SentenceTransformer based on NeuML/pubmedbert-base-embeddings

This is a sentence-transformers model finetuned from NeuML/pubmedbert-base-embeddings on the geo_70k_multiplets_natural_language_annotation 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: NeuML/pubmedbert-base-embeddings
Maximum Sequence Length: 512 tokens
Output Dimensionality: 768 dimensions
Similarity Function: Cosine Similarity
Training Dataset:
- geo_70k_multiplets_natural_language_annotation
Language: code

Model Sources

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

Full Model Architecture

SentenceTransformer(
  (0): MMContextEncoder(
    (text_encoder): BertModel(
      (embeddings): BertEmbeddings(
        (word_embeddings): Embedding(30522, 768, padding_idx=0)
        (position_embeddings): Embedding(512, 768)
        (token_type_embeddings): Embedding(2, 768)
        (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
        (dropout): Dropout(p=0.1, inplace=False)
      )
      (encoder): BertEncoder(
        (layer): ModuleList(
          (0-11): 12 x BertLayer(
            (attention): BertAttention(
              (self): BertSdpaSelfAttention(
                (query): Linear(in_features=768, out_features=768, bias=True)
                (key): Linear(in_features=768, out_features=768, bias=True)
                (value): Linear(in_features=768, out_features=768, bias=True)
                (dropout): Dropout(p=0.1, inplace=False)
              )
              (output): BertSelfOutput(
                (dense): Linear(in_features=768, out_features=768, bias=True)
                (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
                (dropout): Dropout(p=0.1, inplace=False)
              )
            )
            (intermediate): BertIntermediate(
              (dense): Linear(in_features=768, out_features=3072, bias=True)
              (intermediate_act_fn): GELUActivation()
            )
            (output): BertOutput(
              (dense): Linear(in_features=3072, out_features=768, bias=True)
              (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
              (dropout): Dropout(p=0.1, inplace=False)
            )
          )
        )
      )
      (pooler): BertPooler(
        (dense): Linear(in_features=768, out_features=768, bias=True)
        (activation): Tanh()
      )
    )
    (pooling): 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("jo-mengr/mmcontext-pubmedbert-70k")
# Run inference
sentences = [
    'TMSB4X GAPDH LYZ B2M IFI30 ARHGDIB ENSG00000283907 MALAT1 SRGN PSAP LCP1 CXCL8 UBC SH3BGRL3 TUBA1B CTSC S100A4 MYL6 S100A11 ANXA1 CAPG ENSG00000203396 CCL2 CD68 H3-3B MYL12A RAC2 PPIA GLUL TXN SERPINB1 ENSG00000225840 TUBB SLC25A5 HMGB2 S100A6 GSTP1 H2AZ1 NCF2 HSPA5 GPI GRN HSPA8 LAMP2 C6orf62 CAT STMN1 HNRNPH1 KPNA2 LTA4H HLA-B TPP1 CD44 MSRB1 NPC2 ATP6V0D1 H4C3 LITAF SLC2A3 DDX3X AP2M1 PSMB3 CD24 ATP5F1E',
    'This measurement was conducted with Illumina HiSeq 2500. CFU-G/M U2AF1 wild-type 2 cells, which are granulomonocytic progenitor cells that have been transduced with U2AF1 wild-type. These cells are commonly used in studies of common myeloid progenitors, which are CD34-positive cultured cells.',
    'This measurement was conducted with Illumina HiSeq 2500. UM-UC18 bladder cancer cell line, a type of urinary bladder cancer cell line, cultured for study of bladder disease, cancer cell proliferation, and neoplasm.',
]
embeddings = model.encode(sentences)
print(embeddings.shape)
# [3, 768]

# Get the similarity scores for the embeddings
similarities = model.similarity(embeddings, embeddings)
print(similarities)
# tensor([[1.0000, 0.1404, 0.1000],
#         [0.1404, 1.0000, 0.4225],
#         [0.1000, 0.4225, 1.0000]])

Evaluation

Metrics

Triplet

Dataset: geo_70k_multiplets_natural_language_annotation_cell_sentence_2
Evaluated with TripletEvaluator

