Upload folder using huggingface_hub

checkpoint-3000/config.json

@@ -0,0 +1,33 @@
+{
+  "_name_or_path": "/data/tir/projects/tir7/user_data/priyansk/qa_tapex_e2_codet5p-220m_latex",
+  "architectures": [
+    "T5ForConditionalGeneration"
+  ],
+  "bos_token_id": 1,
+  "d_ff": 3072,
+  "d_kv": 64,
+  "d_model": 768,
+  "decoder_start_token_id": 0,
+  "dense_act_fn": "relu",
+  "dropout_rate": 0.1,
+  "eos_token_id": 2,
+  "feed_forward_proj": "relu",
+  "initializer_factor": 1.0,
+  "is_encoder_decoder": true,
+  "is_gated_act": false,
+  "layer_norm_epsilon": 1e-06,
+  "max_length": 1024,
+  "model_type": "t5",
+  "n_positions": 512,
+  "num_decoder_layers": 12,
+  "num_heads": 12,
+  "num_layers": 12,
+  "output_past": true,
+  "pad_token_id": 0,
+  "relative_attention_max_distance": 128,
+  "relative_attention_num_buckets": 32,
+  "torch_dtype": "float32",
+  "transformers_version": "4.17.0",
+  "use_cache": true,
+  "vocab_size": 32100
+}

checkpoint-3000/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fd7e384b56b22431e6b1c1acff02ebd103e36a115d98c4efef1cea6c65d42fcd
+size 1783209146

checkpoint-3000/pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3e7d131b3ca61486ecfd251adf9bd85048186ffb038ab9f0028a18da26462534
+size 891647438

checkpoint-3000/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e82f20000c71859c9df907e3208e084f57fb73e976cbdc565209290b32b1eb88
+size 15006

checkpoint-3000/scaler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:21aba8ed0f38ed1c04994c10a9ca7e9925e55ef2ed51283c43ff8e2cce78585f
+size 988

checkpoint-3000/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:78bd29be37d9912f725f9d3d44533499f3f0ec14930d68bd3fd086c6f4e170be
+size 1064

checkpoint-3000/special_tokens_map.json

@@ -0,0 +1 @@
+{"bos_token": {"content": "<s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true, "special": true}, "eos_token": {"content": "</s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true, "special": true}, "unk_token": {"content": "<unk>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true, "special": true}, "sep_token": {"content": "</s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true, "special": true}, "pad_token": {"content": "<pad>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true, "special": true}, "cls_token": {"content": "<s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true, "special": true}, "mask_token": {"content": "<mask>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, "additional_special_tokens": [{"content": "<extra_id_99>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_98>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_97>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_96>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_95>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_94>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_93>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_92>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_91>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_90>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_89>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_88>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_87>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_86>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_85>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_84>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_83>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_82>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_81>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_80>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_79>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_78>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_77>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_76>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_75>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_74>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_73>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_72>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_71>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_70>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_69>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_68>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_67>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_66>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_65>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_64>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_63>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_62>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_61>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_60>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_59>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_58>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_57>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_56>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_55>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_54>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_53>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_52>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_51>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_50>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_49>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_48>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_47>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_46>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_45>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_44>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_43>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_42>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_41>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_40>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_39>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_38>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_37>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_36>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_35>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_34>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_33>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_32>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_31>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_30>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_29>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_28>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_27>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_26>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_25>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_24>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_23>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_22>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_21>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_20>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_19>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_18>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_17>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_16>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_15>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_14>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_13>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_12>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_11>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_10>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_9>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_8>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_7>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_6>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_5>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_4>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_3>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_2>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_1>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_0>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}]}

checkpoint-3000/tokenizer_config.json

@@ -0,0 +1 @@
+{"errors": "replace", "bos_token": {"content": "<s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true, "special": false, "__type": "AddedToken"}, "eos_token": {"content": "</s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true, "special": false, "__type": "AddedToken"}, "sep_token": {"content": "</s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true, "special": false, "__type": "AddedToken"}, "cls_token": {"content": "<s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true, "special": false, "__type": "AddedToken"}, "unk_token": {"content": "<unk>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true, "special": false, "__type": "AddedToken"}, "pad_token": {"content": "<pad>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true, "special": false, "__type": "AddedToken"}, "mask_token": {"content": "<mask>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": false, "__type": "AddedToken"}, "add_prefix_space": false, "trim_offsets": true, "model_max_length": 512, "special_tokens_map_file": "/data/datasets/hf_cache/transformers/f432e4eb4a7dfc04b533beea5590e11f4b46c86f5630e8a032704ef76d7269c1.b9905d0575bde443a20834122b6e2d48e853b2e36444ce98ddeb43c38097eb3f", "name_or_path": "/data/tir/projects/tir7/user_data/priyansk/qa_tapex_e2_codet5p-220m_latex", "tokenizer_class": "RobertaTokenizer"}

