bert-base-uncased adapter finetuned on GooAQ pairs
This is a sentence-transformers model finetuned from google-bert/bert-base-uncased on the gooaq 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-bert/bert-base-uncased
- Maximum Sequence Length: 512 tokens
- Output Dimensionality: 768 dimensions
- Similarity Function: Cosine Similarity
- Training Dataset:
- Language: en
- License: apache-2.0
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: BertModel
(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})
)
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("tomaarsen/bert-base-uncased-gooaq")
# Run inference
sentences = [
'how can i download youtube videos with internet download manager?',
"['Go to settings and then click on extensions (top left side in chrome).', 'Minimise your browser and open the location (folder) where IDM is installed. ... ', 'Find the file “IDMGCExt. ... ', 'Drag this file to your chrome browser and drop to install the IDM extension.']",
"Coca-Cola might rot your teeth and load your body with sugar and calories, but it's actually an effective and safe first line of treatment for some stomach blockages, researchers say.",
]
embeddings = model.encode(sentences)
print(embeddings.shape)
# [3, 768]
# Get the similarity scores for the embeddings
similarities = model.similarity(embeddings, embeddings)
print(similarities.shape)
# [3, 3]
Evaluation
Metrics
Information Retrieval
- Datasets:
NanoClimateFEVER
,NanoDBPedia
,NanoFEVER
,NanoFiQA2018
,NanoHotpotQA
,NanoMSMARCO
,NanoNFCorpus
,NanoNQ
,NanoQuoraRetrieval
,NanoSCIDOCS
,NanoArguAna
,NanoSciFact
andNanoTouche2020
- Evaluated with
InformationRetrievalEvaluator
Metric | NanoClimateFEVER | NanoDBPedia | NanoFEVER | NanoFiQA2018 | NanoHotpotQA | NanoMSMARCO | NanoNFCorpus | NanoNQ | NanoQuoraRetrieval | NanoSCIDOCS | NanoArguAna | NanoSciFact | NanoTouche2020 |
---|---|---|---|---|---|---|---|---|---|---|---|---|---|
cosine_accuracy@1 | 0.24 | 0.54 | 0.54 | 0.24 | 0.6 | 0.28 | 0.32 | 0.38 | 0.84 | 0.26 | 0.16 | 0.42 | 0.449 |
cosine_accuracy@3 | 0.42 | 0.8 | 0.82 | 0.4 | 0.68 | 0.48 | 0.48 | 0.54 | 0.94 | 0.46 | 0.58 | 0.58 | 0.898 |
cosine_accuracy@5 | 0.46 | 0.84 | 0.9 | 0.5 | 0.74 | 0.54 | 0.5 | 0.62 | 0.98 | 0.6 | 0.74 | 0.62 | 0.9184 |
cosine_accuracy@10 | 0.56 | 0.92 | 0.92 | 0.6 | 0.88 | 0.66 | 0.56 | 0.74 | 1.0 | 0.68 | 0.84 | 0.64 | 0.9796 |
cosine_precision@1 | 0.24 | 0.54 | 0.54 | 0.24 | 0.6 | 0.28 | 0.32 | 0.38 | 0.84 | 0.26 | 0.16 | 0.42 | 0.449 |
cosine_precision@3 | 0.16 | 0.4867 | 0.2733 | 0.16 | 0.2867 | 0.16 | 0.3067 | 0.18 | 0.3867 | 0.2067 | 0.1933 | 0.2067 | 0.4966 |
cosine_precision@5 | 0.108 | 0.444 | 0.184 | 0.14 | 0.192 | 0.108 | 0.244 | 0.128 | 0.248 | 0.184 | 0.148 | 0.14 | 0.4571 |
cosine_precision@10 | 0.07 | 0.39 | 0.096 | 0.088 | 0.118 | 0.066 | 0.184 | 0.076 | 0.13 | 0.126 | 0.084 | 0.076 | 0.3898 |
cosine_recall@1 | 0.1317 | 0.0468 | 0.53 | 0.1147 | 0.3 | 0.28 | 0.0209 | 0.38 | 0.7573 | 0.054 | 0.16 | 0.4 | 0.0348 |
cosine_recall@3 | 0.2083 | 0.1112 | 0.7767 | 0.2287 | 0.43 | 0.48 | 0.0534 | 0.51 | 0.912 | 0.1287 | 0.58 | 0.56 | 0.1111 |
