Training in progress, epoch 16, checkpoint

checkpoint-1648/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-1648/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-1648/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5895dda241f8ec09f872aef3ef9535eb504b2e4dee1c45ba2105d196b9cd6ab8
+size 392536704

checkpoint-1648/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5194847bd4b26502112ccb45936f333b4d9a4bca78a304de2aa88c2987bfad29
+size 785167627

checkpoint-1648/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:159f6612f388c49fda6df5947f08c04939583c57708e9f83d672038ae6189851
+size 14709

checkpoint-1648/scaler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c908b2e1d5cb79baaa46db130df115ab812c125e1549314b7690a35d0a0df673
+size 1383

checkpoint-1648/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:20fbe76287c8205a5822de7b5bc92c045bf00c866efc1a9520592179be2f3b67
+size 1465

checkpoint-1648/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-1648/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-1648/trainer_state.json

@@ -0,0 +1,281 @@
+{
+  "best_global_step": 1339,
+  "best_metric": 5.512476444244385,
+  "best_model_checkpoint": "/home/p318482/syntactic-bootstrapping/model_trained/ads_sh1_fr_13/checkpoint-1339",
+  "epoch": 16.0,
+  "eval_steps": 500,
+  "global_step": 1648,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.009708737864077669,
+      "grad_norm": 16.858198165893555,
+      "learning_rate": 0.0,
+      "loss": 9.7487,
+      "step": 1
+    },
+    {
+      "epoch": 1.0,
+      "grad_norm": 1.9117469787597656,
+      "learning_rate": 2.04e-05,
+      "loss": 8.0282,
+      "step": 103
+    },
+    {
+      "epoch": 1.0,
+      "eval_loss": 7.0722832679748535,
+      "eval_runtime": 3.2339,
+      "eval_samples_per_second": 791.617,
+      "eval_steps_per_second": 3.092,
+      "step": 103
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 0.7483371496200562,
+      "learning_rate": 4.1e-05,
+      "loss": 6.2183,
+      "step": 206
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 5.94571590423584,
+      "eval_runtime": 3.1682,
+      "eval_samples_per_second": 808.025,
+      "eval_steps_per_second": 3.156,
+      "step": 206
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 0.9432477951049805,
+      "learning_rate": 6.16e-05,
+      "loss": 5.7981,
+      "step": 309
+    },
+    {
+      "epoch": 3.0,
+      "eval_loss": 5.829646587371826,
+      "eval_runtime": 3.2539,
+      "eval_samples_per_second": 786.746,
+      "eval_steps_per_second": 3.073,
+      "step": 309
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 0.8913167715072632,
+      "learning_rate": 8.22e-05,
+      "loss": 5.7019,
+      "step": 412
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 5.740334987640381,
+      "eval_runtime": 3.2715,
+      "eval_samples_per_second": 782.514,
+      "eval_steps_per_second": 3.057,
+      "step": 412
+    },
+    {
+      "epoch": 5.0,
+      "grad_norm": 0.8782764077186584,
+      "learning_rate": 9.909090909090911e-05,
+      "loss": 5.617,
+      "step": 515
+    },
+    {
+      "epoch": 5.0,
+      "eval_loss": 5.672503471374512,
+      "eval_runtime": 3.2114,
+      "eval_samples_per_second": 797.148,
+      "eval_steps_per_second": 3.114,
+      "step": 515
+    },
+    {
+      "epoch": 6.0,
+      "grad_norm": 0.9064059853553772,
+      "learning_rate": 9.24025974025974e-05,
+      "loss": 5.5455,
+      "step": 618
+    },
+    {
+      "epoch": 6.0,
+      "eval_loss": 5.621543884277344,
+      "eval_runtime": 3.6511,
+      "eval_samples_per_second": 701.15,
+      "eval_steps_per_second": 2.739,
+      "step": 618
+    },
+    {
+      "epoch": 7.0,
+      "grad_norm": 0.9032197594642639,
+      "learning_rate": 8.571428571428571e-05,
+      "loss": 5.4893,
+      "step": 721
+    },
+    {
+      "epoch": 7.0,