checkpoint-3000/trainer_state.json

@@ -0,0 +1,166 @@
+{
+  "best_metric": 37.6894,
+  "best_model_checkpoint": "/data/tir/projects/tir7/user_data/priyansk/qa_s2_codet5p-220m_latex/checkpoint-2700",
+  "epoch": 25.209095661265028,
+  "global_step": 3000,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 2.52,
+      "learning_rate": 9.160504201680673e-05,
+      "loss": 0.0422,
+      "step": 300
+    },
+    {
+      "epoch": 2.52,
+      "eval_exact_match": 22.1591,
+      "eval_loss": 0.09354351460933685,
+      "eval_runtime": 890.466,
+      "eval_samples_per_second": 0.602,
+      "eval_steps_per_second": 0.038,
+      "step": 300
+    },
+    {
+      "epoch": 5.04,
+      "learning_rate": 8.321008403361344e-05,
+      "loss": 0.0266,
+      "step": 600
+    },
+    {
+      "epoch": 5.04,
+      "eval_exact_match": 28.0303,
+      "eval_loss": 0.08916985988616943,
+      "eval_runtime": 876.6244,
+      "eval_samples_per_second": 0.611,
+      "eval_steps_per_second": 0.039,
+      "step": 600
+    },
+    {
+      "epoch": 7.56,
+      "learning_rate": 7.481512605042018e-05,
+      "loss": 0.0216,
+      "step": 900
+    },
+    {
+      "epoch": 7.56,
+      "eval_exact_match": 30.1136,
+      "eval_loss": 0.08875476568937302,
+      "eval_runtime": 676.1858,
+      "eval_samples_per_second": 0.793,
+      "eval_steps_per_second": 0.05,
+      "step": 900
+    },
+    {
+      "epoch": 10.08,
+      "learning_rate": 6.642016806722689e-05,
+      "loss": 0.0187,
+      "step": 1200
+    },
+    {
+      "epoch": 10.08,
+      "eval_exact_match": 35.0379,
+      "eval_loss": 0.08860571682453156,
+      "eval_runtime": 717.1877,
+      "eval_samples_per_second": 0.747,
+      "eval_steps_per_second": 0.047,
+      "step": 1200
+    },
+    {
+      "epoch": 12.6,
+      "learning_rate": 5.802521008403361e-05,
+      "loss": 0.0165,
+      "step": 1500
+    },
+    {
+      "epoch": 12.6,
+      "eval_exact_match": 33.7121,
+      "eval_loss": 0.09030993282794952,
+      "eval_runtime": 719.2199,
+      "eval_samples_per_second": 0.745,
+      "eval_steps_per_second": 0.047,
+      "step": 1500
+    },
+    {
+      "epoch": 15.13,
+      "learning_rate": 4.963025210084034e-05,
+      "loss": 0.0149,
+      "step": 1800
+    },
+    {
+      "epoch": 15.13,
+      "eval_exact_match": 36.553,
+      "eval_loss": 0.09047655761241913,
+      "eval_runtime": 694.4828,
+      "eval_samples_per_second": 0.772,
+      "eval_steps_per_second": 0.049,
+      "step": 1800
+    },
+    {
+      "epoch": 17.64,
+      "learning_rate": 4.123529411764706e-05,
+      "loss": 0.0136,
+      "step": 2100
+    },
+    {
+      "epoch": 17.64,
+      "eval_exact_match": 35.7955,
+      "eval_loss": 0.09109079837799072,
+      "eval_runtime": 751.6054,
+      "eval_samples_per_second": 0.713,
+      "eval_steps_per_second": 0.045,
+      "step": 2100
+    },
+    {
+      "epoch": 20.17,
+      "learning_rate": 3.284033613445378e-05,
+      "loss": 0.0126,
+      "step": 2400
+    },
+    {
+      "epoch": 20.17,
+      "eval_exact_match": 37.5,
+      "eval_loss": 0.0928933173418045,
+      "eval_runtime": 702.9176,
+      "eval_samples_per_second": 0.763,
+      "eval_steps_per_second": 0.048,
+      "step": 2400
+    },
+    {
+      "epoch": 22.69,
+      "learning_rate": 2.4445378151260506e-05,
+      "loss": 0.0118,
+      "step": 2700
+    },
+    {
+      "epoch": 22.69,
+      "eval_exact_match": 37.6894,
+      "eval_loss": 0.0941813662648201,
+      "eval_runtime": 681.194,
+      "eval_samples_per_second": 0.787,
+      "eval_steps_per_second": 0.05,
+      "step": 2700
+    },
+    {
+      "epoch": 25.21,
+      "learning_rate": 1.6050420168067226e-05,
+      "loss": 0.0112,
+      "step": 3000
+    },
+    {
+      "epoch": 25.21,
+      "eval_exact_match": 37.6894,
+      "eval_loss": 0.09540851414203644,
+      "eval_runtime": 647.9067,
+      "eval_samples_per_second": 0.827,
+      "eval_steps_per_second": 0.052,
+      "step": 3000
+    }
+  ],
+  "max_steps": 3570,
+  "num_train_epochs": 30,
+  "total_flos": 3.758975266848768e+18,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-3000/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2e5aae40c9e4657635d146f9220c617fe930d482b204d32c52af99b4e343d902
+size 3704