Metric	Value
cosine_accuracy	0.6657

Training Details

Training Dataset

geo_70k_multiplets_natural_language_annotation

Dataset: geo_70k_multiplets_natural_language_annotation at 4c62cd1
Size: 61,911 training samples
Columns: anchor, positive, negative_1, and negative_2

Approximate statistics based on the first 1000 samples:

	anchor	positive	negative_1	negative_2
type	string	string	string	string
details	min: 354 characters mean: 392.12 characters max: 601 characters	min: 83 characters mean: 189.5 characters max: 698 characters	min: 100 characters mean: 165.46 characters max: 465 characters	min: 358 characters mean: 435.31 characters max: 456 characters

Samples:

anchor	positive	negative_1	negative_2
`FTH1 FLNA TGFBI SLC7A5 KRT17 ENSG00000283907 COL5A1 FN1 ANXA2 KRT8 PLEC CLU AHNAK SCD TUBA1B ENSG00000225840 HSPA8 KRT7 CD59 FBN1 NEAT1 VIM PFN1 MALAT1 FLNB PRPF8 B2M FTL TMSB4X HSPG2 FASN BASP1 SLC3A2 ACTN4 PXDN SLC38A2 ACLY TUBB THBS1 LDHB SLC38A1 TPM2 ITGA5 PLS3 APP TIMP2 COL3A1 TXNRD1 LMNA ASPH CSRP1 COL12A1 MUC16 ACTN4 LDHA LDHA CPS1 TPM1 CANX COL4A2 PDE3A CANX HIF1A TALAM1`	`This measurement was conducted with Illumina HiSeq 2000. 5-day HeLa cell line with ELAVL1/HuR siRNA1 knockdown, 120 hours post-transfection.`	`This measurement was conducted with Illumina HiSeq 2000. BJ fibroblast cells in a proliferative stage, with polyA RNA subtype.`	`LINC03031 SINHCAFP3 COL1A1 COL6A3 ENSG00000261526 MYL9 COL1A2 MLLT3 CALD1 IGFBP5 ANK3 BTBD9 CCDC80 TPM1 TPM2 ENSG00000233388 MMP14 TAGLN THBS1 HSPA1B SKIL NEDD4L MARCKS MALAT1 ACTN4 ACTN1 PLTP MMP2 PXDN RAB11FIP1 CD63 IGFBP7 TUBB BLM FN1 PTMS MAP7-AS1 MYLK NCALD SPINT2 LMNA FTH1 ENSG00000267469 PRRG3 FSTL1 RNF187 MGLL CAVIN1 COL5A1 COL3A1 FGF23 ENSG00000226824 SPARC ASPH SLC16A3 LPAR4 ENSG00000269888 COL4A2 RRBP1 SH3PXD2A COL4A1 LBH SOX4 LDLR`
