Training in progress, step 8000, checkpoint

checkpoint-8000/config.json

@@ -0,0 +1,32 @@
+{
+  "activation_function": "gelu_new",
+  "architectures": [
+    "GPT2LMHeadModel"
+  ],
+  "attn_pdrop": 0.1,
+  "bos_token_id": 0,
+  "embd_pdrop": 0.1,
+  "eos_token_id": 1,
+  "initializer_range": 0.02,
+  "layer_norm_epsilon": 1e-05,
+  "model_type": "gpt2",
+  "n_ctx": 512,
+  "n_embd": 256,
+  "n_head": 8,
+  "n_inner": 2048,
+  "n_layer": 8,
+  "n_positions": 512,
+  "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,
+  "torch_dtype": "float32",
+  "transformers_version": "4.45.2",
+  "use_cache": true,
+  "vocab_size": 8192
+}

checkpoint-8000/generation_config.json

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

checkpoint-8000/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6b8cb964eb6f210fb37c2c7281acd88182dc6f1d98714a2c6994417e6e59d605
+size 51007160

checkpoint-8000/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5b7392aa0e9fc17189e99ad86e4f1fba212e4f97dc1166cc8fe07f6b11b8e82b
+size 102078202

checkpoint-8000/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:de584a712a042ee7a87388bec1bf7d1f2a2febc8c80675635ad1efe09ab47b44
+size 14244

checkpoint-8000/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a1ae47ac3c7e7d77b096ee52c2399dd2718f92bd5c2d5ef4e626b324ee61395e
+size 1000

checkpoint-8000/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-8000/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-8000/trainer_state.json

@@ -0,0 +1,79 @@
+{
+  "best_metric": 5.171741008758545,
+  "best_model_checkpoint": "/home/p318482/babyLM_controlled/models_trained_last/de_clm/de_childes_42/checkpoint-8000",
+  "epoch": 6.008261359369133,
+  "eval_steps": 2000,
+  "global_step": 8000,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 1.5020653398422832,
+      "eval_loss": 7.03363037109375,
+      "eval_runtime": 3.6768,
+      "eval_samples_per_second": 902.97,
+      "eval_steps_per_second": 56.572,
+      "step": 2000
+    },
+    {
+      "epoch": 3.0041306796845664,
+      "grad_norm": 1.6605541706085205,
+      "learning_rate": 1e-05,
+      "loss": 6.9417,
+      "step": 4000
+    },
+    {
+      "epoch": 3.0041306796845664,
+      "eval_loss": 5.811168670654297,
+      "eval_runtime": 3.7246,
+      "eval_samples_per_second": 891.378,
+      "eval_steps_per_second": 55.845,
+      "step": 4000
+    },
+    {
+      "epoch": 4.50619601952685,
+      "eval_loss": 5.458603382110596,
+      "eval_runtime": 3.8082,
+      "eval_samples_per_second": 871.804,
+      "eval_steps_per_second": 54.619,
+      "step": 6000
+    },
+    {
+      "epoch": 6.008261359369133,
+      "grad_norm": 2.27949595451355,
+      "learning_rate": 2e-05,
+      "loss": 5.2077,
+      "step": 8000
+    },
+    {
+      "epoch": 6.008261359369133,
+      "eval_loss": 5.171741008758545,
+      "eval_runtime": 3.8003,
+      "eval_samples_per_second": 873.606,
+      "eval_steps_per_second": 54.732,
+      "step": 8000
+    }
+  ],
+  "logging_steps": 4000,
+  "max_steps": 100000,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 76,
+  "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": 2068142220115968.0,
+  "train_batch_size": 16,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-8000/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:77f63c790664816f6ed91c4986d1e19906c6b86d33dddcf3c8935838a9bb688c
+size 5368