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": 0,
+  "classifier_dropout": null,
+  "eos_token_id": 1,
+  "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": 4,
+  "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 sha256:3bb4b3720ef4f7b7f47ca3dd45db6276e218ec391f1a56381c172bc3e442392d
+size 59702184

checkpoint-16000/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:843d8feb3d5bac19edc0df3455c8423237afd96c30bd37a67f78c4cc9f30a0b6
+size 119488058

checkpoint-16000/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1055b8f1b2d8773e7f985a3c8689087aeffaa23be02b08c61b0e55df7982eb76
+size 14308

checkpoint-16000/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f1d534a2bb9aa7e14d4f8fd025f70bb885236f8e21c6db210bf254630b820270
+size 1000

checkpoint-16000/special_tokens_map.json

@@ -0,0 +1,34 @@
+{
+  "additional_special_tokens": [
+    "<|endoftext|>"
+  ],
+  "bos_token": {
+    "content": "<s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "</s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "mask_token": {
+    "content": "<mask>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "</s>",
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

checkpoint-16000/tokenizer_config.json

@@ -0,0 +1,63 @@
+{
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<s>",
+      "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": "<pad>",
+      "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": "<|endoftext|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "additional_special_tokens": [
+    "<|endoftext|>"
+  ],
+  "bos_token": "<s>",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "</s>",
+  "mask_token": "<mask>",
+  "model_max_length": 128,
+  "pad_token": "</s>",
+  "tokenizer_class": "PreTrainedTokenizerFast",
+  "unk_token": "<unk>"
+}

checkpoint-16000/trainer_state.json

@@ -0,0 +1,125 @@
+{
+  "best_metric": 6.889880180358887,
+  "best_model_checkpoint": "/home/p318482/babyLM_controlled/models_trained/de_mlm/wikipedia_13/checkpoint-16000",
+  "epoch": 11.725906925613778,
+  "eval_steps": 2000,
+  "global_step": 16000,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 1.4657383657017222,
+      "eval_loss": 8.087221145629883,
+      "eval_runtime": 2.215,
+      "eval_samples_per_second": 1035.232,
+      "eval_steps_per_second": 65.012,
+      "step": 2000
+    },
+    {
+      "epoch": 2.9314767314034444,
+      "grad_norm": 0.9454871416091919,
+      "learning_rate": 1e-05,
+      "loss": 8.1326,
+      "step": 4000
+    },
+    {
+      "epoch": 2.9314767314034444,
+      "eval_loss": 7.397463798522949,
+      "eval_runtime": 2.2812,
+      "eval_samples_per_second": 1005.158,
+      "eval_steps_per_second": 63.124,
+      "step": 4000
+    },
+    {
+      "epoch": 4.397215097105167,
+      "eval_loss": 7.2718353271484375,
+      "eval_runtime": 2.3459,
+      "eval_samples_per_second": 977.446,
+      "eval_steps_per_second": 61.383,
+      "step": 6000
+    },
+    {
+      "epoch": 5.862953462806889,
+      "grad_norm": 1.4085547924041748,
+      "learning_rate": 2e-05,
+      "loss": 7.2846,
+      "step": 8000
+    },
+    {
+      "epoch": 5.862953462806889,
+      "eval_loss": 7.183506965637207,
+      "eval_runtime": 2.2787,
+      "eval_samples_per_second": 1006.271,
+      "eval_steps_per_second": 63.194,
+      "step": 8000
+    },
+    {
+      "epoch": 7.328691828508611,
+      "eval_loss": 7.09923791885376,
+      "eval_runtime": 2.1833,
+      "eval_samples_per_second": 1050.26,
+      "eval_steps_per_second": 65.956,
+      "step": 10000
+    },
+    {
+      "epoch": 8.794430194210333,
+      "grad_norm": 1.4775667190551758,
+      "learning_rate": 2.9995e-05,
+      "loss": 7.1078,
+      "step": 12000
+    },
+    {
+      "epoch": 8.794430194210333,
+      "eval_loss": 7.033052444458008,
+      "eval_runtime": 2.2436,
+      "eval_samples_per_second": 1022.003,
+      "eval_steps_per_second": 64.182,
+      "step": 12000
+    },
+    {
+      "epoch": 10.260168559912056,
+      "eval_loss": 6.949419975280762,
+      "eval_runtime": 2.3034,
+      "eval_samples_per_second": 995.501,
+      "eval_steps_per_second": 62.517,
+      "step": 14000
+    },
+    {
+      "epoch": 11.725906925613778,
+      "grad_norm": 2.364133358001709,
+      "learning_rate": 3.999e-05,
+      "loss": 6.942,
+      "step": 16000
+    },
+    {
+      "epoch": 11.725906925613778,
+      "eval_loss": 6.889880180358887,
+      "eval_runtime": 2.3275,
+      "eval_samples_per_second": 985.162,
+      "eval_steps_per_second": 61.868,
+      "step": 16000
+    }
+  ],
+  "logging_steps": 4000,
+  "max_steps": 100000,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 74,
+  "save_steps": 4000,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 4990343509573632.0,
+  "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 sha256:18d5335e549995a666047ad0aa350511260e15302e12c49cfc12029ab6ff3c4a
+size 5368