checkpoint-3300/config.json

@@ -0,0 +1,33 @@
+{
+  "_name_or_path": "/data/tir/projects/tir7/user_data/priyansk/qa_tapex_e2_codet5p-220m_latex",
+  "architectures": [
+    "T5ForConditionalGeneration"
+  ],
+  "bos_token_id": 1,
+  "d_ff": 3072,
+  "d_kv": 64,
+  "d_model": 768,
+  "decoder_start_token_id": 0,
+  "dense_act_fn": "relu",
+  "dropout_rate": 0.1,
+  "eos_token_id": 2,
+  "feed_forward_proj": "relu",
+  "initializer_factor": 1.0,
+  "is_encoder_decoder": true,
+  "is_gated_act": false,
+  "layer_norm_epsilon": 1e-06,
+  "max_length": 1024,
+  "model_type": "t5",
+  "n_positions": 512,
+  "num_decoder_layers": 12,
+  "num_heads": 12,
+  "num_layers": 12,
+  "output_past": true,
+  "pad_token_id": 0,
+  "relative_attention_max_distance": 128,
+  "relative_attention_num_buckets": 32,
+  "torch_dtype": "float32",
+  "transformers_version": "4.17.0",
+  "use_cache": true,
+  "vocab_size": 32100
+}

checkpoint-3300/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:364929d088f682473d9e9e7bfa662ebe252476f5bd9aa8a8dd4b44b27a8bcfbd
+size 1783209146

checkpoint-3300/pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:88453f89799c2e36c1ffbf7cd26c453f1f74370a866a78f74f08cf79bc1e48f8
+size 891647438

checkpoint-3300/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4138b0cc45f9aaea45e77468351e17cc832e5ea03d6f6171b526fc0915098902
+size 15006

checkpoint-3300/scaler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d02c0cc824d9af81c56696979c838a9c6abecfb71e177390b5106abcdcd6d800
+size 988

checkpoint-3300/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8dbf675ee7b373f5623c07f17825829a12bf8e47b8d82e2d95f97efa899b7fcc
+size 1064

