance Institute for Highway Safety (IIHS) found that SUVs and vans with hood heights exceeding 40 inches are approximately 45% more likely to cause pedestrian fatalities compared to vehicles with hood heights of 30 inches or less and a sloping profile.`	`Vehicles with front ends exceeding 35 inches in height, particularly those lacking a sloping profile, are more likely to cause severe head, torso, and hip injuries to pedestrians.`
`SpringWorks Therapeutics has a portfolio of small molecule targeted oncology product candidates and is conducting clinical trials for rare tumor types and genetically defined cancers.`	`SpringWorks Therapeutics operates in the biopharmaceutical industry, specializing in precision medicine for underserved patient populations.`

Loss: MatryoshkaLoss with these parameters:

{
    "loss": "MultipleNegativesRankingLoss",
    "matryoshka_dims": [
        1024,
        768,
        512,
        256,
        128,
        64
    ],
    "matryoshka_weights": [
        1,
        1,
        1,
        1,
        1,
        1
    ],
    "n_dims_per_step": -1
}

Training Hyperparameters

Non-Default Hyperparameters

eval_strategy: steps
per_device_train_batch_size: 30
per_device_eval_batch_size: 20
gradient_accumulation_steps: 8
learning_rate: 3e-05
num_train_epochs: 2
lr_scheduler_type: cosine
warmup_ratio: 0.2
bf16: True
tf32: True
dataloader_num_workers: 30
load_best_model_at_end: True
optim: adamw_torch_fused
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: 30
per_device_eval_batch_size: 20
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: 3e-05
weight_decay: 0.0
adam_beta1: 0.9
adam_beta2: 0.999
adam_epsilon: 1e-08
max_grad_norm: 1.0
num_train_epochs: 2
max_steps: -1
lr_scheduler_type: cosine
lr_scheduler_kwargs: {}
warmup_ratio: 0.2
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
use_ipex: False
bf16: True
fp16: False
fp16_opt_level: O1
half_precision_backend: auto
bf16_full_eval: False
fp16_full_eval: False
tf32: True
local_rank: 0
ddp_backend: None
tpu_num_cores: None
tpu_metrics_debug: False
debug: []
dataloader_drop_last: False
dataloader_num_workers: 30
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}
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_fused
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: False
