Training in progress, epoch 17, checkpoint

checkpoint-1989/config.json

@@ -0,0 +1,33 @@
+{
+  "activation_function": "gelu_new",
+  "architectures": [
+    "GPT2LMHeadModel"
+  ],
+  "attn_pdrop": 0.1,
+  "bos_token_id": 1,
+  "dtype": "float32",
+  "embd_pdrop": 0.1,
+  "eos_token_id": 2,
+  "initializer_range": 0.02,
+  "layer_norm_epsilon": 1e-05,
+  "model_type": "gpt2",
+  "n_ctx": 1024,
+  "n_embd": 768,
+  "n_head": 12,
+  "n_inner": null,
+  "n_layer": 12,
+  "n_positions": 1024,
+  "pad_token_id": 2,
+  "reorder_and_upcast_attn": false,
+  "resid_pdrop": 0.1,
+  "scale_attn_by_inverse_layer_idx": false,
+  "scale_attn_weights": true,
+  "summary_activation": null,
+  "summary_first_dropout": 0.1,
+  "summary_proj_to_labels": true,
+  "summary_type": "cls_index",
+  "summary_use_proj": true,
+  "transformers_version": "4.56.1",
+  "use_cache": true,
+  "vocab_size": 16000
+}

checkpoint-1989/generation_config.json

@@ -0,0 +1,9 @@
+{
+  "_from_model_config": true,
+  "bos_token_id": 1,
+  "eos_token_id": [
+    2
+  ],
+  "pad_token_id": 2,
+  "transformers_version": "4.56.1"
+}

checkpoint-1989/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0f0cc2534a8ac5839d2d9f50025c42f86f4abc7bdad4a3605b357696baeb9a17
+size 392536704

checkpoint-1989/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cea752524b782427c4f1927ce4e86941851d74e1059fc3eb34fb977e33a80d80
+size 785167627

checkpoint-1989/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:81a270839780e66675553781435007c68042997165b273a299733309015102b4
+size 14709

checkpoint-1989/scaler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f24d8dcc07d2e7e81838884c11407ce60b9f1c47d7903db11e747e4b7c46c8c9
+size 1383

checkpoint-1989/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cbb23672ae3b0bf7dffec6781414a1eda3e2606bd513c9567ce0c485f6224d7e
+size 1465

checkpoint-1989/special_tokens_map.json

@@ -0,0 +1,8 @@
+{
+  "bos_token": "<s>",
+  "eos_token": "</s>",
+  "mask_token": "<mask>",
+  "pad_token": "</s>",
+  "sep_token": "<sep>",
+  "unk_token": "<unk>"
+}

checkpoint-1989/tokenizer_config.json

@@ -0,0 +1,62 @@
+{
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<pad>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "3": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "4": {
+      "content": "<mask>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "5": {
+      "content": "<sep>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "bos_token": "<s>",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "</s>",
+  "extra_special_tokens": {},
+  "mask_token": "<mask>",
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "</s>",
+  "sep_token": "<sep>",
+  "tokenizer_class": "PreTrainedTokenizerFast",
+  "unk_token": "<unk>"
+}

