Training in progress, step 4000, checkpoint

checkpoint-4000/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": 2048,
+  "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-4000/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:16c5517409393142c391e6c12858ddee0171ac4ddbf90141094929b26c338c83
+size 59702184

checkpoint-4000/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:49d9ed573abe764c53d1fc9780b5dccb32e13420a85256fa81d36213a5e5bbf4
+size 119488058

checkpoint-4000/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3ca7b686a4cc62b4c3685e8789313658b08950174ca88afc5c651137237eebda
+size 14308

checkpoint-4000/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9cfe9de6955f4677c6cde35895cc3e06c0e151e1353b09de5f84987a447e7795
+size 1000

checkpoint-4000/special_tokens_map.json

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

checkpoint-4000/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-4000/trainer_state.json

@@ -0,0 +1,65 @@
+{
+  "best_metric": 6.555490493774414,
+  "best_model_checkpoint": "/home/p318482/babyLM_controlled/models_trained_recent/mlm_unmasking/de_mlm_big/childes_30/checkpoint-4000",
+  "epoch": 3.3003300330033003,
+  "eval_steps": 2000,
+  "global_step": 4000,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 1.6501650165016502,
+      "eval_loss": 7.850089073181152,
+      "eval_runtime": 1.8443,
+      "eval_samples_per_second": 1326.761,
+      "eval_steps_per_second": 82.956,
+      "step": 2000
+    },
+    {
+      "epoch": 3.3003300330033003,
+      "grad_norm": 1.7952547073364258,
+      "learning_rate": 4.000000000000001e-06,
+      "loss": 7.8543,
+      "step": 4000
+    },
+    {
+      "epoch": 3.3003300330033003,
+      "eval_loss": 6.555490493774414,
+      "eval_runtime": 1.8231,
+      "eval_samples_per_second": 1342.2,
+      "eval_steps_per_second": 83.922,
+      "step": 4000
+    }
+  ],
+  "logging_steps": 4000,
+  "max_steps": 400000,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 331,
+  "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": 1247308058787840.0,
+  "train_batch_size": 16,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-4000/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:82b00b33ff5a34cb27834747d142985f370938cb613e5baed1026ce1070e9a42
+size 5368