Initial Upload

1_Pooling/config.json

@@ -0,0 +1,7 @@
+{
+  "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
+}

all_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 1.0,
+    "train_loss": 1.4607645520019532,
+    "train_runtime": 81957.8858,
+    "train_samples": 0,
+    "train_samples_per_second": 1220.139,
+    "train_steps_per_second": 1.22
+}

config.json

@@ -0,0 +1,27 @@
+{
+  "_name_or_path": "/dccstor/aashka1/sentence_embeddings/output/NASA_finetuning_v2/NASA-WatBERT-300M-NASA-BioMedical/2GPU/FinetuneNASA_pooling_mean_qlen64_plen256_freeze_embedding_true_lr_2e-5_bsz_500_300000steps/checkpoint-300000",
+  "architectures": [
+    "RobertaModel"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "bos_token_id": 0,
+  "classifier_dropout": null,
+  "eos_token_id": 2,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 768,
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "layer_norm_eps": 1e-05,
+  "max_position_embeddings": 514,
+  "model_type": "roberta",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 12,
+  "pad_token_id": 1,
+  "position_embedding_type": "absolute",
+  "torch_dtype": "float16",
+  "transformers_version": "4.28.1",
+  "type_vocab_size": 1,
+  "use_cache": true,
+  "vocab_size": 50265
+}

log.txt

@@ -0,0 +1,3 @@
+/opt/share/exec/jbsub8 -cores 8+2 -require a100_80gb && hname!=cccxc607 -queue nonstandard -proj embeddings-kd -name NASA-ft -mem 100G -out /dccstor/aashka1/sentence_embeddings/output/NASA_finetuning_v2/Stagewise_stage1/NASA-WatBERT-300M-NASA-BioMedical-300K/2GPU/FinetuneNASA_pooling_mean_qlen64_plen256_freeze_embedding_true_lr_2e-5_bsz_500_100000steps/sent_embed.out -err /dccstor/aashka1/sentence_embeddings/output/NASA_finetuning_v2/Stagewise_stage1/NASA-WatBERT-300M-NASA-BioMedical-300K/2GPU/FinetuneNASA_pooling_mean_qlen64_plen256_freeze_embedding_true_lr_2e-5_bsz_500_100000steps/sent_embed.err PYTHONPATH=/dccstor/aashka1/onekd deepspeed --master_port 29519 /dccstor/aashka1/onekd/onekd/distillation/text_embeddings/simlm/distil_similarity_scores/train_simlm.py --deepspeed /dccstor/aashka1/onekd/onekd/distillation/text_embeddings/simlm/ds_config.json --model_name_or_path /dccstor/aashka1/sentence_embeddings/output/NASA_finetuning_v2/NASA-WatBERT-300M-NASA-BioMedical/2GPU/FinetuneNASA_pooling_mean_qlen64_plen256_freeze_embedding_true_lr_2e-5_bsz_500_300000steps/checkpoint-300000 --freeze_pos_emb true --pooling_source mean --per_device_train_batch_size 500 --per_device_eval_batch_size 32 --add_pooler False --t 0.02 --seed 42 --do_train --data_dir /dccstor/aashka1/sentence_embeddings/data/sent_embed_data/ --train_data_config //dccstor/phalanx/aashka/sentence-embeddings-NASA/configs/data_config_20M_NASA_stage1_nogooaq_nohotpot_nofever.json --validation_file /dccstor/retrieve-rerank2/data/embeddings-training-data/simlm/dev_w_docs.jsonl --fp16 --q_max_len 64 --p_max_len 256 --train_n_passages 1 --dataloader_num_workers 2 --num_train_epochs 1 --learning_rate 2e-5 --use_scaled_loss True --warmup_steps 1000 --share_encoder True --logging_steps 1000 --output_dir /dccstor/aashka1/sentence_embeddings/output/NASA_finetuning_v2/Stagewise_stage1/NASA-WatBERT-300M-NASA-BioMedical-300K/2GPU/FinetuneNASA_pooling_mean_qlen64_plen256_freeze_embedding_true_lr_2e-5_bsz_500_100000steps --save_total_limit 3 --save_strategy steps --save_steps 100000 --remove_unused_columns False --overwrite_output_dir --disable_tqdm False --max_steps 100000 --report_to none
+# bsub -q nonstandard -g /aashka/_/embeddings-kd -J NASA-ft -M 102400 -hl -n 8 -R "select[a100_80gb && hname!=cccxc607] rusage[mem=112640] span[ptile=8] affinity[core(1)]" -oo /dccstor/aashka1/sentence_embeddings/output/NASA_finetuning_v2/Stagewise_stage1/NASA-WatBERT-300M-NASA-BioMedical-300K/2GPU/FinetuneNASA_pooling_mean_qlen64_plen256_freeze_embedding_true_lr_2e-5_bsz_500_100000steps/sent_embed.out -eo /dccstor/aashka1/sentence_embeddings/output/NASA_finetuning_v2/Stagewise_stage1/NASA-WatBERT-300M-NASA-BioMedical-300K/2GPU/FinetuneNASA_pooling_mean_qlen64_plen256_freeze_embedding_true_lr_2e-5_bsz_500_100000steps/sent_embed.err -gpu  num=2:mode=exclusive_process PYTHONPATH=/dccstor/aashka1/onekd deepspeed --master_port 29519 /dccstor/aashka1/onekd/onekd/distillation/text_embeddings/simlm/distil_similarity_scores/train_simlm.py --deepspeed /dccstor/aashka1/onekd/onekd/distillation/text_embeddings/simlm/ds_config.json --model_name_or_path /dccstor/aashka1/sentence_embeddings/output/NASA_finetuning_v2/NASA-WatBERT-300M-NASA-BioMedical/2GPU/FinetuneNASA_pooling_mean_qlen64_plen256_freeze_embedding_true_lr_2e-5_bsz_500_300000steps/checkpoint-300000 --freeze_pos_emb true --pooling_source mean --per_device_train_batch_size 500 --per_device_eval_batch_size 32 --add_pooler False --t 0.02 --seed 42 --do_train --data_dir /dccstor/aashka1/sentence_embeddings/data/sent_embed_data/ --train_data_config //dccstor/phalanx/aashka/sentence-embeddings-NASA/configs/data_config_20M_NASA_stage1_nogooaq_nohotpot_nofever.json --validation_file /dccstor/retrieve-rerank2/data/embeddings-training-data/simlm/dev_w_docs.jsonl --fp16 --q_max_len 64 --p_max_len 256 --train_n_passages 1 --dataloader_num_workers 2 --num_train_epochs 1 --learning_rate 2e-5 --use_scaled_loss True --warmup_steps 1000 --share_encoder True --logging_steps 1000 --output_dir /dccstor/aashka1/sentence_embeddings/output/NASA_finetuning_v2/Stagewise_stage1/NASA-WatBERT-300M-NASA-BioMedical-300K/2GPU/FinetuneNASA_pooling_mean_qlen64_plen256_freeze_embedding_true_lr_2e-5_bsz_500_100000steps --save_total_limit 3 --save_strategy steps --save_steps 100000 --remove_unused_columns False --overwrite_output_dir --disable_tqdm False --max_steps 100000 --report_to none
+Job <722256> is submitted to queue <nonstandard>.

