Training in progress, step 2

adapter_config.json

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

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:84ffef693e76e44506b26d938714d20a6b455ee71c516a59ad9d1697b87868cf
+oid sha256:aaa09614d53c95426e1db17f54651e9e174b5d66cd6aa8eb3820da83de7b1d24
 size 335604696

all_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 0.008888888888888889,
-    "eval_accuracy": 0.30833333333333335,
-    "eval_loss": 0.69873046875,
-    "eval_runtime": 12.9332,
-    "eval_samples_per_second": 15.464,
-    "eval_steps_per_second": 1.933
+    "eval_accuracy": 0.6481481481481481,
+    "eval_loss": 0.6923828125,
+    "eval_runtime": 16.3982,
+    "eval_samples_per_second": 12.196,
+    "eval_steps_per_second": 3.049
 }

eval_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 0.008888888888888889,
-    "eval_accuracy": 0.30833333333333335,
-    "eval_loss": 0.69873046875,
-    "eval_runtime": 12.9332,
-    "eval_samples_per_second": 15.464,
-    "eval_steps_per_second": 1.933
+    "eval_accuracy": 0.6481481481481481,
+    "eval_loss": 0.6923828125,
+    "eval_runtime": 16.3982,
+    "eval_samples_per_second": 12.196,
+    "eval_steps_per_second": 3.049
 }

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7ed6741bda4959c7a4525e886843c26cd1f59fbb620407eeae07f38a2965fb27
+oid sha256:358fd4a5ac2a959d81f91e781f4a6699fb17f7df6aff3cd7b17cf68827043b85
 size 5432