Training in progress, step 16000, checkpoint

checkpoint-16000/config.json

@@ -0,0 +1,28 @@
+{
+  "architectures": [
+    "RobertaForMaskedLM"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "bos_token_id": 3,
+  "classifier_dropout": null,
+  "eos_token_id": 4,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 256,
+  "initializer_range": 0.02,
+  "intermediate_size": 1024,
+  "layer_norm_eps": 1e-05,
+  "mask_token_id": 6,
+  "max_position_embeddings": 512,
+  "model_type": "roberta",
+  "num_attention_heads": 8,
+  "num_hidden_layers": 8,
+  "pad_token_id": 1,
+  "position_embedding_type": "absolute",
+  "tie_word_embeddings": false,
+  "torch_dtype": "float32",
+  "transformers_version": "4.45.2",
+  "type_vocab_size": 2,
+  "use_cache": true,
+  "vocab_size": 8192
+}

checkpoint-16000/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:457b3b872046388944cdf22da550eecdd304753d7ce38136e67c5eb304d6a6e4
+size 42892200

checkpoint-16000/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f9c414c1b0838ae84260650593faea488f395e4af7c781a4e27b08f48aec1a0d
+size 85868090

checkpoint-16000/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:789a07eb223b1d4fcafae88d6e3b0a0cc13777e536f93c03e0a78a63f88d8f6f
+size 14244

checkpoint-16000/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1e0480c1ad5b3e4b47a3c01da45b211859b181e54301e949d089a1b445c79daa
+size 1000

checkpoint-16000/special_tokens_map.json

@@ -0,0 +1,6 @@
+{
+  "bos_token": "[PAD]",
+  "eos_token": "[EOS]",
+  "mask_token": "[MASK]",
+  "pad_token": "[CLS]"
+}

checkpoint-16000/tokenizer_config.json

@@ -0,0 +1,67 @@
+{
+  "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
+    },
+    "6": {
+      "content": "[MASK]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "bos_token": "[PAD]",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "[EOS]",
+  "mask_token": "[MASK]",
+  "model_max_length": 128,
+  "pad_token": "[CLS]",
+  "tokenizer_class": "PreTrainedTokenizerFast"
+}

checkpoint-16000/trainer_state.json

@@ -0,0 +1,134 @@
+{
+  "best_metric": 2.7470672130584717,
+  "best_model_checkpoint": "/home/p318482/babyLM_controlled/models_trained_recent/mlm_unmasking/fr_mlm_new/childes_mlm_unmasking_context_42/checkpoint-16000",
+  "epoch": 28.1195079086116,
+  "eval_steps": 2000,
+  "global_step": 16000,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 3.51493848857645,
+      "eval_loss": 5.561293601989746,
+      "eval_runtime": 1.7076,
+      "eval_samples_per_second": 1305.334,
+      "eval_steps_per_second": 81.986,
+      "step": 2000
+    },
+    {
+      "epoch": 7.0298769771529,
+      "grad_norm": 1.3643518686294556,
+      "learning_rate": 4e-05,
+      "loss": 6.2537,
+      "step": 4000
+    },
+    {
+      "epoch": 7.0298769771529,
+      "eval_loss": 5.49988317489624,
+      "eval_runtime": 1.7023,
+      "eval_samples_per_second": 1309.415,
+      "eval_steps_per_second": 82.242,
+      "step": 4000
+    },
+    {
+      "epoch": 10.54481546572935,
+      "eval_loss": 5.408681392669678,
+      "eval_runtime": 1.7195,
+      "eval_samples_per_second": 1296.321,
+      "eval_steps_per_second": 81.42,
+      "step": 6000
+    },
+    {
+      "epoch": 14.0597539543058,
+      "grad_norm": 3.336256265640259,
+      "learning_rate": 8e-05,
+      "loss": 5.329,
+      "step": 8000
+    },
+    {
+      "epoch": 14.0597539543058,
+      "eval_loss": 4.505647659301758,
+      "eval_runtime": 1.6952,
+      "eval_samples_per_second": 1314.863,
+      "eval_steps_per_second": 82.584,
+      "step": 8000
+    },
+    {
+      "epoch": 17.57469244288225,
+      "eval_loss": 3.6001980304718018,
+      "eval_runtime": 1.6952,
+      "eval_samples_per_second": 1314.862,
+      "eval_steps_per_second": 82.584,
+      "step": 10000
+    },
+    {
+      "epoch": 21.0896309314587,
+      "grad_norm": 2.967787504196167,
+      "learning_rate": 0.00012,
+      "loss": 3.7571,
+      "step": 12000
+    },
+    {
+      "epoch": 21.0896309314587,
+      "eval_loss": 3.1521711349487305,
+      "eval_runtime": 1.6965,
+      "eval_samples_per_second": 1313.906,
+      "eval_steps_per_second": 82.524,
+      "step": 12000
+    },
+    {
+      "epoch": 24.604569420035148,
+      "eval_loss": 2.8882946968078613,
+      "eval_runtime": 1.7031,
+      "eval_samples_per_second": 1308.779,
+      "eval_steps_per_second": 82.202,
+      "step": 14000
+    },
+    {
+      "epoch": 28.1195079086116,
+      "grad_norm": 2.9542622566223145,
+      "learning_rate": 0.00016,
+      "loss": 2.9513,
+      "step": 16000
+    },
+    {
+      "epoch": 28.1195079086116,
+      "eval_loss": 2.7470672130584717,
+      "eval_runtime": 1.7009,
+      "eval_samples_per_second": 1310.516,
+      "eval_steps_per_second": 82.311,
+      "step": 16000
+    }
+  ],
+  "logging_steps": 4000,
+  "max_steps": 400000,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 703,
+  "save_steps": 4000,
+  "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": 3336855044751360.0,
+  "train_batch_size": 16,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-16000/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:864f26ae3ed21d609fcbef4afeaac5049d9afc52c206fb700dc34dd3284525ec
+size 5496