Training in progress, epoch 0, checkpoint

last-checkpoint/adapter_config.json

@@ -20,13 +20,13 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "q_proj",
-    "gate_proj",
-    "o_proj",
     "up_proj",
+    "k_proj",
+    "gate_proj",
     "down_proj",
-    "v_proj",
-    "k_proj"
+    "o_proj",
+    "q_proj",
+    "v_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:eaf406d8cdb6b52f8a340431d451dc1fc4681e2ed0bc3663717273a1b9b3ef12
+oid sha256:3c945ca30ec1e375834aa87855da4255d894deb40e5b9f52325c3ea6fe5a0705
 size 9048

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3a9285a5e0e1816273de107fab9026259478a32528f645b209923a81b449ca3d
+oid sha256:16ae1faee0ed776955ab77fd04343fcc207c1381d2412499eebedc5c1ed7e3aa
 size 28134

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3371d8262583e71fbd2dbc164d88e0c7a9f17a572e2f30bb0e3d8e809fca6486
+oid sha256:89dbad7a5f4eb47d4580585d91b020d1a7f5dc2fd4d54175561c1eb48d2175e4
 size 14244

last-checkpoint/trainer_state.json

@@ -10,17 +10,17 @@
   "log_history": [
     {
       "epoch": 0.0002666666666666667,
-      "grad_norm": 0.013214402832090855,
+      "grad_norm": 0.01467735692858696,
       "learning_rate": 0.0,
       "loss": 11.9308,
       "step": 1
     },
     {
       "epoch": 0.0002666666666666667,
-      "eval_loss": 11.933879852294922,
-      "eval_runtime": 0.0614,
-      "eval_samples_per_second": 260.741,
-      "eval_steps_per_second": 65.185,
+      "eval_loss": 11.933913230895996,
+      "eval_runtime": 0.0593,
+      "eval_samples_per_second": 269.652,
+      "eval_steps_per_second": 67.413,
       "step": 1
     }
   ],

last-checkpoint/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9e1debdbe4095d32d1d33ca4aa71553472896ec45d93796674ecc45e1e93db66
+oid sha256:ffc137fa98ebb6d0b8662ffb83c7ce3e658bcb1935535cbd825760cff33994a6
 size 6776