`LINC03031 SINHCAFP3 ENSG00000261526 COL6A3 COL1A1 TPM1 MLLT3 MYL9 BTBD9 ENSG00000233388 COL1A2 CALD1 TPM2 TUBB HSPA1B ANK3 TAGLN IGFBP5 NEDD4L PRR11 ACTN4 SKIL MALAT1 FN1 PTMS MMP14 THBS1 ACTN1 NCALD PXDN CCDC80 MARCKS SPINT2 PRRG3 VIM RAB3B TUBB MAP7-AS1 SERPINE1 CD63 LMNA ASPH ENSG00000226824 RAB11FIP1 SLC16A3 CAVIN1 COL4A1 LPAR4 TOP2A FGF23 BLM COL4A2 MGLL NR6A1 FSTL1 MMP2 ENSG00000269888 ENSG00000267484 LOXL2 PLTP MYLK KIAA0825 ENSG00000267469 FTH1`	`This measurement was conducted with Illumina HiSeq 2000. BJ fibroblast cells in a proliferative stage, with polyA RNA subtype.`	`This measurement was conducted with Illumina HiSeq 2000. 5-day HeLa cell line with ELAVL1/HuR siRNA1 knockdown, 120 hours post-transfection.`	`LINC03031 SINHCAFP3 COL1A1 COL6A3 ENSG00000261526 MYL9 COL1A2 MLLT3 CALD1 IGFBP5 ANK3 BTBD9 CCDC80 TPM1 TPM2 ENSG00000233388 MMP14 TAGLN THBS1 HSPA1B SKIL NEDD4L MARCKS MALAT1 ACTN4 ACTN1 PLTP MMP2 PXDN RAB11FIP1 CD63 IGFBP7 TUBB BLM FN1 PTMS MAP7-AS1 MYLK NCALD SPINT2 LMNA FTH1 ENSG00000267469 PRRG3 FSTL1 RNF187 MGLL CAVIN1 COL5A1 COL3A1 FGF23 ENSG00000226824 SPARC ASPH SLC16A3 LPAR4 ENSG00000269888 COL4A2 RRBP1 SH3PXD2A COL4A1 LBH SOX4 LDLR`
`LINC03031 SINHCAFP3 COL1A1 COL6A3 ENSG00000261526 MYL9 COL1A2 MLLT3 CALD1 IGFBP5 ANK3 BTBD9 CCDC80 TPM1 TPM2 ENSG00000233388 MMP14 TAGLN THBS1 HSPA1B SKIL NEDD4L MARCKS MALAT1 ACTN4 ACTN1 PLTP MMP2 PXDN RAB11FIP1 CD63 IGFBP7 TUBB BLM FN1 PTMS MAP7-AS1 MYLK NCALD SPINT2 LMNA FTH1 ENSG00000267469 PRRG3 FSTL1 RNF187 MGLL CAVIN1 COL5A1 COL3A1 FGF23 ENSG00000226824 SPARC ASPH SLC16A3 LPAR4 ENSG00000269888 COL4A2 RRBP1 SH3PXD2A COL4A1 LBH SOX4 LDLR`	`This measurement was conducted with Illumina HiSeq 2000. BJ fibroblast cells at a confluent growth stage, with polyA RNA subtype.`	`This measurement was conducted with Illumina HiSeq 2000. 5-day HeLa cell line with ELAVL1/HuR siRNA1 knockdown, 120 hours post-transfection.`	`LINC03031 SINHCAFP3 ENSG00000261526 COL6A3 COL1A1 TPM1 MLLT3 MYL9 BTBD9 ENSG00000233388 COL1A2 CALD1 TPM2 TUBB HSPA1B ANK3 TAGLN IGFBP5 NEDD4L PRR11 ACTN4 SKIL MALAT1 FN1 PTMS MMP14 THBS1 ACTN1 NCALD PXDN CCDC80 MARCKS SPINT2 PRRG3 VIM RAB3B TUBB MAP7-AS1 SERPINE1 CD63 LMNA ASPH ENSG00000226824 RAB11FIP1 SLC16A3 CAVIN1 COL4A1 LPAR4 TOP2A FGF23 BLM COL4A2 MGLL NR6A1 FSTL1 MMP2 ENSG00000269888 ENSG00000267484 LOXL2 PLTP MYLK KIAA0825 ENSG00000267469 FTH1`