hub_always_push: False
gradient_checkpointing: False
gradient_checkpointing_kwargs: None
include_inputs_for_metrics: False
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
dispatch_batches: None
split_batches: 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
eval_use_gather_object: False
batch_sampler: no_duplicates
multi_dataset_batch_sampler: proportional

Training Logs

Click to expand

Epoch	Step	Training Loss	dim_1024_cosine_map@100	dim_128_cosine_map@100	dim_256_cosine_map@100	dim_512_cosine_map@100	dim_64_cosine_map@100	dim_768_cosine_map@100
0.0023	1	1.8313	-	-	-	-	-	-
0.0046	2	1.9678	-	-	-	-	-	-
0.0069	3	0.8038	-	-	-	-	-	-
0.0092	4	0.7993	-	-	-	-	-	-
0.0115	5	0.7926	-	-	-	-	-	-
0.0138	6	0.9348	-	-	-	-	-	-
0.0161	7	0.8707	-	-	-	-	-	-
0.0185	8	0.7293	-	-	-	-	-	-
0.0208	9	0.6618	-	-	-	-	-	-
0.0231	10	0.846	-	-	-	-	-	-
0.0254	11	0.6836	-	-	-	-	-	-
0.0277	12	0.7034	-	-	-	-	-	-
0.0300	13	0.7987	-	-	-	-	-	-
0.0323	14	0.6443	-	-	-	-	-	-
0.0346	15	0.5975	-	-	-	-	-	-
0.0369	16	0.4471	-	-	-	-	-	-
0.0392	17	0.4739	-	-	-	-	-	-
0.0415	18	0.4136	-	-	-	-	-	-
0.0438	19	0.3865	-	-	-	-	-	-
0.0461	20	0.3421	-	-	-	-	-	-
0.0484	21	0.5076	-	-	-	-	-	-
0.0507	22	0.1878	-	-	-	-	-	-
0.0531	23	0.3597	-	-	-	-	-	-
0.0554	24	0.23	-	-	-	-	-	-
0.0577	25	0.1331	-	-	-	-	-	-
0.0600	26	0.1793	-	-	-	-	-	-
0.0623	27	0.1309	-	-	-	-	-	-
0.0646	28	0.1077	-	-	-	-	-	-
0.0669	29	0.1681	-	-	-	-	-	-
0.0692	30	0.055	-	-	-	-	-	-
0.0715	31	0.1062	-	-	-	-	-	-
0.0738	32	0.0672	-	-	-	-	-	-
0.0761	33	0.067	-	-	-	-	-	-
0.0784	34	0.0953	-	-	-	-	-	-
0.0807	35	0.0602	-	-	-	-	-	-
0.0830	36	0.1312	-	-	-	-	-	-
0.0854	37	0.0356	-	-	-	-	-	-
0.0877	38	0.0707	-	-	-	-	-	-
0.0900	39	0.1525	-	-	-	-	-	-
0.0923	40	0.0362	-	-	-	-	-	-
0.0946	41	0.253	-	-	-	-	-	-
0.0969	42	0.0572	-	-	-	-	-	-
0.0992	43	0.1031	-	-	-	-	-	-
0.1015	44	0.1023	-	-	-	-	-	-
0.1038	45	0.052	-	-	-	-	-	-
0.1061	46	0.0614	-	-	-	-	-	-
0.1084	47	0.1256	-	-	-	-	-	-
0.1107	48	0.1624	-	-	-	-	-	-
0.1130	49	0.0363	-	-	-	-	-	-
0.1153	50	0.2001	0.8949	0.8940	0.8947	0.8950	0.8864	0.8972
0.1176	51	0.0846	-	-	-	-	-	-
0.1200	52	0.0338	-	-	-	-	-	-
0.1223	53	0.0648	-	-	-	-	-	-
0.1246	54	0.1232	-	-	-	-	-	-
0.1269	55	0.0318	-	-	-	-	-	-
0.1292	56	0.1148	-	-	-	-	-	-
0.1315	57	0.0826	-	-	-	-	-	-
0.1338	58	0.034	-	-	-	-	-	-
