Upload folder using huggingface_hub

config.json

@@ -0,0 +1,33 @@
+{
+  "activation_function": "gelu_new",
+  "architectures": [
+    "GPT2LMHeadModel"
+  ],
+  "attn_implementation": "sdpa",
+  "attn_pdrop": 0.1,
+  "bos_token_id": 2,
+  "embd_pdrop": 0.1,
+  "eos_token_id": 2,
+  "initializer_range": 0.02,
+  "layer_norm_epsilon": 1e-05,
+  "model_type": "gpt2",
+  "n_ctx": 2048,
+  "n_embd": 768,
+  "n_head": 12,
+  "n_inner": null,
+  "n_layer": 12,
+  "n_positions": 2048,
+  "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": "4.51.3",
+  "use_cache": true,
+  "vocab_size": 200000
+}

eval_final_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 7.530470000941886,
+    "eval_final_loss": 4.826417446136475,
+    "eval_final_ppl": 124.76318820728994,
+    "eval_final_runtime": 4.9262,
+    "eval_final_samples_per_second": 77.747,
+    "eval_final_steps_per_second": 9.744
+}

generation_config.json

@@ -0,0 +1,6 @@
+{
+  "_from_model_config": true,
+  "bos_token_id": 2,
+  "eos_token_id": 2,
+  "transformers_version": "4.51.3"
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b471e7c5c4dc9acbc1b388784ff06d3c88f8a5fb8eda4dcc36bbd20e650b7142
+size 960930432

tokenizer.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:61fd80f7594c686a6b737473aada795fdb56da7afb15fea91a35694fe464811a
+size 3839372