cosine_recall@5 | 0.2417 | 0.1583 | 0.8567 | 0.3166 | 0.48 | 0.54 | 0.0639 | 0.6 | 0.946 | 0.1897 | 0.74 | 0.605 | 0.1656 |
cosine_recall@10 | 0.2967 | 0.2555 | 0.8867 | 0.3986 | 0.59 | 0.66 | 0.0805 | 0.71 | 0.9793 | 0.2587 | 0.84 | 0.64 | 0.2685 |
cosine_ndcg@10 | 0.2552 | 0.4644 | 0.7349 | 0.2926 | 0.5292 | 0.468 | 0.2307 | 0.5387 | 0.9158 | 0.2418 | 0.5045 | 0.538 | 0.4323 |
cosine_mrr@10 | 0.3379 | 0.6797 | 0.6962 | 0.3385 | 0.6639 | 0.4079 | 0.3979 | 0.4905 | 0.901 | 0.3804 | 0.3963 | 0.5057 | 0.6533 |
cosine_map@100 | 0.2076 | 0.3253 | 0.6788 | 0.2372 | 0.4523 | 0.4276 | 0.0966 | 0.4896 | 0.8894 | 0.1865 | 0.4007 | 0.508 | 0.3349 |
Nano BEIR
- Dataset:
NanoBEIR_mean
- Evaluated with
NanoBEIREvaluator
Metric | Value |
---|---|
cosine_accuracy@1 | 0.4053 |
cosine_accuracy@3 | 0.6214 |
cosine_accuracy@5 | 0.6891 |
cosine_accuracy@10 | 0.7677 |
cosine_precision@1 | 0.4053 |
cosine_precision@3 | 0.2695 |
cosine_precision@5 | 0.2096 |
cosine_precision@10 | 0.1457 |
cosine_recall@1 | 0.2469 |
cosine_recall@3 | 0.3915 |
cosine_recall@5 | 0.4541 |
cosine_recall@10 | 0.528 |
cosine_ndcg@10 | 0.4728 |
cosine_mrr@10 | 0.5269 |
cosine_map@100 | 0.4027 |
Training Details
Training Dataset
gooaq
- Dataset: gooaq at b089f72
- Size: 3,012,496 training samples
- Columns:
question
andanswer
- Approximate statistics based on the first 1000 samples:
question answer type string string details - min: 8 tokens
- mean: 11.86 tokens
- max: 21 tokens
- min: 14 tokens
- mean: 60.48 tokens
- max: 138 tokens
- Samples:
question answer what is the difference between broilers and layers?
An egg laying poultry is called egger or layer whereas broilers are reared for obtaining meat. So a layer should be able to produce more number of large sized eggs, without growing too much. On the other hand, a broiler should yield more meat and hence should be able to grow well.
what is the difference between chronological order and spatial order?
As a writer, you should always remember that unlike chronological order and the other organizational methods for data, spatial order does not take into account the time. Spatial order is primarily focused on the location. All it does is take into account the location of objects and not the time.
is kamagra same as viagra?
Kamagra is thought to contain the same active ingredient as Viagra, sildenafil citrate. In theory, it should work in much the same way as Viagra, taking about 45 minutes to take effect, and lasting for around 4-6 hours. However, this will vary from person to person.
- Loss:
MatryoshkaLoss
with these parameters:{ "loss": "CachedMultipleNegativesRankingLoss", "matryoshka_dims": [ 768, 512, 256, 128, 64, 32 ], "matryoshka_weights": [ 1, 1, 1, 1, 1, 1 ], "n_dims_per_step": -1 }
Evaluation Dataset
gooaq
- Dataset: gooaq at b089f72
- Size: 3,012,496 evaluation samples
- Columns:
question
andanswer
- Approximate statistics based on the first 1000 samples:
question answer type string string details - min: 8 tokens
- mean: 11.88 tokens
- max: 22 tokens
- min: 14 tokens
- mean: 61.03 tokens
- max: 127 tokens
- Samples:
question answer how do i program my directv remote with my tv?