modules.json

@@ -0,0 +1,20 @@
+[
+  {
+    "idx": 0,
+    "name": "0",
+    "path": "",
+    "type": "sentence_transformers.models.Transformer"
+  },
+  {
+    "idx": 1,
+    "name": "1",
+    "path": "1_Pooling",
+    "type": "sentence_transformers.models.Pooling"
+  },
+  {
+    "idx": 2,
+    "name": "2",
+    "path": "2_Normalize",
+    "type": "sentence_transformers.models.Normalize"
+  }
+]

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0143a266f7ab79c9490c39331c7c39f6ca4c1e4f9519bfd738de18e53e64722d
+size 249322753

sentence_bert_config.json

@@ -0,0 +1,4 @@
+{
+  "max_seq_length": 512,
+  "do_lower_case": false
+}

special_tokens_map.json

@@ -0,0 +1,15 @@
+{
+  "bos_token": "<s>",
+  "cls_token": "<s>",
+  "eos_token": "</s>",
+  "mask_token": {
+    "content": "<mask>",
+    "lstrip": true,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "<pad>",
+  "sep_token": "</s>",
+  "unk_token": "<unk>"
+}

tokenizer_config.json

@@ -0,0 +1,15 @@
+{
+  "add_prefix_space": false,
+  "bos_token": "<s>",
+  "clean_up_tokenization_spaces": true,
+  "cls_token": "<s>",
+  "eos_token": "</s>",
+  "errors": "replace",
+  "mask_token": "<mask>",
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "<pad>",
+  "sep_token": "</s>",
+  "tokenizer_class": "RobertaTokenizer",
+  "trim_offsets": true,
+  "unk_token": "<unk>"
+}

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 1.0,
+    "train_loss": 1.4607645520019532,
+    "train_runtime": 81957.8858,
+    "train_samples": 0,
+    "train_samples_per_second": 1220.139,
+    "train_steps_per_second": 1.22
+}