0.1361	59	0.0492	-	-	-	-	-	-
0.1384	60	0.0427	-	-	-	-	-	-
0.1407	61	0.0709	-	-	-	-	-	-
0.1430	62	0.0494	-	-	-	-	-	-
0.1453	63	0.0554	-	-	-	-	-	-
0.1476	64	0.061	-	-	-	-	-	-
0.1499	65	0.1155	-	-	-	-	-	-
0.1522	66	0.0419	-	-	-	-	-	-
0.1546	67	0.0185	-	-	-	-	-	-
0.1569	68	0.0559	-	-	-	-	-	-
0.1592	69	0.0219	-	-	-	-	-	-
0.1615	70	0.0302	-	-	-	-	-	-
0.1638	71	0.0322	-	-	-	-	-	-
0.1661	72	0.0604	-	-	-	-	-	-
0.1684	73	0.038	-	-	-	-	-	-
0.1707	74	0.0971	-	-	-	-	-	-
0.1730	75	0.0384	-	-	-	-	-	-
0.1753	76	0.0887	-	-	-	-	-	-
0.1776	77	0.0495	-	-	-	-	-	-
0.1799	78	0.0203	-	-	-	-	-	-
0.1822	79	0.0669	-	-	-	-	-	-
0.1845	80	0.0319	-	-	-	-	-	-
0.1869	81	0.0177	-	-	-	-	-	-
0.1892	82	0.0303	-	-	-	-	-	-
0.1915	83	0.037	-	-	-	-	-	-
0.1938	84	0.0122	-	-	-	-	-	-
0.1961	85	0.0377	-	-	-	-	-	-
0.1984	86	0.0578	-	-	-	-	-	-
0.2007	87	0.0347	-	-	-	-	-	-
0.2030	88	0.1288	-	-	-	-	-	-
0.2053	89	0.0964	-	-	-	-	-	-
0.2076	90	0.0172	-	-	-	-	-	-
0.2099	91	0.0726	-	-	-	-	-	-
0.2122	92	0.0225	-	-	-	-	-	-
0.2145	93	0.1011	-	-	-	-	-	-
0.2168	94	0.0248	-	-	-	-	-	-
0.2191	95	0.0431	-	-	-	-	-	-
0.2215	96	0.0243	-	-	-	-	-	-
0.2238	97	0.0221	-	-	-	-	-	-
0.2261	98	0.0529	-	-	-	-	-	-
0.2284	99	0.0459	-	-	-	-	-	-
0.2307	100	0.0869	0.9026	0.8967	0.8950	0.9003	0.8915	0.9009
0.2330	101	0.0685	-	-	-	-	-	-
0.2353	102	0.0801	-	-	-	-	-	-
0.2376	103	0.025	-	-	-	-	-	-
0.2399	104	0.0556	-	-	-	-	-	-
0.2422	105	0.0146	-	-	-	-	-	-
0.2445	106	0.0335	-	-	-	-	-	-
0.2468	107	0.0441	-	-	-	-	-	-
0.2491	108	0.0187	-	-	-	-	-	-
0.2514	109	0.1027	-	-	-	-	-	-
0.2537	110	0.0189	-	-	-	-	-	-
0.2561	111	0.1262	-	-	-	-	-	-
0.2584	112	0.1193	-	-	-	-	-	-
0.2607	113	0.0285	-	-	-	-	-	-
0.2630	114	0.0226	-	-	-	-	-	-
0.2653	115	0.1209	-	-	-	-	-	-
0.2676	116	0.0765	-	-	-	-	-	-
0.2699	117	0.1405	-	-	-	-	-	-
0.2722	118	0.0629	-	-	-	-	-	-
0.2745	119	0.0413	-	-	-	-	-	-
0.2768	120	0.0572	-	-	-	-	-	-
0.2791	121	0.0192	-	-	-	-	-	-
0.2814	122	0.0949	-	-	-	-	-	-
0.2837	123	0.0398	-	-	-	-	-	-
0.2860	124	0.0596	-	-	-	-	-	-
0.2884	125	0.0243	-	-	-	-	-	-
0.2907	126	0.0636	-	-	-	-	-	-
0.2930	127	0.0367	-	-	-	-	-	-
0.2953	128	0.0542	-	-	-	-	-	-
0.2976	129	0.0149	-	-	-	-	-	-
0.2999	130	0.097	-	-	-	-	-	-
0.3022	131	0.0213	-	-	-	-	-	-
0.3045	132	0.027	-	-	-	-	-	-
0.3068	133	0.0577	-	-	-	-	-	-
0.3091	134	0.0143	-	-	-	-	-	-
