Training in progress, epoch 0, checkpoint

last-checkpoint/adapter_config.json

@@ -20,13 +20,13 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "o_proj",
-    "q_proj",
     "down_proj",
+    "up_proj",
     "v_proj",
     "k_proj",
-    "up_proj",
-    "gate_proj"
+    "gate_proj",
+    "o_proj",
+    "q_proj"
   ],
   "task_type": "CAUSAL_LM",
   "use_dora": false,

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6efb93ec184caebb21b76edcc2ded7acbb4844eb4e971f2cf9198f46cd44839a
+oid sha256:a7e711b7b4034b2f3bd9799e3738e807081a18d9dd9d4402e426cefbb04d49c9
 size 9048

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a0550e9f0ad9a1ad615359ae2c7f3d476f95a14186e0fa3dd682502702db8511
+oid sha256:e58e24af97e29d770d88764fcd1061468363d04933ab5bc770d00cb8fc8b69f1
 size 28134

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:89dbad7a5f4eb47d4580585d91b020d1a7f5dc2fd4d54175561c1eb48d2175e4
+oid sha256:b789b44e35b8f0004dd2e31bfb1af056b8528eeb7ffa2f97d314e99f11187ccd
 size 14244

last-checkpoint/trainer_state.json

@@ -10,17 +10,17 @@
   "log_history": [
     {
       "epoch": 0.0002666666666666667,
-      "grad_norm": 0.014666087925434113,
+      "grad_norm": 0.014549034647643566,
       "learning_rate": 0.0,
       "loss": 11.9308,
       "step": 1
     },
     {
       "epoch": 0.0002666666666666667,
-      "eval_loss": 11.933889389038086,
-      "eval_runtime": 0.062,
-      "eval_samples_per_second": 257.898,
-      "eval_steps_per_second": 64.474,
+      "eval_loss": 11.933904647827148,
+      "eval_runtime": 0.0581,
+      "eval_samples_per_second": 275.597,
+      "eval_steps_per_second": 68.899,
       "step": 1
     }
   ],

last-checkpoint/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:47fa42679f335647845a74974f7e50c2bd6a7ce30b3ac5b0c8963b209e83c057
+oid sha256:a34c87415c9e639533cb4d7b940fdf4014edaec46abd1ff3438eb94417519db3
 size 6776