End of training

README.md

@@ -1,5 +1,5 @@
 ---
-base_model: AngelRaychev/0.5B-policy-iteration_1
+base_model: AngelRaychev/0.5B-policy-iteration_0
 library_name: transformers
 model_name: 0.5B-policy-iteration_1
 tags:
@@ -11,7 +11,7 @@ licence: license
 
 # Model Card for 0.5B-policy-iteration_1
 
-This model is a fine-tuned version of [AngelRaychev/0.5B-policy-iteration_1](https://huggingface.co/AngelRaychev/0.5B-policy-iteration_1).
+This model is a fine-tuned version of [AngelRaychev/0.5B-policy-iteration_0](https://huggingface.co/AngelRaychev/0.5B-policy-iteration_0).
 It has been trained using [TRL](https://github.com/huggingface/trl).
 
 ## Quick start

config.json

@@ -11,7 +11,7 @@
   "intermediate_size": 4864,
   "max_position_embeddings": 32768,
   "max_window_layers": 24,
-  "model_card": "\nFinal Loss: 0.1864\nBatch Size: 128\nLearning Rate: 2e-05\nDataset Size: 12000\n",
+  "model_card": "\nFinal Loss: 0.1322\nBatch Size: 256\nLearning Rate: 5e-05\nDataset Size: 12000\n",
   "model_type": "qwen2",
   "num_attention_heads": 14,
   "num_hidden_layers": 24,

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:dc78273a56022f074818e0d16161820ac4f2dac511b02132da715ce9b0225407
+oid sha256:63f1ae89b0acabcd1fc5db8f0ce8353a20ff8321b30e4b74f6ea034e53ed4e8b
 size 988163026

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:af82e20d6685d8bda2a261d6cac813e0b25393aa6cfc65f951a5fefc3de380c7
+oid sha256:ef97afc57949b18fd2288785835675a046bec5061a59df155b12db886a97da6e
 size 5624