checkpoint-1989/trainer_state.json

@@ -0,0 +1,296 @@
+{
+  "best_global_step": 1989,
+  "best_metric": 3.8605313301086426,
+  "best_model_checkpoint": "/home/p318482/syntactic-bootstrapping/model_trained/adl_o_fr_13/checkpoint-1989",
+  "epoch": 17.0,
+  "eval_steps": 500,
+  "global_step": 1989,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.008547008547008548,
+      "grad_norm": 15.76026439666748,
+      "learning_rate": 0.0,
+      "loss": 9.782,
+      "step": 1
+    },
+    {
+      "epoch": 1.0,
+      "grad_norm": 1.2579153776168823,
+      "learning_rate": 2.32e-05,
+      "loss": 7.6799,
+      "step": 117
+    },
+    {
+      "epoch": 1.0,
+      "eval_loss": 6.5062150955200195,
+      "eval_runtime": 3.5136,
+      "eval_samples_per_second": 806.871,
+      "eval_steps_per_second": 3.415,
+      "step": 117
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 0.9850084185600281,
+      "learning_rate": 4.660000000000001e-05,
+      "loss": 5.8927,
+      "step": 234
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 5.511380672454834,
+      "eval_runtime": 3.5345,
+      "eval_samples_per_second": 802.096,
+      "eval_steps_per_second": 3.395,
+      "step": 234
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 1.2361372709274292,
+      "learning_rate": 7e-05,
+      "loss": 5.2809,
+      "step": 351
+    },
+    {
+      "epoch": 3.0,
+      "eval_loss": 5.0717010498046875,
+      "eval_runtime": 3.5917,
+      "eval_samples_per_second": 789.322,
+      "eval_steps_per_second": 3.341,
+      "step": 351
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 1.223827600479126,
+      "learning_rate": 9.340000000000001e-05,
+      "loss": 4.9132,
+      "step": 468
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 4.758579254150391,
+      "eval_runtime": 3.5226,
+      "eval_samples_per_second": 804.792,
+      "eval_steps_per_second": 3.407,
+      "step": 468
+    },
+    {
+      "epoch": 5.0,
+      "grad_norm": 1.2095705270767212,
+      "learning_rate": 9.53846153846154e-05,
+      "loss": 4.6357,
+      "step": 585
+    },
+    {
+      "epoch": 5.0,
+      "eval_loss": 4.519491195678711,
+      "eval_runtime": 3.5622,
+      "eval_samples_per_second": 795.858,
+      "eval_steps_per_second": 3.369,
+      "step": 585
+    },
+    {
+      "epoch": 6.0,
+      "grad_norm": 1.2360373735427856,
+      "learning_rate": 8.895604395604396e-05,
+      "loss": 4.4116,
+      "step": 702
+    },
+    {
+      "epoch": 6.0,
+      "eval_loss": 4.33683443069458,
+      "eval_runtime": 3.5156,
+      "eval_samples_per_second": 806.409,
+      "eval_steps_per_second": 3.413,
+      "step": 702
+    },
+    {
+      "epoch": 7.0,
+      "grad_norm": 1.260266661643982,
+      "learning_rate": 8.252747252747253e-05,
+      "loss": 4.238,
+      "step": 819
+    },
+    {
+      "epoch": 7.0,
+      "eval_loss": 4.204611301422119,
+      "eval_runtime": 3.7767,
+      "eval_samples_per_second": 750.66,
+      "eval_steps_per_second": 3.177,
+      "step": 819
+    },
+    {
+      "epoch": 8.0,
+      "grad_norm": 1.240233302116394,
+      "learning_rate": 7.60989010989011e-05,
+      "loss": 4.1018,
+      "step": 936
+    },
+    {
+      "epoch": 8.0,
+      "eval_loss": 4.109301567077637,
+      "eval_runtime": 3.7089,
+      "eval_samples_per_second": 764.375,
+      "eval_steps_per_second": 3.235,
+      "step": 936
+    },
+    {
+      "epoch": 9.0,
+      "grad_norm": 1.2881073951721191,
+      "learning_rate": 6.967032967032967e-05,
+      "loss": 3.9923,
+      "step": 1053
+    },
+    {