0.3114	135	0.0285	-	-	-	-	-	-
0.3137	136	0.033	-	-	-	-	-	-
0.3160	137	0.0412	-	-	-	-	-	-
0.3183	138	0.0125	-	-	-	-	-	-
0.3206	139	0.0512	-	-	-	-	-	-
0.3230	140	0.0189	-	-	-	-	-	-
0.3253	141	0.124	-	-	-	-	-	-
0.3276	142	0.0118	-	-	-	-	-	-
0.3299	143	0.017	-	-	-	-	-	-
0.3322	144	0.025	-	-	-	-	-	-
0.3345	145	0.0187	-	-	-	-	-	-
0.3368	146	0.0141	-	-	-	-	-	-
0.3391	147	0.0325	-	-	-	-	-	-
0.3414	148	0.0582	-	-	-	-	-	-
0.3437	149	0.0611	-	-	-	-	-	-
0.3460	150	0.0261	0.9047	0.8995	0.9003	0.9022	0.8998	0.9032
0.3483	151	0.014	-	-	-	-	-	-
0.3506	152	0.0077	-	-	-	-	-	-
0.3529	153	0.022	-	-	-	-	-	-
0.3552	154	0.0328	-	-	-	-	-	-
0.3576	155	0.0124	-	-	-	-	-	-
0.3599	156	0.0103	-	-	-	-	-	-
0.3622	157	0.0607	-	-	-	-	-	-
0.3645	158	0.0121	-	-	-	-	-	-
0.3668	159	0.0761	-	-	-	-	-	-
0.3691	160	0.0981	-	-	-	-	-	-
0.3714	161	0.1071	-	-	-	-	-	-
0.3737	162	0.1307	-	-	-	-	-	-
0.3760	163	0.0524	-	-	-	-	-	-
0.3783	164	0.0726	-	-	-	-	-	-
0.3806	165	0.0636	-	-	-	-	-	-
0.3829	166	0.0428	-	-	-	-	-	-
0.3852	167	0.0111	-	-	-	-	-	-
0.3875	168	0.0542	-	-	-	-	-	-
0.3899	169	0.0193	-	-	-	-	-	-
0.3922	170	0.0095	-	-	-	-	-	-
0.3945	171	0.0464	-	-	-	-	-	-
0.3968	172	0.0167	-	-	-	-	-	-
0.3991	173	0.0209	-	-	-	-	-	-
0.4014	174	0.0359	-	-	-	-	-	-
0.4037	175	0.071	-	-	-	-	-	-
0.4060	176	0.0189	-	-	-	-	-	-
0.4083	177	0.0448	-	-	-	-	-	-
0.4106	178	0.0161	-	-	-	-	-	-
0.4129	179	0.0427	-	-	-	-	-	-
0.4152	180	0.0229	-	-	-	-	-	-
0.4175	181	0.0274	-	-	-	-	-	-
0.4198	182	0.0173	-	-	-	-	-	-
0.4221	183	0.0123	-	-	-	-	-	-
0.4245	184	0.0395	-	-	-	-	-	-
0.4268	185	0.015	-	-	-	-	-	-
0.4291	186	0.0168	-	-	-	-	-	-
0.4314	187	0.0165	-	-	-	-	-	-
0.4337	188	0.0412	-	-	-	-	-	-
0.4360	189	0.0961	-	-	-	-	-	-
0.4383	190	0.0551	-	-	-	-	-	-
0.4406	191	0.0685	-	-	-	-	-	-
0.4429	192	0.1561	-	-	-	-	-	-
0.4452	193	0.0333	-	-	-	-	-	-
0.4475	194	0.0567	-	-	-	-	-	-
0.4498	195	0.0081	-	-	-	-	-	-
0.4521	196	0.0297	-	-	-	-	-	-
0.4544	197	0.0131	-	-	-	-	-	-
0.4567	198	0.0322	-	-	-	-	-	-
0.4591	199	0.0224	-	-	-	-	-	-
0.4614	200	0.0068	0.8989	0.8941	0.8983	0.8985	0.8975	0.9002
0.4637	201	0.0115	-	-	-	-	-	-
0.4660	202	0.0098	-	-	-	-	-	-
0.4683	203	0.101	-	-	-	-	-	-
0.4706	204	0.0282	-	-	-	-	-	-
0.4729	205	0.0721	-	-	-	-	-	-
0.4752	206	0.0123	-	-	-	-	-	-
0.4775	207	0.1014	-	-	-	-	-	-
0.4798	208	0.0257	-	-	-	-	-	-
