Training in progress, step 11000, checkpoint

checkpoint-11000/config.json

@@ -0,0 +1,39 @@
+{
+  "_name_or_path": "gpt2",
+  "activation_function": "gelu_new",
+  "architectures": [
+    "GPT2LMHeadModel"
+  ],
+  "attn_pdrop": 0.1,
+  "bos_token_id": 5,
+  "embd_pdrop": 0.1,
+  "eos_token_id": 4,
+  "initializer_range": 0.02,
+  "layer_norm_epsilon": 1e-05,
+  "model_type": "gpt2",
+  "n_ctx": 1024,
+  "n_embd": 128,
+  "n_head": 4,
+  "n_inner": null,
+  "n_layer": 4,
+  "n_positions": 128,
+  "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,
+  "task_specific_params": {
+    "text-generation": {
+      "do_sample": true,
+      "max_length": 50
+    }
+  },
+  "torch_dtype": "float32",
+  "transformers_version": "4.45.2",
+  "use_cache": true,
+  "vocab_size": 8192
+}

checkpoint-11000/generation_config.json

@@ -0,0 +1,6 @@
+{
+  "_from_model_config": true,
+  "bos_token_id": 5,
+  "eos_token_id": 4,
+  "transformers_version": "4.45.2"
+}

checkpoint-11000/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8906a28ee6ac31b331b028b1cb394ed182095f33356ca52836de1015c02bd193
+size 7438256

checkpoint-11000/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cff43b232a216ec5158ec0b220c073537228b9018ade82ffc2bfee6d33e89e1b
+size 14910010

checkpoint-11000/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c98800bd7329d283daf5e4549c9ea6183cb7d7e41949d133fbf4e44b686013ce
+size 14244

checkpoint-11000/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9363d58c6e301b6ab42d2fc44778c095c88e2b1d11cddf1431b5fa9a8541e488
+size 1192

checkpoint-11000/special_tokens_map.json

@@ -0,0 +1,17 @@
+{
+  "bos_token": {
+    "content": "[BOS]",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "[EOS]",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "[EOS]"
+}

checkpoint-11000/tokenizer_config.json

@@ -0,0 +1,58 @@
+{
+  "added_tokens_decoder": {
+    "0": {
+      "content": "[UNK]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "[CLS]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "[SEP]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "3": {
+      "content": "[PAD]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "4": {
+      "content": "[EOS]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "5": {
+      "content": "[BOS]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "bos_token": "[BOS]",
+  "clean_up_tokenization_spaces": true,
+  "eos_token": "[EOS]",
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "[EOS]",
+  "tokenizer_class": "PreTrainedTokenizerFast"
+}