['Press MENU on your remote.', 'Select Settings & Help > Settings > Remote Control > Program Remote.', 'Choose the device (TV, audio, DVD) you wish to program. ... ', 'Follow the on-screen prompts to complete programming.']
are rodrigues fruit bats nocturnal?
Before its numbers were threatened by habitat destruction, storms, and hunting, some of those groups could number 500 or more members. Sunrise, sunset. Rodrigues fruit bats are most active at dawn, at dusk, and at night.
why does your heart rate increase during exercise bbc bitesize?
During exercise there is an increase in physical activity and muscle cells respire more than they do when the body is at rest. The heart rate increases during exercise. The rate and depth of breathing increases - this makes sure that more oxygen is absorbed into the blood, and more carbon dioxide is removed from it.
- Loss:
MatryoshkaLoss
with these parameters:{ "loss": "CachedMultipleNegativesRankingLoss", "matryoshka_dims": [ 768, 512, 256, 128, 64, 32 ], "matryoshka_weights": [ 1, 1, 1, 1, 1, 1 ], "n_dims_per_step": -1 }
Training Hyperparameters
Non-Default Hyperparameters
eval_strategy
: stepsper_device_train_batch_size
: 1024per_device_eval_batch_size
: 1024learning_rate
: 2e-05num_train_epochs
: 1warmup_ratio
: 0.1seed
: 12bf16
: Truebatch_sampler
: no_duplicates
All Hyperparameters
Click to expand
overwrite_output_dir
: Falsedo_predict
: Falseeval_strategy
: stepsprediction_loss_only
: Trueper_device_train_batch_size
: 1024per_device_eval_batch_size
: 1024per_gpu_train_batch_size
: Noneper_gpu_eval_batch_size
: Nonegradient_accumulation_steps
: 1eval_accumulation_steps
: Nonetorch_empty_cache_steps
: Nonelearning_rate
: 2e-05weight_decay
: 0.0adam_beta1
: 0.9adam_beta2
: 0.999adam_epsilon
: 1e-08max_grad_norm
: 1.0num_train_epochs
: 1max_steps
: -1lr_scheduler_type
: linearlr_scheduler_kwargs
: {}warmup_ratio
: 0.1warmup_steps
: 0log_level
: passivelog_level_replica
: warninglog_on_each_node
: Truelogging_nan_inf_filter
: Truesave_safetensors
: Truesave_on_each_node
: Falsesave_only_model
: Falserestore_callback_states_from_checkpoint
: Falseno_cuda
: Falseuse_cpu
: Falseuse_mps_device
: Falseseed
: 12data_seed
: Nonejit_mode_eval
: Falseuse_ipex
: Falsebf16
: Truefp16
: Falsefp16_opt_level
: O1half_precision_backend
: autobf16_full_eval
: Falsefp16_full_eval
: Falsetf32
: Nonelocal_rank
: 0ddp_backend
: Nonetpu_num_cores
: Nonetpu_metrics_debug
: Falsedebug
: []dataloader_drop_last
: Falsedataloader_num_workers
: 0dataloader_prefetch_factor
: Nonepast_index
: -1disable_tqdm
: Falseremove_unused_columns
: Truelabel_names
: Noneload_best_model_at_end
: Falseignore_data_skip
: Falsefsdp
: []fsdp_min_num_params
: 0fsdp_config
: {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}fsdp_transformer_layer_cls_to_wrap
: Noneaccelerator_config
: {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}deepspeed
: Nonelabel_smoothing_factor
: 0.0optim
: adamw_torchoptim_args
: Noneadafactor
: Falsegroup_by_length
: Falselength_column_name
: lengthddp_find_unused_parameters
: Noneddp_bucket_cap_mb
: Noneddp_broadcast_buffers
: Falsedataloader_pin_memory
: Truedataloader_persistent_workers
: Falseskip_memory_metrics
: Trueuse_legacy_prediction_loop
: Falsepush_to_hub
: Falseresume_from_checkpoint
: Nonehub_model_id
: Nonehub_strategy
: every_savehub_private_repo
: Falsehub_always_push