Loss: MultipleNegativesRankingLoss with these parameters:

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

Evaluation Dataset

geo_70k_multiplets_natural_language_annotation

Dataset: geo_70k_multiplets_natural_language_annotation at 4c62cd1
Size: 6,901 evaluation samples
Columns: anchor, positive, negative_1, and negative_2

Approximate statistics based on the first 1000 samples:

	anchor	positive	negative_1	negative_2
type	string	string	string	string
details	min: 351 characters mean: 395.91 characters max: 472 characters	min: 78 characters mean: 191.46 characters max: 983 characters	min: 90 characters mean: 217.63 characters max: 702 characters	min: 364 characters mean: 385.93 characters max: 427 characters

Samples:

anchor	positive	negative_1	negative_2
`IGF2 FN1 TF A2M C3 GAPDH AGT SCD LYZ SELENOP TUBB CP FTH1 RELN DSP SPTBN1 TMEM123 UBC LDHA LDHA TUBA1B MAP4K4 APLP2 PDIA6 HSPA5 H1-0 CCN2 SERPINF1 RRBP1 HNRNPH1 PPIA DHCR24 HSP90AA1 ACSL4 SLC38A2 APOE COL27A1 EIF3C BSG TOP2A RBM39 TGOLN2 ERBB3 CLU PNN MBNL3 GLUD1 LGALS3BP AHNAK PSAP CANX SNRNP200 NEAT1 CTSB ENSG00000283907 FAT1 PLOD2 TFPI NCOA4 LAMA5 FARP1 SLC2A1 ILF3 CALM3`	`This measurement was conducted with Illumina HiSeq 2000. 15-year-old male HepG2 immortalized cell line with hepatocellular carcinoma, transiently expressing shRNA targeting PKM2 for RNA-seq study.`	`This measurement was conducted with Illumina HiSeq 2000. 15-year-old male patient with hepatocellular carcinoma; HNRNPC knocked down via shRNA in HepG2 (immortalized cell line) for RNA-seq analysis.`	`FTH1 ENSG00000283907 GAPDH MALAT1 TALAM1 HSPA5 HMGA1 UBC PSAP TUBB ENSG00000225840 TUBA1B WARS1 HSPA9 PIM1 AARS1 NEAT1 CANX ASNS VEGFA GARS1 FLNA HSPA8 GLUL SND1 ANP32B TMSB4X SARS1 DDIT4 UBR4 HSP90AA1 ENSG00000258017 TXNIP PPIA EIF3C GSTP1 SCD EPB41 IARS1 GANAB STMN1 MCM7 PRKCB SLC25A6 EPRS1 MARS1 HNRNPF SLC25A5 TUBB COPA AHNAK VAT1 LDHA LDHA XIST HADHA PHGDH LCP1 GTPBP2 LONP1 TARS1 SLC1A5 LDHB SFPQ`
`B2M TMSB4X HLA-B HSPA8 ETS1 HLA-E SARAF UBC HLA-A ENSG00000283907 HLA-A HSP90AA1 HLA-A LCP1 HLA-C GAPDH TMEM123 ENSG00000237550 PPIA CORO1A ARHGDIB TMSB10 FTH1 CDC42SE2 TXNIP CXCR4 SUN2 STAT1 CD44 ITM2B TUBB HLA-B TUBA1B IL32 LDHB TLE5 PCBP1 MBNL1 RIPOR2 H3-3B JAK1 HLA-B PSAP BTG1 LBH HSPA5 MYL6 HLA-C CANX RAC2 SAMHD1 SLC38A1 DDX3X HLA-C HNRNPF EPB41 SLC25A6 CTDSP2 SLC38A2 MATR3 SH3BGRL3 ESYT1 CCL5 TRAM1`	`This measurement was conducted with Illumina HiSeq 2000. 16-year-old female's T cells from a control group, stimulated with ag85 at timepoint 0, and primary cells.`	`This measurement was conducted with Illumina HiSeq 2000. 17-year-old male's monocytes stimulated with mTb, taken at 180 days post-stimulation, as part of the control group in a study.`	`CXCL8 B2M ENSG00000225840 HLA-DRA SRGN TXN H3-3B ID2 PLAUR PSAP ANXA2 GSTO1 DNAJB9 LGALS3 ADA UBC GAPDH SAT1 CTSL MYL6 NSMAF PMAIP1 FTH1 CD44 BID TMSB4X ARL8B ATP6V1B2 CD74 NAMPT HLA-E ISG20 CD83 CHCHD2 IFI30 ANXA1 BZW1 CSTB CXCL2 TXNRD1 NR3C1 NINJ1 EIF5 SLC3A2 GLIPR1 CDKN1A NRIP3 SLC7A11 TMED2 G0S2 LITAF LCP1 TMSB10 CREM HSPA8 HLA-B HSPA5 CD59 ENSG00000237550 NCOA4 LGALS1 ENSG00000203396 IFNGR1 ARF4`
`THBS1 FLNA FTH1 FN1 GAPDH HLA-B TUBB HSP90AA1 PLEC TUBA1B AHNAK HLA-C TLN1 ANXA2 AXL HSPA8 TXNRD1 ACTN1 ACTN4 SLC7A5 TUBB CAVIN1 HUWE1 MPRIP MKI67 MYL6 LASP1 TGFBI TOP2A PPIA B2M LMNA TMSB4X TUBB4B KPNA2 COL6A2 GPRC5A COL5A1 HMGA1 NCAPD2 COL4A2 TMSB10 ENSG00000283907 SPTBN1 LAMC1 HSPA5 VCL ASPM PRPF8 UBR4 SPTAN1 ANLN CENPF CLIC4 COTL1 UBC AKR1B1 CALM3 PLS3 TRIO HIF1A CANX TGFB2 KTN1`	`This measurement was conducted with Illumina HiSeq 2500. UM-UC18 bladder cancer cell line, a type of urinary bladder cancer cell line, cultured for study of bladder disease, cancer cell proliferation, and neoplasm.`	`This measurement was conducted with NextSeq 500. HeLa cells with PARP knockdown treatment.`	`GAPDH HSPA8 TUBA1B SRGN HSP90AA1 TMSB4X TUBB ENSG00000237550 PPIA LYZ HNRNPH1 EMB LDHB HSPA5 LCP1 ARHGDIB STMN1 CANX MACROH2A1 TUBB ILF3 FLNA UBC GPI MCM4 B2M MAT2A TFRC MBNL1 H2AZ1 MCM7 XRCC5 HSPA9 IMPDH2 HMGA1 SFPQ CAT AHNAK CORO1A PRPF8 BSG MAN2B1 NOP56 SLC25A5 HMGB2 APEX1 H3-3B GPX1 COLGALT1 GANAB TCP1 HNRNPF SLC25A6 CAPRIN1 APLP2 MYC TUBB4B LDHA LDHA KIAA0100 GSTP1 SRSF2 TOP2A MKI67`