+      "epoch": 9.0,
+      "eval_loss": 4.043199062347412,
+      "eval_runtime": 3.7053,
+      "eval_samples_per_second": 765.121,
+      "eval_steps_per_second": 3.239,
+      "step": 1053
+    },
+    {
+      "epoch": 10.0,
+      "grad_norm": 1.3990460634231567,
+      "learning_rate": 6.324175824175824e-05,
+      "loss": 3.8977,
+      "step": 1170
+    },
+    {
+      "epoch": 10.0,
+      "eval_loss": 3.9879415035247803,
+      "eval_runtime": 3.6744,
+      "eval_samples_per_second": 771.554,
+      "eval_steps_per_second": 3.266,
+      "step": 1170
+    },
+    {
+      "epoch": 11.0,
+      "grad_norm": 1.4328550100326538,
+      "learning_rate": 5.6813186813186814e-05,
+      "loss": 3.816,
+      "step": 1287
+    },
+    {
+      "epoch": 11.0,
+      "eval_loss": 3.9500930309295654,
+      "eval_runtime": 3.5602,
+      "eval_samples_per_second": 796.308,
+      "eval_steps_per_second": 3.371,
+      "step": 1287
+    },
+    {
+      "epoch": 12.0,
+      "grad_norm": 1.544657588005066,
+      "learning_rate": 5.038461538461539e-05,
+      "loss": 3.744,
+      "step": 1404
+    },
+    {
+      "epoch": 12.0,
+      "eval_loss": 3.9190826416015625,
+      "eval_runtime": 3.5225,
+      "eval_samples_per_second": 804.83,
+      "eval_steps_per_second": 3.407,
+      "step": 1404
+    },
+    {
+      "epoch": 13.0,
+      "grad_norm": 1.545511245727539,
+      "learning_rate": 4.3956043956043955e-05,
+      "loss": 3.6797,
+      "step": 1521
+    },
+    {
+      "epoch": 13.0,
+      "eval_loss": 3.8957431316375732,
+      "eval_runtime": 4.1595,
+      "eval_samples_per_second": 681.572,
+      "eval_steps_per_second": 2.885,
+      "step": 1521
+    },
+    {
+      "epoch": 14.0,
+      "grad_norm": 1.6040838956832886,
+      "learning_rate": 3.752747252747253e-05,
+      "loss": 3.6209,
+      "step": 1638
+    },
+    {
+      "epoch": 14.0,
+      "eval_loss": 3.8795065879821777,
+      "eval_runtime": 3.5163,
+      "eval_samples_per_second": 806.24,
+      "eval_steps_per_second": 3.413,
+      "step": 1638
+    },
+    {
+      "epoch": 15.0,
+      "grad_norm": 1.714130163192749,
+      "learning_rate": 3.10989010989011e-05,
+      "loss": 3.5691,
+      "step": 1755
+    },
+    {
+      "epoch": 15.0,
+      "eval_loss": 3.870738983154297,
+      "eval_runtime": 3.5807,
+      "eval_samples_per_second": 791.746,
+      "eval_steps_per_second": 3.351,
+      "step": 1755
+    },
+    {
+      "epoch": 16.0,
+      "grad_norm": 1.7997115850448608,
+      "learning_rate": 2.4670329670329673e-05,
+      "loss": 3.5224,
+      "step": 1872
+    },
+    {
+      "epoch": 16.0,
+      "eval_loss": 3.8642091751098633,
+      "eval_runtime": 3.5595,
+      "eval_samples_per_second": 796.459,
+      "eval_steps_per_second": 3.371,
+      "step": 1872
+    },
+    {
+      "epoch": 17.0,
+      "grad_norm": 1.8778117895126343,
+      "learning_rate": 1.824175824175824e-05,
+      "loss": 3.4816,
+      "step": 1989
+    },
+    {
+      "epoch": 17.0,
+      "eval_loss": 3.8605313301086426,
+      "eval_runtime": 3.7382,
+      "eval_samples_per_second": 758.379,
+      "eval_steps_per_second": 3.21,
+      "step": 1989
+    }
+  ],
+  "logging_steps": 500,
+  "max_steps": 2340,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 20,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 3.3123729604608e+16,
+  "train_batch_size": 256,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-1989/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:498692f4a859e8aaa0a4227ffca730af463b7fef8283fed1a31c7ad56d195f9e
+size 5905