Training in progress, step 26000, checkpoint

checkpoint-26000/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": 128,
+  "n_embd": 128,
+  "n_head": 8,
+  "n_inner": null,
+  "n_layer": 8,
+  "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": 16384
+}

checkpoint-26000/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-26000/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3713e3f5a8a32ba7c97f6e2980f63f786e9a1bf8d94d7f02f73f84396446ee8c
+size 14809584

checkpoint-26000/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:eeb6594411729d19d212d6bddc66c494a4bd5a85c12d9e62651f534cdc64458c
+size 29680378

checkpoint-26000/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:877292d22740dab71a60eba368c24f266a89afae796557be7d08efa3e050bb84
+size 13990

checkpoint-26000/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:88758702d491557b0198389be3dada205f4b7e578089958e251f82889e0ade9a
+size 1192

checkpoint-26000/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-26000/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": false,
+  "eos_token": "[EOS]",
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "[EOS]",
+  "tokenizer_class": "PreTrainedTokenizerFast"
+}

checkpoint-26000/trainer_state.json

@@ -0,0 +1,237 @@
+{
+  "best_metric": 0.15186844766139984,
+  "best_model_checkpoint": "/Users/frapadovani/Desktop/babyLM_controlled/models_trained/convergence_french/random_sentence_french/checkpoint-26000",
+  "epoch": 0.04701865555579668,
+  "eval_steps": 2000,
+  "global_step": 26000,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.003616819658138206,
+      "grad_norm": 0.14744240045547485,
+      "learning_rate": 0.0001,
+      "loss": 0.7697,
+      "step": 2000
+    },
+    {
+      "epoch": 0.003616819658138206,
+      "eval_loss": 0.19154727458953857,
+      "eval_runtime": 91.812,
+      "eval_samples_per_second": 397.933,
+      "eval_steps_per_second": 24.877,
+      "step": 2000
+    },
+    {
+      "epoch": 0.007233639316276412,
+      "grad_norm": 0.1623348593711853,
+      "learning_rate": 0.0001,
+      "loss": 0.1884,
+      "step": 4000
+    },
+    {
+      "epoch": 0.007233639316276412,
+      "eval_loss": 0.17771275341510773,
+      "eval_runtime": 99.0721,
+      "eval_samples_per_second": 368.772,
+      "eval_steps_per_second": 23.054,
+      "step": 4000
+    },
+    {
+      "epoch": 0.010850458974414617,
+      "grad_norm": 0.13657265901565552,
+      "learning_rate": 0.0001,
+      "loss": 0.178,
+      "step": 6000
+    },
+    {
+      "epoch": 0.010850458974414617,
+      "eval_loss": 0.16984443366527557,
+      "eval_runtime": 90.3285,
+      "eval_samples_per_second": 404.468,
+      "eval_steps_per_second": 25.285,
+      "step": 6000
+    },
+    {
+      "epoch": 0.014467278632552824,
+      "grad_norm": 0.15028506517410278,
+      "learning_rate": 0.0001,
+      "loss": 0.1714,
+      "step": 8000
+    },
+    {
+      "epoch": 0.014467278632552824,
+      "eval_loss": 0.16548627614974976,
+      "eval_runtime": 90.9787,
+      "eval_samples_per_second": 401.578,
+      "eval_steps_per_second": 25.105,
+      "step": 8000
+    },
+    {
+      "epoch": 0.01808409829069103,
+      "grad_norm": 0.17982865869998932,
+      "learning_rate": 0.0001,
+      "loss": 0.1669,
+      "step": 10000
+    },
+    {
+      "epoch": 0.01808409829069103,
+      "eval_loss": 0.16203773021697998,
+      "eval_runtime": 87.7097,
+      "eval_samples_per_second": 416.544,
+      "eval_steps_per_second": 26.04,
+      "step": 10000
+    },
+    {
+      "epoch": 0.021700917948829234,
+      "grad_norm": 0.14700624346733093,
+      "learning_rate": 0.0001,
+      "loss": 0.1647,
+      "step": 12000