: Falsegradient_checkpointing
: Falsegradient_checkpointing_kwargs
: Noneinclude_inputs_for_metrics
: Falseinclude_for_metrics
: []eval_do_concat_batches
: Truefp16_backend
: autopush_to_hub_model_id
: Nonepush_to_hub_organization
: Nonemp_parameters
:auto_find_batch_size
: Falsefull_determinism
: Falsetorchdynamo
: Noneray_scope
: lastddp_timeout
: 1800torch_compile
: Falsetorch_compile_backend
: Nonetorch_compile_mode
: Nonedispatch_batches
: Nonesplit_batches
: Noneinclude_tokens_per_second
: Falseinclude_num_input_tokens_seen
: Falseneftune_noise_alpha
: Noneoptim_target_modules
: Nonebatch_eval_metrics
: Falseeval_on_start
: Falseuse_liger_kernel
: Falseeval_use_gather_object
: Falseaverage_tokens_across_devices
: Falseprompts
: Nonebatch_sampler
: no_duplicatesmulti_dataset_batch_sampler
: proportional
Training Logs
Epoch | Step | Training Loss | Validation Loss | NanoClimateFEVER_cosine_ndcg@10 | NanoDBPedia_cosine_ndcg@10 | NanoFEVER_cosine_ndcg@10 | NanoFiQA2018_cosine_ndcg@10 | NanoHotpotQA_cosine_ndcg@10 | NanoMSMARCO_cosine_ndcg@10 | NanoNFCorpus_cosine_ndcg@10 | NanoNQ_cosine_ndcg@10 | NanoQuoraRetrieval_cosine_ndcg@10 | NanoSCIDOCS_cosine_ndcg@10 | NanoArguAna_cosine_ndcg@10 | NanoSciFact_cosine_ndcg@10 | NanoTouche2020_cosine_ndcg@10 | NanoBEIR_mean_cosine_ndcg@10 |
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
0 | 0 | - | - | 0.1046 | 0.2182 | 0.1573 | 0.0575 | 0.2597 | 0.1602 | 0.0521 | 0.0493 | 0.7310 | 0.1320 | 0.2309 | 0.1240 | 0.0970 | 0.1826 |
0.0010 | 1 | 28.4268 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
0.0256 | 25 | 24.7252 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
0.0512 | 50 | 13.3628 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
0.0768 | 75 | 7.843 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
0.1024 | 100 | 5.7393 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
0.1279 | 125 | 4.6576 | 2.3368 | 0.2890 | 0.4610 | 0.7408 | 0.2882 | 0.5446 | 0.4091 | 0.2179 | 0.4664 | 0.9079 | 0.2394 | 0.5433 | 0.5003 | 0.4318 | 0.4646 |
0.1535 | 150 | 4.0846 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
0.1791 | 175 | 3.7129 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
0.2047 | 200 | 3.4899 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
0.2303 | 225 | 3.3263 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
0.2559 | 250 | 3.2013 | 1.6545 | 0.2622 | 0.4744 | 0.7456 | 0.2934 | 0.5371 | 0.4326 | 0.2290 | 0.5157 | 0.9130 | 0.2577 | 0.5189 | 0.5155 | 0.4302 | 0.4712 |
0.2815 | 275 | 2.9109 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
0.3071 | 300 | 2.9064 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
0.3327 | 325 | 2.8215 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
0.3582 | 350 | 2.7893 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
0.3838 | 375 | 2.6663 | 1.4146 | 0.2629 | 0.4657 | 0.7330 | 0.2853 | 0.5299 | 0.4346 | 0.2311 | 0.5216 | 0.9172 | 0.2513 | 0.5133 | 0.5429 | 0.4287 | 0.4706 |
0.4094 | 400 | 2.6672 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
0.4350 | 425 | 2.5587 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
0.4606 | 450 | 2.5001 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
0.4862 | 475 | 2.4476 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
0.5118 | 500 | 2.4127 | 1.2843 | 0.2565 | 0.4668 | 0.7289 | 0.2838 | 0.5392 | 0.4599 | 0.2284 | 0.5238 | 0.9021 | 0.2416 | 0.4971 | 0.5349 | 0.4320 | 0.4688 |