0.4821	209	0.1126	-	-	-	-	-	-
0.4844	210	0.0586	-	-	-	-	-	-
0.4867	211	0.0307	-	-	-	-	-	-
0.4890	212	0.0226	-	-	-	-	-	-
0.4913	213	0.0471	-	-	-	-	-	-
0.4937	214	0.025	-	-	-	-	-	-
0.4960	215	0.0799	-	-	-	-	-	-
0.4983	216	0.0173	-	-	-	-	-	-
0.5006	217	0.0208	-	-	-	-	-	-
0.5029	218	0.0461	-	-	-	-	-	-
0.5052	219	0.0592	-	-	-	-	-	-
0.5075	220	0.0076	-	-	-	-	-	-
0.5098	221	0.0156	-	-	-	-	-	-
0.5121	222	0.0149	-	-	-	-	-	-
0.5144	223	0.0138	-	-	-	-	-	-
0.5167	224	0.0526	-	-	-	-	-	-
0.5190	225	0.0689	-	-	-	-	-	-
0.5213	226	0.0191	-	-	-	-	-	-
0.5236	227	0.0094	-	-	-	-	-	-
0.5260	228	0.0125	-	-	-	-	-	-
0.5283	229	0.0632	-	-	-	-	-	-
0.5306	230	0.0773	-	-	-	-	-	-
0.5329	231	0.0147	-	-	-	-	-	-
0.5352	232	0.0145	-	-	-	-	-	-
0.5375	233	0.0068	-	-	-	-	-	-
0.5398	234	0.0673	-	-	-	-	-	-
0.5421	235	0.0131	-	-	-	-	-	-
0.5444	236	0.0217	-	-	-	-	-	-
0.5467	237	0.0126	-	-	-	-	-	-
0.5490	238	0.0172	-	-	-	-	-	-
0.5513	239	0.0122	-	-	-	-	-	-
0.5536	240	0.0175	-	-	-	-	-	-
0.5559	241	0.0184	-	-	-	-	-	-
0.5582	242	0.0422	-	-	-	-	-	-
0.5606	243	0.0106	-	-	-	-	-	-
0.5629	244	0.071	-	-	-	-	-	-
0.5652	245	0.0089	-	-	-	-	-	-
0.5675	246	0.0099	-	-	-	-	-	-
0.5698	247	0.0133	-	-	-	-	-	-
0.5721	248	0.0627	-	-	-	-	-	-
0.5744	249	0.0248	-	-	-	-	-	-
0.5767	250	0.0349	0.8970	0.8968	0.8961	0.8961	0.8952	0.8963
0.5790	251	0.0145	-	-	-	-	-	-
0.5813	252	0.0052	-	-	-	-	-	-
0.5836	253	0.0198	-	-	-	-	-	-
0.5859	254	0.0065	-	-	-	-	-	-
0.5882	255	0.007	-	-	-	-	-	-
0.5905	256	0.0072	-	-	-	-	-	-
0.5928	257	0.1878	-	-	-	-	-	-
0.5952	258	0.0091	-	-	-	-	-	-
0.5975	259	0.0421	-	-	-	-	-	-
0.5998	260	0.0166	-	-	-	-	-	-
0.6021	261	0.0909	-	-	-	-	-	-
0.6044	262	0.0107	-	-	-	-	-	-
0.6067	263	0.0191	-	-	-	-	-	-
0.6090	264	0.0168	-	-	-	-	-	-
0.6113	265	0.0814	-	-	-	-	-	-
0.6136	266	0.0736	-	-	-	-	-	-
0.6159	267	0.0297	-	-	-	-	-	-
0.6182	268	0.016	-	-	-	-	-	-
0.6205	269	0.0201	-	-	-	-	-	-
0.6228	270	0.0111	-	-	-	-	-	-
0.6251	271	0.0164	-	-	-	-	-	-
0.6275	272	0.0106	-	-	-	-	-	-
0.6298	273	0.0287	-	-	-	-	-	-
0.6321	274	0.0595	-	-	-	-	-	-
0.6344	275	0.0446	-	-	-	-	-	-
0.6367	276	0.0203	-	-	-	-	-	-
0.6390	277	0.0079	-	-	-	-	-	-
0.6413	278	0.0345	-	-	-	-	-	-
0.6436	279	0.0461	-	-	-	-	-	-
0.6459	280	0.0803	-	-	-	-	-	-
0.6482	281	0.0218	-	-	-	-	-	-
0.6505	282	0.0288	-	-	-	-	-	-
0.6528	283	0.0745	-	-	-	-	-	-
0.6551	284	0.0102	-	-	-	-	-	-