+    },
+    {
+      "epoch": 0.021700917948829234,
+      "eval_loss": 0.1595447063446045,
+      "eval_runtime": 87.4086,
+      "eval_samples_per_second": 417.979,
+      "eval_steps_per_second": 26.13,
+      "step": 12000
+    },
+    {
+      "epoch": 0.02531773760696744,
+      "grad_norm": 0.1563248634338379,
+      "learning_rate": 0.0001,
+      "loss": 0.1608,
+      "step": 14000
+    },
+    {
+      "epoch": 0.02531773760696744,
+      "eval_loss": 0.1577664017677307,
+      "eval_runtime": 87.7156,
+      "eval_samples_per_second": 416.517,
+      "eval_steps_per_second": 26.039,
+      "step": 14000
+    },
+    {
+      "epoch": 0.02893455726510565,
+      "grad_norm": 0.1511106789112091,
+      "learning_rate": 0.0001,
+      "loss": 0.1601,
+      "step": 16000
+    },
+    {
+      "epoch": 0.02893455726510565,
+      "eval_loss": 0.15645398199558258,
+      "eval_runtime": 90.1398,
+      "eval_samples_per_second": 405.315,
+      "eval_steps_per_second": 25.338,
+      "step": 16000
+    },
+    {
+      "epoch": 0.032551376923243856,
+      "grad_norm": 0.16017386317253113,
+      "learning_rate": 0.0001,
+      "loss": 0.1607,
+      "step": 18000
+    },
+    {
+      "epoch": 0.032551376923243856,
+      "eval_loss": 0.1552029848098755,
+      "eval_runtime": 87.911,
+      "eval_samples_per_second": 415.591,
+      "eval_steps_per_second": 25.981,
+      "step": 18000
+    },
+    {
+      "epoch": 0.03616819658138206,
+      "grad_norm": 0.12754733860492706,
+      "learning_rate": 0.0001,
+      "loss": 0.1565,
+      "step": 20000
+    },
+    {
+      "epoch": 0.03616819658138206,
+      "eval_loss": 0.15438680350780487,
+      "eval_runtime": 87.7234,
+      "eval_samples_per_second": 416.479,
+      "eval_steps_per_second": 26.036,
+      "step": 20000
+    },
+    {
+      "epoch": 0.039785016239520264,
+      "grad_norm": 0.14869365096092224,
+      "learning_rate": 0.0001,
+      "loss": 0.1555,
+      "step": 22000
+    },
+    {
+      "epoch": 0.039785016239520264,
+      "eval_loss": 0.15326616168022156,
+      "eval_runtime": 87.1572,
+      "eval_samples_per_second": 419.185,
+      "eval_steps_per_second": 26.206,
+      "step": 22000
+    },
+    {
+      "epoch": 0.04340183589765847,
+      "grad_norm": 0.1375146508216858,
+      "learning_rate": 0.0001,
+      "loss": 0.1549,
+      "step": 24000
+    },
+    {
+      "epoch": 0.04340183589765847,
+      "eval_loss": 0.15261560678482056,
+      "eval_runtime": 87.2311,
+      "eval_samples_per_second": 418.83,
+      "eval_steps_per_second": 26.183,
+      "step": 24000
+    },
+    {
+      "epoch": 0.04701865555579668,
+      "grad_norm": 0.15134082734584808,
+      "learning_rate": 0.0001,
+      "loss": 0.1544,
+      "step": 26000
+    },
+    {
+      "epoch": 0.04701865555579668,
+      "eval_loss": 0.15186844766139984,
+      "eval_runtime": 87.5341,
+      "eval_samples_per_second": 417.38,
+      "eval_steps_per_second": 26.093,
+      "step": 26000
+    }
+  ],
+  "logging_steps": 2000,
+  "max_steps": 552972,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 1,
+  "save_steps": 2000,
+  "stateful_callbacks": {
+    "EarlyStoppingCallback": {
+      "args": {
+        "early_stopping_patience": 3,
+        "early_stopping_threshold": 0.001
+      },
+      "attributes": {
+        "early_stopping_patience_counter": 2
+      }
+    },
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 506845986816000.0,
+  "train_batch_size": 16,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-26000/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f5e294196a968439ae6368c7b1ef6eb6e2124702ff760349c5ba2ea15623271d
+size 5496