0.5374 | 525 | 2.414 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
0.5629 | 550 | 2.3723 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
0.5885 | 575 | 2.3418 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
0.6141 | 600 | 2.2862 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
0.6397 | 625 | 2.207 | 1.2078 | 0.2613 | 0.4542 | 0.7382 | 0.2817 | 0.5230 | 0.4664 | 0.2282 | 0.5266 | 0.9095 | 0.2453 | 0.5127 | 0.5414 | 0.4239 | 0.4702 |
0.6653 | 650 | 2.2305 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
0.6909 | 675 | 2.2409 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
0.7165 | 700 | 2.2001 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
0.7421 | 725 | 2.1923 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
0.7677 | 750 | 2.195 | 1.1538 | 0.2549 | 0.4671 | 0.7333 | 0.2804 | 0.5265 | 0.4659 | 0.2321 | 0.5331 | 0.9086 | 0.2429 | 0.5070 | 0.5430 | 0.4369 | 0.4717 |
0.7932 | 775 | 2.1826 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
0.8188 | 800 | 2.1754 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
0.8444 | 825 | 2.1141 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
0.8700 | 850 | 2.1572 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
0.8956 | 875 | 2.1126 | 1.1256 | 0.2505 | 0.4622 | 0.7293 | 0.2857 | 0.5286 | 0.4823 | 0.2308 | 0.5397 | 0.9158 | 0.2412 | 0.5050 | 0.5365 | 0.4387 | 0.4728 |
0.9212 | 900 | 2.0755 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
0.9468 | 925 | 2.1032 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
0.9724 | 950 | 2.1211 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
0.9980 | 975 | 2.0826 | - | - | - | - | - | - | - | - | - | - | - | - | - | - | - |
1.0 | 977 | - | - | 0.2552 | 0.4644 | 0.7349 | 0.2926 | 0.5292 | 0.4680 | 0.2307 | 0.5387 | 0.9158 | 0.2418 | 0.5045 | 0.5380 | 0.4323 | 0.4728 |
Environmental Impact
Carbon emissions were measured using CodeCarbon.
- Energy Consumed: 0.624 kWh
- Carbon Emitted: 0.243 kg of CO2
- Hours Used: 1.619 hours
Training Hardware
- On Cloud: No
- GPU Model: 1 x NVIDIA GeForce RTX 3090
- CPU Model: 13th Gen Intel(R) Core(TM) i7-13700K
- RAM Size: 31.78 GB
Framework Versions
- Python: 3.11.6
- Sentence Transformers: 3.4.0.dev0
- Transformers: 4.46.2
- PyTorch: 2.5.0+cu121
- Accelerate: 0.35.0.dev0
- Datasets: 2.20.0
- Tokenizers: 0.20.3
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}
}
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}
}
- Downloads last month
- 20
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social
visibility and check back later, or deploy to Inference Endpoints (dedicated)
instead.
Model tree for tomaarsen/bert-base-uncased-gooaq
Base model
google-bert/bert-base-uncasedDataset used to train tomaarsen/bert-base-uncased-gooaq
Evaluation results
- Cosine Accuracy@1 on NanoClimateFEVERself-reported0.240
- Cosine Accuracy@3 on NanoClimateFEVERself-reported0.420
- Cosine Accuracy@5 on NanoClimateFEVERself-reported0.460
- Cosine Accuracy@10 on NanoClimateFEVERself-reported0.560
- Cosine Precision@1 on NanoClimateFEVERself-reported0.240
- Cosine Precision@3 on NanoClimateFEVERself-reported0.160
- Cosine Precision@5 on NanoClimateFEVERself-reported0.108
- Cosine Precision@10 on NanoClimateFEVERself-reported0.070
- Cosine Recall@1 on NanoClimateFEVERself-reported0.132
- Cosine Recall@3 on NanoClimateFEVERself-reported0.208