checkpoint-3300/special_tokens_map.json

@@ -0,0 +1 @@
+{"bos_token": {"content": "<s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true, "special": true}, "eos_token": {"content": "</s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true, "special": true}, "unk_token": {"content": "<unk>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true, "special": true}, "sep_token": {"content": "</s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true, "special": true}, "pad_token": {"content": "<pad>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true, "special": true}, "cls_token": {"content": "<s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true, "special": true}, "mask_token": {"content": "<mask>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, "additional_special_tokens": [{"content": "<extra_id_99>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_98>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_97>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_96>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_95>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_94>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_93>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_92>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_91>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_90>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_89>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_88>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_87>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_86>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_85>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_84>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_83>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_82>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_81>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_80>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_79>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_78>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_77>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_76>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_75>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_74>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_73>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_72>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_71>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_70>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_69>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_68>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_67>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_66>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_65>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_64>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_63>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_62>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_61>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_60>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_59>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_58>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_57>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_56>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_55>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_54>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_53>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_52>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_51>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_50>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_49>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_48>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_47>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_46>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_45>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_44>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_43>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_42>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_41>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_40>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_39>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_38>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_37>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_36>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_35>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_34>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_33>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_32>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_31>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_30>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_29>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_28>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_27>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_26>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_25>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_24>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_23>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_22>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_21>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_20>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_19>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_18>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_17>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_16>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_15>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_14>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_13>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_12>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_11>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_10>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_9>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_8>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_7>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_6>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_5>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_4>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_3>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_2>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_1>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}, {"content": "<extra_id_0>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": true}]}

checkpoint-3300/tokenizer_config.json

@@ -0,0 +1 @@
+{"errors": "replace", "bos_token": {"content": "<s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true, "special": false, "__type": "AddedToken"}, "eos_token": {"content": "</s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true, "special": false, "__type": "AddedToken"}, "sep_token": {"content": "</s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true, "special": false, "__type": "AddedToken"}, "cls_token": {"content": "<s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true, "special": false, "__type": "AddedToken"}, "unk_token": {"content": "<unk>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true, "special": false, "__type": "AddedToken"}, "pad_token": {"content": "<pad>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true, "special": false, "__type": "AddedToken"}, "mask_token": {"content": "<mask>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "special": false, "__type": "AddedToken"}, "add_prefix_space": false, "trim_offsets": true, "model_max_length": 512, "special_tokens_map_file": "/data/datasets/hf_cache/transformers/f432e4eb4a7dfc04b533beea5590e11f4b46c86f5630e8a032704ef76d7269c1.b9905d0575bde443a20834122b6e2d48e853b2e36444ce98ddeb43c38097eb3f", "name_or_path": "/data/tir/projects/tir7/user_data/priyansk/qa_tapex_e2_codet5p-220m_latex", "tokenizer_class": "RobertaTokenizer"}