0.6574	285	0.0626	-	-	-	-	-	-
0.6597	286	0.0606	-	-	-	-	-	-
0.6621	287	0.0319	-	-	-	-	-	-
0.6644	288	0.0303	-	-	-	-	-	-
0.6667	289	0.0216	-	-	-	-	-	-
0.6690	290	0.0417	-	-	-	-	-	-
0.6713	291	0.0061	-	-	-	-	-	-
0.6736	292	0.0386	-	-	-	-	-	-
0.6759	293	0.0117	-	-	-	-	-	-
0.6782	294	0.0283	-	-	-	-	-	-
0.6805	295	0.013	-	-	-	-	-	-
0.6828	296	0.1237	-	-	-	-	-	-
0.6851	297	0.0878	-	-	-	-	-	-
0.6874	298	0.0158	-	-	-	-	-	-
0.6897	299	0.0562	-	-	-	-	-	-
0.6920	300	0.0871	0.9022	0.9027	0.9074	0.9055	0.8990	0.9027
0.6943	301	0.0657	-	-	-	-	-	-
0.6967	302	0.0239	-	-	-	-	-	-
0.6990	303	0.0053	-	-	-	-	-	-
0.7013	304	0.0237	-	-	-	-	-	-
0.7036	305	0.0182	-	-	-	-	-	-
0.7059	306	0.0135	-	-	-	-	-	-
0.7082	307	0.0059	-	-	-	-	-	-
0.7105	308	0.0061	-	-	-	-	-	-
0.7128	309	0.0072	-	-	-	-	-	-
0.7151	310	0.0319	-	-	-	-	-	-
0.7174	311	0.1183	-	-	-	-	-	-
0.7197	312	0.0447	-	-	-	-	-	-
0.7220	313	0.0369	-	-	-	-	-	-
0.7243	314	0.0462	-	-	-	-	-	-
0.7266	315	0.0233	-	-	-	-	-	-
0.7290	316	0.0114	-	-	-	-	-	-
0.7313	317	0.0179	-	-	-	-	-	-
0.7336	318	0.0203	-	-	-	-	-	-
0.7359	319	0.0071	-	-	-	-	-	-
0.7382	320	0.1297	-	-	-	-	-	-
0.7405	321	0.0249	-	-	-	-	-	-
0.7428	322	0.063	-	-	-	-	-	-
0.7451	323	0.0479	-	-	-	-	-	-
0.7474	324	0.1483	-	-	-	-	-	-
0.7497	325	0.0058	-	-	-	-	-	-
0.7520	326	0.0191	-	-	-	-	-	-
0.7543	327	0.0855	-	-	-	-	-	-
0.7566	328	0.0156	-	-	-	-	-	-
0.7589	329	0.0147	-	-	-	-	-	-
0.7612	330	0.0124	-	-	-	-	-	-
0.7636	331	0.0242	-	-	-	-	-	-
0.7659	332	0.0433	-	-	-	-	-	-
0.7682	333	0.0103	-	-	-	-	-	-
0.7705	334	0.0833	-	-	-	-	-	-
0.7728	335	0.0082	-	-	-	-	-	-
0.7751	336	0.0122	-	-	-	-	-	-
0.7774	337	0.031	-	-	-	-	-	-
0.7797	338	0.0116	-	-	-	-	-	-
0.7820	339	0.0947	-	-	-	-	-	-
0.7843	340	0.0323	-	-	-	-	-	-
0.7866	341	0.0177	-	-	-	-	-	-
0.7889	342	0.0487	-	-	-	-	-	-
0.7912	343	0.0123	-	-	-	-	-	-
0.7935	344	0.0075	-	-	-	-	-	-
0.7958	345	0.0061	-	-	-	-	-	-
0.7982	346	0.0057	-	-	-	-	-	-
0.8005	347	0.1108	-	-	-	-	-	-
0.8028	348	0.0104	-	-	-	-	-	-
0.8051	349	0.0131	-	-	-	-	-	-
0.8074	350	0.0229	0.9053	0.9041	0.9033	0.9066	0.8965	0.9052
0.8097	351	0.0478	-	-	-	-	-	-
0.8120	352	0.0127	-	-	-	-	-	-
0.8143	353	0.1143	-	-	-	-	-	-
0.8166	354	0.0365	-	-	-	-	-	-
0.8189	355	0.0418	-	-	-	-	-	-
0.8212	356	0.0494	-	-	-	-	-	-
0.8235	357	0.0082	-	-	-	-	-	-
0.8258	358	0.0212	-	-	-	-	-	-