Loss: MultipleNegativesRankingLoss with these parameters:

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

Training Hyperparameters

Non-Default Hyperparameters

eval_strategy: steps
per_device_train_batch_size: 256
per_device_eval_batch_size: 256
learning_rate: 0.1
num_train_epochs: 4
warmup_ratio: 0.1
bf16: True
gradient_checkpointing: True

All Hyperparameters

Click to expand

overwrite_output_dir: False
do_predict: False
eval_strategy: steps
prediction_loss_only: True
per_device_train_batch_size: 256
per_device_eval_batch_size: 256
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: 0.1
weight_decay: 0.0
adam_beta1: 0.9
adam_beta2: 0.999
adam_epsilon: 1e-08
max_grad_norm: 1.0
num_train_epochs: 4
max_steps: -1
lr_scheduler_type: linear
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
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: 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}
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
hub_revision: None
gradient_checkpointing: True
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
liger_kernel_config: None
eval_use_gather_object: False
average_tokens_across_devices: False
prompts: None
batch_sampler: batch_sampler
multi_dataset_batch_sampler: proportional
router_mapping: {}
learning_rate_mapping: {}

Training Logs

Epoch	Step	Training Loss	geo 70k multiplets natural language annotation loss	geo_70k_multiplets_natural_language_annotation_cell_sentence_2_cosine_accuracy
0.4132	100	19.2596	20.4720	0.6657
0.8264	200	19.2622	20.4720	0.6657
1.2397	300	19.2606	20.4720	0.6657
1.6529	400	19.2546	20.4720	0.6657
2.0661	500	19.2629	20.4720	0.6657
2.4793	600	19.2551	20.4720	0.6657
2.8926	700	19.2647	20.4720	0.6657
3.3058	800	19.2696	20.4720	0.6657
3.7190	900	19.2583	20.4720	0.6657

Framework Versions

Python: 3.11.6
Sentence Transformers: 5.0.0
Transformers: 4.55.0.dev0
PyTorch: 2.5.1+cu121
Accelerate: 1.9.0
Datasets: 2.19.1
Tokenizers: 0.21.4

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",
}

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}
}

jo-mengr
/

mmcontext-pubmedbert-70k

SentenceTransformer based on NeuML/pubmedbert-base-embeddings

Model Details

Model Description

Model Sources

Full Model Architecture

Usage

Direct Usage (Sentence Transformers)

Evaluation

Metrics

Triplet

Training Details

Training Dataset

geo_70k_multiplets_natural_language_annotation

Evaluation Dataset

geo_70k_multiplets_natural_language_annotation

Training Hyperparameters

Non-Default Hyperparameters

All Hyperparameters

Training Logs

Framework Versions

Citation

BibTeX

Sentence Transformers

MultipleNegativesRankingLoss

Model tree for jo-mengr/mmcontext-pubmedbert-70k

Evaluation results