Add checkpoint

README.md

@@ -0,0 +1,70 @@
+---
+library_name: transformers
+tags:
+- generated_from_trainer
+metrics:
+- accuracy
+model-index:
+- name: BABYLM_50000_41
+  results: []
+---
+
+<!-- This model card has been generated automatically according to the information the Trainer had access to. You
+should probably proofread and complete it, then remove this comment. -->
+
+# BABYLM_50000_41
+
+This model was trained from scratch on an unknown dataset.
+It achieves the following results on the evaluation set:
+- Loss: 3.8863
+- Accuracy: 0.3690
+
+## Model description
+
+More information needed
+
+## Intended uses & limitations
+
+More information needed
+
+## Training and evaluation data
+
+More information needed
+
+## Training procedure
+
+### Training hyperparameters
+
+The following hyperparameters were used during training:
+- learning_rate: 0.0006
+- train_batch_size: 32
+- eval_batch_size: 32
+- seed: 41
+- optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
+- lr_scheduler_type: linear
+- lr_scheduler_warmup_steps: 1000
+- num_epochs: 10.0
+- mixed_precision_training: Native AMP
+
+### Training results
+
+| Training Loss | Epoch | Step  | Validation Loss | Accuracy |
+|:-------------:|:-----:|:-----:|:---------------:|:--------:|
+| 1.1472        | 1.0   | 1008  | 4.4067          | 0.3239   |
+| 1.0265        | 2.0   | 2016  | 4.3543          | 0.3260   |
+| 1.011         | 3.0   | 3024  | 4.3273          | 0.3275   |
+| 1.0037        | 4.0   | 4032  | 4.2573          | 0.3327   |
+| 0.9846        | 5.0   | 5040  | 4.2131          | 0.3388   |
+| 0.9628        | 6.0   | 6048  | 4.1228          | 0.3467   |
+| 0.9383        | 7.0   | 7056  | 4.0391          | 0.3538   |
+| 0.9143        | 8.0   | 8064  | 3.9592          | 0.3621   |
+| 0.8929        | 9.0   | 9072  | 3.9107          | 0.3665   |
+| 0.88          | 10.0  | 10080 | 3.8863          | 0.3690   |
+
+
+### Framework versions
+
+- Transformers 4.51.3
+- Pytorch 2.5.1+cu121
+- Datasets 3.6.0
+- Tokenizers 0.21.1

all_results.json

@@ -0,0 +1,16 @@
+{
+    "epoch": 10.0,
+    "eval_accuracy": 0.3690443281951633,
+    "eval_loss": 3.8863136768341064,
+    "eval_runtime": 859.3565,
+    "eval_samples": 33202,
+    "eval_samples_per_second": 38.636,
+    "eval_steps_per_second": 1.208,
+    "perplexity": 48.730917106191576,
+    "total_flos": 8.426145447936e+16,
+    "train_loss": 0.9752902931637234,
+    "train_runtime": 19950.9665,
+    "train_samples": 32248,
+    "train_samples_per_second": 16.164,
+    "train_steps_per_second": 0.505
+}

config.json

@@ -0,0 +1,40 @@
+{
+  "activation_function": "gelu_new",
+  "architectures": [
+    "GPT2LMHeadModel"
+  ],
+  "attn_pdrop": 0.1,
+  "bos_token_id": 1,
+  "embd_pdrop": 0.1,
+  "eos_token_id": 2,
+  "initializer_range": 0.02,
+  "layer_norm_epsilon": 1e-05,
+  "model_type": "gpt2",
+  "n_ctx": 1024,
+  "n_embd": 768,
+  "n_head": 12,
+  "n_inner": null,
+  "n_layer": 12,
+  "n_positions": 512,
+  "pad_token_id": 0,
+  "prefix": "<|endoftext|>",
+  "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": "4.51.3",
+  "use_cache": true,
+  "vocab_size": 50000
+}

eval_results.json

@@ -0,0 +1,10 @@
+{
+    "epoch": 10.0,
+    "eval_accuracy": 0.3690443281951633,
+    "eval_loss": 3.8863136768341064,
+    "eval_runtime": 859.3565,
+    "eval_samples": 33202,
+    "eval_samples_per_second": 38.636,
+    "eval_steps_per_second": 1.208,
+    "perplexity": 48.730917106191576
+}

generation_config.json

@@ -0,0 +1,7 @@
+{
+  "_from_model_config": true,
+  "bos_token_id": 1,
+  "eos_token_id": 2,
+  "pad_token_id": 0,
+  "transformers_version": "4.51.3"
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b287aa778e7d9ef4497b323f43c16c7300420aba4bd685cd4acc2fc95b8bee2e
+size 495411840

special_tokens_map.json

@@ -0,0 +1,6 @@
+{
+  "bos_token": "<|endoftext|>",
+  "eos_token": "<|endoftext|>",
+  "pad_token": "<pad>",
+  "unk_token": "<|endoftext|>"
+}

tokenizer_config.json

@@ -0,0 +1,49 @@
+{
+  "add_prefix_space": true,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<pad>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<|endoftext|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "<s>",
+      "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
+    }
+  },
+  "bos_token": "<|endoftext|>",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|endoftext|>",
+  "extra_special_tokens": {},
+  "max_length": null,
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_to_multiple_of": null,
+  "pad_token": "<pad>",
+  "pad_token_type_id": 0,
+  "padding_side": "right",
+  "tokenizer_class": "GPT2Tokenizer",
+  "unk_token": "<|endoftext|>"
+}