0.8281	359	0.0106	-	-	-	-	-	-
0.8304	360	0.1009	-	-	-	-	-	-
0.8328	361	0.0316	-	-	-	-	-	-
0.8351	362	0.0313	-	-	-	-	-	-
0.8374	363	0.0108	-	-	-	-	-	-
0.8397	364	0.0198	-	-	-	-	-	-
0.8420	365	0.0112	-	-	-	-	-	-
0.8443	366	0.0197	-	-	-	-	-	-
0.8466	367	0.058	-	-	-	-	-	-
0.8489	368	0.0187	-	-	-	-	-	-
0.8512	369	0.0196	-	-	-	-	-	-
0.8535	370	0.0586	-	-	-	-	-	-
0.8558	371	0.0099	-	-	-	-	-	-
0.8581	372	0.0248	-	-	-	-	-	-
0.8604	373	0.0183	-	-	-	-	-	-
0.8627	374	0.0268	-	-	-	-	-	-
0.8651	375	0.0154	-	-	-	-	-	-
0.8674	376	0.0868	-	-	-	-	-	-
0.8697	377	0.0264	-	-	-	-	-	-
0.8720	378	0.0639	-	-	-	-	-	-
0.8743	379	0.1036	-	-	-	-	-	-
0.8766	380	0.0334	-	-	-	-	-	-
0.8789	381	0.04	-	-	-	-	-	-
0.8812	382	0.0095	-	-	-	-	-	-
0.8835	383	0.0371	-	-	-	-	-	-
0.8858	384	0.0585	-	-	-	-	-	-
0.8881	385	0.0353	-	-	-	-	-	-
0.8904	386	0.0095	-	-	-	-	-	-
0.8927	387	0.0126	-	-	-	-	-	-
0.8950	388	0.0384	-	-	-	-	-	-
0.8973	389	0.018	-	-	-	-	-	-
0.8997	390	0.057	-	-	-	-	-	-
0.9020	391	0.0371	-	-	-	-	-	-
0.9043	392	0.0475	-	-	-	-	-	-
0.9066	393	0.0972	-	-	-	-	-	-
0.9089	394	0.0189	-	-	-	-	-	-
0.9112	395	0.0993	-	-	-	-	-	-
0.9135	396	0.0527	-	-	-	-	-	-
0.9158	397	0.0466	-	-	-	-	-	-
0.9181	398	0.0383	-	-	-	-	-	-
0.9204	399	0.0322	-	-	-	-	-	-
0.9227	400	0.0651	0.9077	0.9074	0.9073	0.9077	0.9023	0.9078
0.9250	401	0.0055	-	-	-	-	-	-
0.9273	402	0.0083	-	-	-	-	-	-
0.9296	403	0.0062	-	-	-	-	-	-
0.9319	404	0.0085	-	-	-	-	-	-
0.9343	405	0.0179	-	-	-	-	-	-
0.9366	406	0.0041	-	-	-	-	-	-
0.9389	407	0.0978	-	-	-	-	-	-
0.9412	408	0.0068	-	-	-	-	-	-
0.9435	409	0.0145	-	-	-	-	-	-
0.9458	410	0.0098	-	-	-	-	-	-
0.9481	411	0.032	-	-	-	-	-	-
0.9504	412	0.0232	-	-	-	-	-	-
0.9527	413	0.0149	-	-	-	-	-	-
0.9550	414	0.0175	-	-	-	-	-	-
0.9573	415	0.0099	-	-	-	-	-	-
0.9596	416	0.0121	-	-	-	-	-	-
0.9619	417	0.108	-	-	-	-	-	-
0.9642	418	0.012	-	-	-	-	-	-
0.9666	419	0.0102	-	-	-	-	-	-
0.9689	420	0.0108	-	-	-	-	-	-
0.9712	421	0.2258	-	-	-	-	-	-
0.9735	422	0.0037	-	-	-	-	-	-
0.9758	423	0.0186	-	-	-	-	-	-
0.9781	424	0.0446	-	-	-	-	-	-
0.9804	425	0.1558	-	-	-	-	-	-
0.9827	426	0.023	-	-	-	-	-	-
0.9850	427	0.0075	-	-	-	-	-	-
0.9873	428	0.0095	-	-	-	-	-	-
0.9896	429	0.0141	-	-	-	-	-	-
0.9919	430	0.0617	-	-	-	-	-	-
0.9942	431	0.0961	-	-	-	-	-	-
0.9965	432	0.0058	-	-	-	-	-	-
0.9988	433	0.0399	-	-	-	-	-	-
