End of training

README.md

@@ -1,5 +1,5 @@
 ---
-base_model: AngelRaychev/1.5B-policy-iteration_0
+base_model: AngelRaychev/1.5B-policy-iteration_1
 library_name: transformers
 model_name: 1.5B-policy-iteration_1
 tags:
@@ -11,7 +11,7 @@ licence: license
 
 # Model Card for 1.5B-policy-iteration_1
 
-This model is a fine-tuned version of [AngelRaychev/1.5B-policy-iteration_0](https://huggingface.co/AngelRaychev/1.5B-policy-iteration_0).
+This model is a fine-tuned version of [AngelRaychev/1.5B-policy-iteration_1](https://huggingface.co/AngelRaychev/1.5B-policy-iteration_1).
 It has been trained using [TRL](https://github.com/huggingface/trl).
 
 ## Quick start

config.json

@@ -11,7 +11,7 @@
   "intermediate_size": 8960,
   "max_position_embeddings": 131072,
   "max_window_layers": 28,
-  "model_card": "\nFinal Loss: 0.1511\nBatch Size: 128\nLearning Rate: 5e-05\nDataset Size: 12000\n",
+  "model_card": "\nFinal Loss: 0.1191\nBatch Size: 128\nLearning Rate: 2e-05\nDataset Size: 12000\n",
   "model_type": "qwen2",
   "num_attention_heads": 12,
   "num_hidden_layers": 28,

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:367a94f556dc5a20a5d38b0f5fecd8d8735c49a5ccd4c83340cef93c3eb0796c
+oid sha256:c5ed8b1c575f7004b2a8e20b5ba77d2cca1c10408af1ecc44436242d5c506063
 size 3087542418

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f7e6f8038e2971c4021b1029ebc2b09d36658ca526e9b43678814b4088b5117b
+oid sha256:8d6b3a3c47144ece8459524a9751875282489b14b993cf1c9e48cab88ecd976a
 size 5624