End of training

README.md

@@ -1,17 +1,20 @@
 ---
 base_model: Qwen/Qwen2.5-Coder-7B-Instruct
+datasets: flattened_successful_trajectories/v0
 library_name: transformers
 model_name: Qwen2.5-Coder-7B-Instruct-Solver-RFT
 tags:
 - generated_from_trainer
+- kordn
 - sft
 - trl
+- cogzero
 licence: license
 ---
 
 # Model Card for Qwen2.5-Coder-7B-Instruct-Solver-RFT
 
-This model is a fine-tuned version of [Qwen/Qwen2.5-Coder-7B-Instruct](https://huggingface.co/Qwen/Qwen2.5-Coder-7B-Instruct).
+This model is a fine-tuned version of [Qwen/Qwen2.5-Coder-7B-Instruct](https://huggingface.co/Qwen/Qwen2.5-Coder-7B-Instruct) on the [flattened_successful_trajectories/v0](https://huggingface.co/datasets/flattened_successful_trajectories/v0) dataset.
 It has been trained using [TRL](https://github.com/huggingface/trl).
 
 ## Quick start

all_results.json

@@ -1,4 +1,9 @@
 {
+    "eval_loss": 0.11205815523862839,
+    "eval_runtime": 53.6979,
+    "eval_samples": 2246,
+    "eval_samples_per_second": 41.827,
+    "eval_steps_per_second": 5.233,
     "total_flos": 7.689061516716278e+17,
     "train_loss": 0.0504409685922677,
     "train_runtime": 7612.3259,

config.json

@@ -52,7 +52,7 @@
   "tie_word_embeddings": false,
   "torch_dtype": "bfloat16",
   "transformers_version": "4.53.3",
-  "use_cache": false,
+  "use_cache": true,
   "use_sliding_window": false,
   "vocab_size": 152064
 }

eval_results.json

@@ -0,0 +1,7 @@
+{
+    "eval_loss": 0.11205815523862839,
+    "eval_runtime": 53.6979,
+    "eval_samples": 2246,
+    "eval_samples_per_second": 41.827,
+    "eval_steps_per_second": 5.233
+}