1.0012	434	0.0063	-	-	-	-	-	-
1.0035	435	0.0288	-	-	-	-	-	-
1.0058	436	0.0041	-	-	-	-	-	-
1.0081	437	0.0071	-	-	-	-	-	-
1.0104	438	0.0233	-	-	-	-	-	-
1.0127	439	0.0135	-	-	-	-	-	-
1.0150	440	0.1015	-	-	-	-	-	-
1.0173	441	0.0045	-	-	-	-	-	-
1.0196	442	0.0088	-	-	-	-	-	-
1.0219	443	0.0086	-	-	-	-	-	-
1.0242	444	0.0072	-	-	-	-	-	-
1.0265	445	0.0147	-	-	-	-	-	-
1.0288	446	0.025	-	-	-	-	-	-
1.0311	447	0.0067	-	-	-	-	-	-
1.0334	448	0.0066	-	-	-	-	-	-
1.0358	449	0.0062	-	-	-	-	-	-
1.0381	450	0.0068	0.9091	0.9083	0.9045	0.9038	0.8983	0.9072
1.0404	451	0.0126	-	-	-	-	-	-
1.0427	452	0.0082	-	-	-	-	-	-
1.0450	453	0.0034	-	-	-	-	-	-
1.0473	454	0.04	-	-	-	-	-	-
1.0496	455	0.0235	-	-	-	-	-	-
1.0519	456	0.24	-	-	-	-	-	-
1.0542	457	0.0514	-	-	-	-	-	-
1.0565	458	0.0152	-	-	-	-	-	-
1.0588	459	0.0476	-	-	-	-	-	-
1.0611	460	0.0037	-	-	-	-	-	-
1.0634	461	0.0066	-	-	-	-	-	-
1.0657	462	0.0065	-	-	-	-	-	-
1.0681	463	0.0097	-	-	-	-	-	-
1.0704	464	0.0053	-	-	-	-	-	-
1.0727	465	0.0397	-	-	-	-	-	-
1.0750	466	0.0089	-	-	-	-	-	-
1.0773	467	0.0238	-	-	-	-	-	-
1.0796	468	0.0078	-	-	-	-	-	-
1.0819	469	0.0108	-	-	-	-	-	-
1.0842	470	0.0094	-	-	-	-	-	-
1.0865	471	0.0034	-	-	-	-	-	-
1.0888	472	0.0165	-	-	-	-	-	-
1.0911	473	0.0407	-	-	-	-	-	-
1.0934	474	0.0339	-	-	-	-	-	-
1.0957	475	0.0645	-	-	-	-	-	-
1.0980	476	0.0052	-	-	-	-	-	-
1.1003	477	0.0643	-	-	-	-	-	-
1.1027	478	0.0113	-	-	-	-	-	-
1.1050	479	0.007	-	-	-	-	-	-
1.1073	480	0.0062	-	-	-	-	-	-
1.1096	481	0.0232	-	-	-	-	-	-
1.1119	482	0.0374	-	-	-	-	-	-
1.1142	483	0.0582	-	-	-	-	-	-
1.1165	484	0.0396	-	-	-	-	-	-
1.1188	485	0.0041	-	-	-	-	-	-
1.1211	486	0.0064	-	-	-	-	-	-
1.1234	487	0.0248	-	-	-	-	-	-
1.1257	488	0.0052	-	-	-	-	-	-
1.1280	489	0.0095	-	-	-	-	-	-
1.1303	490	0.0681	-	-	-	-	-	-
1.1326	491	0.0082	-	-	-	-	-	-
1.1349	492	0.0279	-	-	-	-	-	-
1.1373	493	0.008	-	-	-	-	-	-
1.1396	494	0.0032	-	-	-	-	-	-
1.1419	495	0.041	-	-	-	-	-	-
1.1442	496	0.0089	-	-	-	-	-	-
1.1465	497	0.0289	-	-	-	-	-	-
1.1488	498	0.0232	-	-	-	-	-	-
1.1511	499	0.059	-	-	-	-	-	-
1.1534	500	0.0053	0.9039	0.9059	0.9032	0.9046	0.8995	0.9050

Framework Versions

Python: 3.11.9
Sentence Transformers: 3.0.1
Transformers: 4.44.2
PyTorch: 2.4.0+cu121
Accelerate: 0.33.0
Datasets: 2.19.2
Tokenizers: 0.19.1

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}
}