+      "eval_loss": 5.587393283843994,
+      "eval_runtime": 3.1849,
+      "eval_samples_per_second": 803.801,
+      "eval_steps_per_second": 3.14,
+      "step": 721
+    },
+    {
+      "epoch": 8.0,
+      "grad_norm": 1.1316813230514526,
+      "learning_rate": 7.902597402597402e-05,
+      "loss": 5.4408,
+      "step": 824
+    },
+    {
+      "epoch": 8.0,
+      "eval_loss": 5.563743591308594,
+      "eval_runtime": 3.1849,
+      "eval_samples_per_second": 803.796,
+      "eval_steps_per_second": 3.14,
+      "step": 824
+    },
+    {
+      "epoch": 9.0,
+      "grad_norm": 1.0942347049713135,
+      "learning_rate": 7.233766233766233e-05,
+      "loss": 5.3968,
+      "step": 927
+    },
+    {
+      "epoch": 9.0,
+      "eval_loss": 5.543261528015137,
+      "eval_runtime": 3.1862,
+      "eval_samples_per_second": 803.458,
+      "eval_steps_per_second": 3.139,
+      "step": 927
+    },
+    {
+      "epoch": 10.0,
+      "grad_norm": 1.0711631774902344,
+      "learning_rate": 6.564935064935066e-05,
+      "loss": 5.3536,
+      "step": 1030
+    },
+    {
+      "epoch": 10.0,
+      "eval_loss": 5.528441429138184,
+      "eval_runtime": 3.1759,
+      "eval_samples_per_second": 806.079,
+      "eval_steps_per_second": 3.149,
+      "step": 1030
+    },
+    {
+      "epoch": 11.0,
+      "grad_norm": 1.2122929096221924,
+      "learning_rate": 5.8961038961038965e-05,
+      "loss": 5.3114,
+      "step": 1133
+    },
+    {
+      "epoch": 11.0,
+      "eval_loss": 5.518833160400391,
+      "eval_runtime": 3.1786,
+      "eval_samples_per_second": 805.397,
+      "eval_steps_per_second": 3.146,
+      "step": 1133
+    },
+    {
+      "epoch": 12.0,
+      "grad_norm": 1.308700442314148,
+      "learning_rate": 5.2272727272727274e-05,
+      "loss": 5.268,
+      "step": 1236
+    },
+    {
+      "epoch": 12.0,
+      "eval_loss": 5.516317844390869,
+      "eval_runtime": 3.1957,
+      "eval_samples_per_second": 801.08,
+      "eval_steps_per_second": 3.129,
+      "step": 1236
+    },
+    {
+      "epoch": 13.0,
+      "grad_norm": 1.3491945266723633,
+      "learning_rate": 4.5584415584415584e-05,
+      "loss": 5.2246,
+      "step": 1339
+    },
+    {
+      "epoch": 13.0,
+      "eval_loss": 5.512476444244385,
+      "eval_runtime": 3.4806,
+      "eval_samples_per_second": 735.507,
+      "eval_steps_per_second": 2.873,
+      "step": 1339
+    },
+    {
+      "epoch": 14.0,
+      "grad_norm": 1.4462554454803467,
+      "learning_rate": 3.88961038961039e-05,
+      "loss": 5.1809,
+      "step": 1442
+    },
+    {
+      "epoch": 14.0,
+      "eval_loss": 5.516213893890381,
+      "eval_runtime": 3.2146,
+      "eval_samples_per_second": 796.378,
+      "eval_steps_per_second": 3.111,
+      "step": 1442
+    },
+    {
+      "epoch": 15.0,
+      "grad_norm": 1.5322386026382446,
+      "learning_rate": 3.220779220779221e-05,
+      "loss": 5.1367,
+      "step": 1545
+    },
+    {
+      "epoch": 15.0,
+      "eval_loss": 5.516896724700928,
+      "eval_runtime": 3.297,
+      "eval_samples_per_second": 776.452,
+      "eval_steps_per_second": 3.033,
+      "step": 1545
+    },
+    {
+      "epoch": 16.0,
+      "grad_norm": 1.6749299764633179,
+      "learning_rate": 2.551948051948052e-05,
+      "loss": 5.0935,
+      "step": 1648
+    },
+    {
+      "epoch": 16.0,
+      "eval_loss": 5.523487091064453,
+      "eval_runtime": 3.1982,
+      "eval_samples_per_second": 800.448,
+      "eval_steps_per_second": 3.127,
+      "step": 1648
+    }
+  ],
+  "logging_steps": 500,
+  "max_steps": 2060,
+  "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": 2.7368772599808e+16,
+  "train_batch_size": 256,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-1648/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c6ea9dd4549136ce0710957e49c4d2b5ec11bfb6eb9e92285dc4d43b109e6b78
+size 5905