train_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 10.0,
+    "total_flos": 8.426145447936e+16,
+    "train_loss": 0.9752902931637234,
+    "train_runtime": 19950.9665,
+    "train_samples": 32248,
+    "train_samples_per_second": 16.164,
+    "train_steps_per_second": 0.505
+}

trainer_state.json

@@ -0,0 +1,203 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 10.0,
+  "eval_steps": 500,
+  "global_step": 10080,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.9920634920634921,
+      "grad_norm": 17054.458984375,
+      "learning_rate": 0.0005993999999999999,
+      "loss": 1.1472,
+      "step": 1000
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.3239240886981203,
+      "eval_loss": 4.4066948890686035,
+      "eval_runtime": 855.2718,
+      "eval_samples_per_second": 38.82,
+      "eval_steps_per_second": 1.214,
+      "step": 1008
+    },
+    {
+      "epoch": 1.9841269841269842,
+      "grad_norm": 29671.1875,
+      "learning_rate": 0.0005339867841409691,
+      "loss": 1.0265,
+      "step": 2000
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.3259714508038384,
+      "eval_loss": 4.354335784912109,
+      "eval_runtime": 855.8298,
+      "eval_samples_per_second": 38.795,
+      "eval_steps_per_second": 1.213,
+      "step": 2016
+    },
+    {
+      "epoch": 2.9761904761904763,
+      "grad_norm": 70240.5390625,
+      "learning_rate": 0.0004679074889867841,
+      "loss": 1.011,
+      "step": 3000
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.32745557614417636,
+      "eval_loss": 4.327347755432129,
+      "eval_runtime": 853.4908,
+      "eval_samples_per_second": 38.901,
+      "eval_steps_per_second": 1.216,
+      "step": 3024
+    },
+    {
+      "epoch": 3.9682539682539684,
+      "grad_norm": 46378.90234375,
+      "learning_rate": 0.0004018281938325991,
+      "loss": 1.0037,
+      "step": 4000
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.33272428004301724,
+      "eval_loss": 4.257251262664795,
+      "eval_runtime": 855.0895,
+      "eval_samples_per_second": 38.829,
+      "eval_steps_per_second": 1.214,
+      "step": 4032
+    },
+    {
+      "epoch": 4.9603174603174605,
+      "grad_norm": 28076.61328125,
+      "learning_rate": 0.00033574889867841405,
+      "loss": 0.9846,
+      "step": 5000
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.33883436159210933,
+      "eval_loss": 4.213110446929932,
+      "eval_runtime": 854.9015,
+      "eval_samples_per_second": 38.837,
+      "eval_steps_per_second": 1.214,
+      "step": 5040
+    },
+    {
+      "epoch": 5.9523809523809526,
+      "grad_norm": 14078.001953125,
+      "learning_rate": 0.000269669603524229,
+      "loss": 0.9628,
+      "step": 6000
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.34673329159550076,
+      "eval_loss": 4.1228461265563965,
+      "eval_runtime": 857.4048,
+      "eval_samples_per_second": 38.724,
+      "eval_steps_per_second": 1.211,
+      "step": 6048
+    },
+    {
+      "epoch": 6.944444444444445,
+      "grad_norm": 9524.201171875,
+      "learning_rate": 0.00020359030837004404,
+      "loss": 0.9383,
+      "step": 7000
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.3538012175014567,
+      "eval_loss": 4.039106845855713,
+      "eval_runtime": 854.4996,
+      "eval_samples_per_second": 38.855,
+      "eval_steps_per_second": 1.215,
+      "step": 7056
+    },
+    {
+      "epoch": 7.936507936507937,
+      "grad_norm": 7510.40869140625,
+      "learning_rate": 0.000137511013215859,
+      "loss": 0.9143,
+      "step": 8000
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.3621231644852932,
+      "eval_loss": 3.9591543674468994,
+      "eval_runtime": 858.8934,
+      "eval_samples_per_second": 38.657,
+      "eval_steps_per_second": 1.209,
+      "step": 8064
+    },
+    {
+      "epoch": 8.928571428571429,
+      "grad_norm": 7499.3291015625,
+      "learning_rate": 7.1431718061674e-05,
+      "loss": 0.8929,
+      "step": 9000
+    },
+    {
+      "epoch": 9.0,
+      "eval_accuracy": 0.3664532976168766,
+      "eval_loss": 3.910747528076172,
+      "eval_runtime": 856.101,
+      "eval_samples_per_second": 38.783,
+      "eval_steps_per_second": 1.212,
+      "step": 9072
+    },
+    {
+      "epoch": 9.920634920634921,
+      "grad_norm": 7264.61669921875,
+      "learning_rate": 5.352422907488987e-06,
+      "loss": 0.88,
+      "step": 10000
+    },
+    {
+      "epoch": 10.0,
+      "eval_accuracy": 0.3690443281951633,
+      "eval_loss": 3.8863136768341064,
+      "eval_runtime": 867.6713,
+      "eval_samples_per_second": 38.266,
+      "eval_steps_per_second": 1.196,
+      "step": 10080
+    },
+    {
+      "epoch": 10.0,
+      "step": 10080,
+      "total_flos": 8.426145447936e+16,
+      "train_loss": 0.9752902931637234,
+      "train_runtime": 19950.9665,
+      "train_samples_per_second": 16.164,
+      "train_steps_per_second": 0.505
+    }
+  ],
+  "logging_steps": 1000,
+  "max_steps": 10080,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 10,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 8.426145447936e+16,
+  "train_batch_size": 32,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d183ea0ceb947c9b9b82399970949b0bc2739d78350970eb0c0df5c5e9693b77
+size 5304