Training in progress, step 5

README.md

@@ -1,6 +1,6 @@
 ---
 base_model: meta-llama/Llama-3.1-8B-Instruct
-library_name: transformers
+library_name: peft
 model_name: output
 tags:
 - generated_from_trainer
@@ -34,6 +34,7 @@ This model was trained with Reward.
 
 ### Framework versions
 
+- PEFT 0.14.0
 - TRL: 0.14.0
 - Transformers: 4.48.2
 - Pytorch: 2.5.1

adapter_config.json

@@ -25,11 +25,11 @@
   "target_modules": [
     "v_proj",
     "o_proj",
-    "up_proj",
-    "k_proj",
-    "gate_proj",
     "down_proj",
-    "q_proj"
+    "q_proj",
+    "k_proj",
+    "up_proj",
+    "gate_proj"
   ],
   "task_type": "CAUSAL_LM",
   "use_dora": false,

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7374334d3ef7c37b90b38309f5dec367aeea0f48a309ad9356dfdb298b9f15cd
+oid sha256:6647db167f1caded3586034e58c464a35fecd1abed1f373098d4906945cc5dce
 size 167832240

all_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 1.0,
-    "eval_accuracy": 0.0,
-    "eval_loss": 0.705078125,
-    "eval_runtime": 0.3434,
-    "eval_samples_per_second": 11.647,
-    "eval_steps_per_second": 2.912
+    "eval_accuracy": 1.0,
+    "eval_loss": 0.68310546875,
+    "eval_runtime": 0.3355,
+    "eval_samples_per_second": 11.923,
+    "eval_steps_per_second": 2.981
 }

eval_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 1.0,
-    "eval_accuracy": 0.0,
-    "eval_loss": 0.705078125,
-    "eval_runtime": 0.3434,
-    "eval_samples_per_second": 11.647,
-    "eval_steps_per_second": 2.912
+    "eval_accuracy": 1.0,
+    "eval_loss": 0.68310546875,
+    "eval_runtime": 0.3355,
+    "eval_samples_per_second": 11.923,
+    "eval_steps_per_second": 2.981
 }