checkpoint-3300/trainer_state.json

@@ -0,0 +1,181 @@
+{
+  "best_metric": 38.2576,
+  "best_model_checkpoint": "/data/tir/projects/tir7/user_data/priyansk/qa_s2_codet5p-220m_latex/checkpoint-3300",
+  "epoch": 27.7276529012023,
+  "global_step": 3300,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 2.52,
+      "learning_rate": 9.160504201680673e-05,
+      "loss": 0.0422,
+      "step": 300
+    },
+    {
+      "epoch": 2.52,
+      "eval_exact_match": 22.1591,
+      "eval_loss": 0.09354351460933685,
+      "eval_runtime": 890.466,
+      "eval_samples_per_second": 0.602,
+      "eval_steps_per_second": 0.038,
+      "step": 300
+    },
+    {
+      "epoch": 5.04,
+      "learning_rate": 8.321008403361344e-05,
+      "loss": 0.0266,
+      "step": 600
+    },
+    {
+      "epoch": 5.04,
+      "eval_exact_match": 28.0303,
+      "eval_loss": 0.08916985988616943,
+      "eval_runtime": 876.6244,
+      "eval_samples_per_second": 0.611,
+      "eval_steps_per_second": 0.039,
+      "step": 600
+    },
+    {
+      "epoch": 7.56,
+      "learning_rate": 7.481512605042018e-05,
+      "loss": 0.0216,
+      "step": 900
+    },
+    {
+      "epoch": 7.56,
+      "eval_exact_match": 30.1136,
+      "eval_loss": 0.08875476568937302,
+      "eval_runtime": 676.1858,
+      "eval_samples_per_second": 0.793,
+      "eval_steps_per_second": 0.05,
+      "step": 900
+    },
+    {
+      "epoch": 10.08,
+      "learning_rate": 6.642016806722689e-05,
+      "loss": 0.0187,
+      "step": 1200
+    },
+    {
+      "epoch": 10.08,
+      "eval_exact_match": 35.0379,
+      "eval_loss": 0.08860571682453156,
+      "eval_runtime": 717.1877,
+      "eval_samples_per_second": 0.747,
+      "eval_steps_per_second": 0.047,
+      "step": 1200
+    },
+    {
+      "epoch": 12.6,
+      "learning_rate": 5.802521008403361e-05,
+      "loss": 0.0165,
+      "step": 1500
+    },
+    {
+      "epoch": 12.6,
+      "eval_exact_match": 33.7121,
+      "eval_loss": 0.09030993282794952,
+      "eval_runtime": 719.2199,
+      "eval_samples_per_second": 0.745,
+      "eval_steps_per_second": 0.047,
+      "step": 1500
+    },
+    {
+      "epoch": 15.13,
+      "learning_rate": 4.963025210084034e-05,
+      "loss": 0.0149,
+      "step": 1800
+    },
+    {
+      "epoch": 15.13,
+      "eval_exact_match": 36.553,
+      "eval_loss": 0.09047655761241913,
+      "eval_runtime": 694.4828,
+      "eval_samples_per_second": 0.772,
+      "eval_steps_per_second": 0.049,
+      "step": 1800
+    },
+    {
+      "epoch": 17.64,
+      "learning_rate": 4.123529411764706e-05,
+      "loss": 0.0136,
+      "step": 2100
+    },
+    {
+      "epoch": 17.64,
+      "eval_exact_match": 35.7955,
+      "eval_loss": 0.09109079837799072,
+      "eval_runtime": 751.6054,
+      "eval_samples_per_second": 0.713,
+      "eval_steps_per_second": 0.045,
+      "step": 2100
+    },
+    {
+      "epoch": 20.17,
+      "learning_rate": 3.284033613445378e-05,
+      "loss": 0.0126,
+      "step": 2400
+    },
+    {
+      "epoch": 20.17,
+      "eval_exact_match": 37.5,
+      "eval_loss": 0.0928933173418045,
+      "eval_runtime": 702.9176,
+      "eval_samples_per_second": 0.763,
+      "eval_steps_per_second": 0.048,
+      "step": 2400
+    },
+    {
+      "epoch": 22.69,
+      "learning_rate": 2.4445378151260506e-05,
+      "loss": 0.0118,
+      "step": 2700
+    },
+    {
+      "epoch": 22.69,
+      "eval_exact_match": 37.6894,
+      "eval_loss": 0.0941813662648201,
+      "eval_runtime": 681.194,
+      "eval_samples_per_second": 0.787,
+      "eval_steps_per_second": 0.05,
+      "step": 2700
+    },
+    {
+      "epoch": 25.21,
+      "learning_rate": 1.6050420168067226e-05,
+      "loss": 0.0112,
+      "step": 3000
+    },
+    {
+      "epoch": 25.21,
+      "eval_exact_match": 37.6894,
+      "eval_loss": 0.09540851414203644,
+      "eval_runtime": 647.9067,
+      "eval_samples_per_second": 0.827,
+      "eval_steps_per_second": 0.052,
+      "step": 3000
+    },
+    {
+      "epoch": 27.73,
+      "learning_rate": 7.655462184873946e-06,
+      "loss": 0.0107,
+      "step": 3300
+    },
+    {
+      "epoch": 27.73,
+      "eval_exact_match": 38.2576,
+      "eval_loss": 0.09561759978532791,
+      "eval_runtime": 642.7497,
+      "eval_samples_per_second": 0.834,
+      "eval_steps_per_second": 0.053,
+      "step": 3300
+    }
+  ],
+  "max_steps": 3570,
+  "num_train_epochs": 30,
+  "total_flos": 4.134522033788682e+18,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-3300/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2e5aae40c9e4657635d146f9220c617fe930d482b204d32c52af99b4e343d902
+size 3704

runs/Apr08_18-34-17_babel-1-23/1712615663.1967785/events.out.tfevents.1712615663.babel-1-23

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4e27120a52c4b5342c5a9fc6830d91bb2e1e7497b1e13344a3122062fdad5f17
+size 4958

runs/Apr08_18-34-17_babel-1-23/events.out.tfevents.1712615663.babel-1-23

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7eef2bdce65a726b9ceeb7a7f422fe4bae688566ab6799b53dd203793a35d60a
+size 9087