checkpoint-11000/trainer_state.json

@@ -0,0 +1,284 @@
+{
+  "best_metric": 3.5645763874053955,
+  "best_model_checkpoint": "/home/p318482/babyLM_controlled/models_trained/reduce_on_plateau_convergence/random_transcript_conv/checkpoint-11000",
+  "epoch": 0.2793367023032581,
+  "eval_steps": 1000,
+  "global_step": 11000,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.012697122831966277,
+      "grad_norm": 0.7364352941513062,
+      "learning_rate": 0.0001,
+      "loss": 5.9868,
+      "step": 500
+    },
+    {
+      "epoch": 0.025394245663932554,
+      "grad_norm": 1.3591667413711548,
+      "learning_rate": 0.0001,
+      "loss": 4.7264,
+      "step": 1000
+    },
+    {
+      "epoch": 0.025394245663932554,
+      "eval_loss": 4.496066570281982,
+      "eval_runtime": 1.8776,
+      "eval_samples_per_second": 932.025,
+      "eval_steps_per_second": 29.292,
+      "step": 1000
+    },
+    {
+      "epoch": 0.03809136849589883,
+      "grad_norm": 0.9609436988830566,
+      "learning_rate": 0.0001,
+      "loss": 4.4273,
+      "step": 1500
+    },
+    {
+      "epoch": 0.05078849132786511,
+      "grad_norm": 1.208341360092163,
+      "learning_rate": 0.0001,
+      "loss": 4.2766,
+      "step": 2000
+    },
+    {
+      "epoch": 0.05078849132786511,
+      "eval_loss": 4.21427583694458,
+      "eval_runtime": 1.8912,
+      "eval_samples_per_second": 925.334,
+      "eval_steps_per_second": 29.082,
+      "step": 2000
+    },
+    {
+      "epoch": 0.06348561415983138,
+      "grad_norm": 1.4926683902740479,
+      "learning_rate": 0.0001,
+      "loss": 4.2317,
+      "step": 2500
+    },
+    {
+      "epoch": 0.07618273699179766,
+      "grad_norm": 1.3337597846984863,
+      "learning_rate": 0.0001,
+      "loss": 4.1231,
+      "step": 3000
+    },
+    {
+      "epoch": 0.07618273699179766,
+      "eval_loss": 4.032348155975342,
+      "eval_runtime": 1.8734,
+      "eval_samples_per_second": 934.155,
+      "eval_steps_per_second": 29.359,
+      "step": 3000
+    },
+    {
+      "epoch": 0.08887985982376394,
+      "grad_norm": 1.7601583003997803,
+      "learning_rate": 0.0001,
+      "loss": 4.0229,
+      "step": 3500
+    },
+    {
+      "epoch": 0.10157698265573022,
+      "grad_norm": 1.3109067678451538,
+      "learning_rate": 0.0001,
+      "loss": 4.0307,
+      "step": 4000
+    },
+    {
+      "epoch": 0.10157698265573022,
+      "eval_loss": 3.9221267700195312,
+      "eval_runtime": 1.8807,
+      "eval_samples_per_second": 930.486,
+      "eval_steps_per_second": 29.244,
+      "step": 4000
+    },
+    {
+      "epoch": 0.11427410548769648,
+      "grad_norm": 1.6231205463409424,
+      "learning_rate": 0.0001,
+      "loss": 3.9838,
+      "step": 4500
+    },
+    {
+      "epoch": 0.12697122831966276,
+      "grad_norm": 1.6827378273010254,
+      "learning_rate": 0.0001,
+      "loss": 3.8887,
+      "step": 5000
+    },
+    {
+      "epoch": 0.12697122831966276,
+      "eval_loss": 3.857755661010742,
+      "eval_runtime": 1.8825,
+      "eval_samples_per_second": 929.612,
+      "eval_steps_per_second": 29.216,
+      "step": 5000
+    },
+    {
+      "epoch": 0.13966835115162904,
+      "grad_norm": 1.4547417163848877,
+      "learning_rate": 0.0001,
+      "loss": 3.8996,
+      "step": 5500
+    },
+    {
+      "epoch": 0.15236547398359532,
+      "grad_norm": 1.4457532167434692,
+      "learning_rate": 0.0001,
+      "loss": 3.8689,
+      "step": 6000
+    },
+    {
+      "epoch": 0.15236547398359532,
+      "eval_loss": 3.779956579208374,
+      "eval_runtime": 1.8835,
+      "eval_samples_per_second": 929.122,
+      "eval_steps_per_second": 29.201,
+      "step": 6000
+    },
+    {
+      "epoch": 0.1650625968155616,
+      "grad_norm": 1.6317448616027832,
+      "learning_rate": 0.0001,
+      "loss": 3.8136,
+      "step": 6500
+    },
+    {
+      "epoch": 0.17775971964752788,
+      "grad_norm": 1.5866583585739136,
+      "learning_rate": 0.0001,
+      "loss": 3.7808,
+      "step": 7000
+    },
+    {
+      "epoch": 0.17775971964752788,
+      "eval_loss": 3.7245218753814697,
+      "eval_runtime": 2.2249,
+      "eval_samples_per_second": 786.548,
+      "eval_steps_per_second": 24.72,
+      "step": 7000
+    },
+    {
+      "epoch": 0.19045684247949415,
+      "grad_norm": 1.6744858026504517,
+      "learning_rate": 0.0001,
+      "loss": 3.7905,
+      "step": 7500
+    },
+    {
+      "epoch": 0.20315396531146043,
+      "grad_norm": 1.818229079246521,
+      "learning_rate": 0.0001,
+      "loss": 3.742,
+      "step": 8000
+    },
+    {
+      "epoch": 0.20315396531146043,
+      "eval_loss": 3.685417413711548,
+      "eval_runtime": 1.8966,
+      "eval_samples_per_second": 922.71,
+      "eval_steps_per_second": 28.999,
+      "step": 8000
+    },
+    {
+      "epoch": 0.2158510881434267,
+      "grad_norm": 1.648987054824829,
+      "learning_rate": 0.0001,
+      "loss": 3.6802,
+      "step": 8500
+    },
+    {
+      "epoch": 0.22854821097539296,
+      "grad_norm": 1.5762256383895874,
+      "learning_rate": 0.0001,
+      "loss": 3.7303,
+      "step": 9000
+    },
+    {
+      "epoch": 0.22854821097539296,
+      "eval_loss": 3.6259124279022217,
+      "eval_runtime": 1.8817,
+      "eval_samples_per_second": 930.029,
+      "eval_steps_per_second": 29.229,
+      "step": 9000
+    },
+    {
+      "epoch": 0.24124533380735924,
+      "grad_norm": 1.7662715911865234,
+      "learning_rate": 0.0001,
+      "loss": 3.6795,
+      "step": 9500
+    },
+    {
+      "epoch": 0.2539424566393255,
+      "grad_norm": 1.4495737552642822,
+      "learning_rate": 0.0001,
+      "loss": 3.5985,
+      "step": 10000
+    },
+    {
+      "epoch": 0.2539424566393255,
+      "eval_loss": 3.6000101566314697,
+      "eval_runtime": 1.878,
+      "eval_samples_per_second": 931.832,
+      "eval_steps_per_second": 29.286,
+      "step": 10000
+    },
+    {
+      "epoch": 0.2666395794712918,
+      "grad_norm": 1.66859769821167,
+      "learning_rate": 0.0001,
+      "loss": 3.6678,
+      "step": 10500
+    },
+    {
+      "epoch": 0.2793367023032581,
+      "grad_norm": 1.732608437538147,
+      "learning_rate": 0.0001,
+      "loss": 3.6448,
+      "step": 11000
+    },
+    {
+      "epoch": 0.2793367023032581,
+      "eval_loss": 3.5645763874053955,
+      "eval_runtime": 1.8955,
+      "eval_samples_per_second": 923.251,
+      "eval_steps_per_second": 29.016,
+      "step": 11000
+    }
+  ],
+  "logging_steps": 500,
+  "max_steps": 39379,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 1,
+  "save_steps": 1000,
+  "stateful_callbacks": {
+    "EarlyStoppingCallback": {
+      "args": {
+        "early_stopping_patience": 3,
+        "early_stopping_threshold": 0.001
+      },
+      "attributes": {
+        "early_stopping_patience_counter": 0
+      }
+    },
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 214469443584000.0,
+  "train_batch_size": 32,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-11000/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e446b19bfdd079c96f93f25aba6c168e7eaf20eb6abb93593c603320e1446ee9
+size 5432