End of training

README.md

@@ -1,9 +1,11 @@
 ---
 base_model: unsloth/phi-4
+datasets: open-r1/verifiable-coding-problems-python
 library_name: transformers
 model_name: unsloth-phi-4-Instruct-LORA-Open-R1-Code-GRPO-b2-as8
 tags:
 - generated_from_trainer
+- open-r1
 - trl
 - grpo
 licence: license
@@ -11,7 +13,7 @@ licence: license
 
 # Model Card for unsloth-phi-4-Instruct-LORA-Open-R1-Code-GRPO-b2-as8
 
-This model is a fine-tuned version of [unsloth/phi-4](https://huggingface.co/unsloth/phi-4).
+This model is a fine-tuned version of [unsloth/phi-4](https://huggingface.co/unsloth/phi-4) on the [open-r1/verifiable-coding-problems-python](https://huggingface.co/datasets/open-r1/verifiable-coding-problems-python) dataset.
 It has been trained using [TRL](https://github.com/huggingface/trl).
 
 ## Quick start

config.json

@@ -0,0 +1,32 @@
+{
+  "_attn_implementation_autoset": true,
+  "architectures": [
+    "LlamaForCausalLM"
+  ],
+  "attention_bias": false,
+  "attention_dropout": 0.0,
+  "bos_token_id": 100257,
+  "eos_token_id": 100265,
+  "head_dim": 128,
+  "hidden_act": "silu",
+  "hidden_size": 5120,
+  "initializer_range": 0.02,
+  "intermediate_size": 17920,
+  "max_position_embeddings": 16384,
+  "mlp_bias": false,
+  "model_type": "llama",
+  "num_attention_heads": 40,
+  "num_hidden_layers": 40,
+  "num_key_value_heads": 10,
+  "original_max_position_embeddings": 16384,
+  "pad_token_id": 100351,
+  "pretraining_tp": 1,
+  "rms_norm_eps": 1e-05,
+  "rope_scaling": null,
+  "rope_theta": 250000,
+  "tie_word_embeddings": false,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.50.0",
+  "use_cache": true,
+  "vocab_size": 100352
+}