Training in progress, step 12, checkpoint

checkpoint-12/adapter_config.json

@@ -23,13 +23,13 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "down_proj",
-    "q_proj",
-    "up_proj",
     "k_proj",
-    "gate_proj",
     "o_proj",
-    "v_proj"
+    "v_proj",
+    "gate_proj",
+    "up_proj",
+    "down_proj",
+    "q_proj"
   ],
   "task_type": "CAUSAL_LM",
   "use_dora": false,

checkpoint-12/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b87bd054e221cbe254a03bb4b69a3d653d51d03f46f553ba7bf32733dce5ebf6
+oid sha256:c81ba4577f2b90ebf944c08cd92158beaf7c433a56828aeb2d2670b1ac57f2d2
 size 83945296

checkpoint-12/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0ee78646aa04c0a66cc8488020d20fcc7008c9c8cfcd0d5854de962724298c87
+oid sha256:bc3066cc1d1029a905ce53416b23d2545c51cd844adc7f6327bfd833b2a1ae83
 size 168149074

checkpoint-12/trainer_state.json

@@ -10,118 +10,118 @@
   "log_history": [
     {
       "epoch": 0.0006200589055960316,
-      "grad_norm": 0.7733756303787231,
+      "grad_norm": 0.7731499075889587,
       "learning_rate": 4.000000000000001e-06,
       "loss": 1.166,
       "step": 1
     },
     {
       "epoch": 0.0012401178111920632,
-      "grad_norm": 0.34648507833480835,
+      "grad_norm": 0.346652090549469,
       "learning_rate": 8.000000000000001e-06,
       "loss": 0.9594,
       "step": 2
     },
     {
       "epoch": 0.0018601767167880949,
-      "grad_norm": 0.3973380923271179,
+      "grad_norm": 0.39937612414360046,
       "learning_rate": 1.2e-05,
-      "loss": 0.8281,
+      "loss": 0.828,
       "step": 3
     },
     {
       "epoch": 0.0018601767167880949,
-      "eval_loss": 1.0490500926971436,
-      "eval_runtime": 28.8129,
-      "eval_samples_per_second": 3.471,
-      "eval_steps_per_second": 3.471,
+      "eval_loss": 1.0491173267364502,
+      "eval_runtime": 26.0459,
+      "eval_samples_per_second": 3.839,
+      "eval_steps_per_second": 3.839,
       "step": 3
     },
     {
       "epoch": 0.0024802356223841263,
-      "grad_norm": 0.342887282371521,
+      "grad_norm": 0.3457990288734436,
       "learning_rate": 1.6000000000000003e-05,
-      "loss": 0.9881,
+      "loss": 0.988,
       "step": 4
     },
     {
       "epoch": 0.0031002945279801583,
-      "grad_norm": 0.46208909153938293,
+      "grad_norm": 0.513918399810791,
       "learning_rate": 2e-05,
       "loss": 0.8319,
       "step": 5
     },
     {
       "epoch": 0.0037203534335761897,
-      "grad_norm": 0.4722931981086731,
+      "grad_norm": 0.47683727741241455,
       "learning_rate": 1.9200000000000003e-05,
-      "loss": 0.9364,
+      "loss": 0.9366,
       "step": 6
     },
     {
       "epoch": 0.0037203534335761897,
-      "eval_loss": 1.0428661108016968,
-      "eval_runtime": 26.3402,
-      "eval_samples_per_second": 3.796,
-      "eval_steps_per_second": 3.796,
+      "eval_loss": 1.0431987047195435,
+      "eval_runtime": 26.2352,
+      "eval_samples_per_second": 3.812,
+      "eval_steps_per_second": 3.812,
       "step": 6
     },
     {
       "epoch": 0.004340412339172222,
-      "grad_norm": 0.4379231333732605,
+      "grad_norm": 0.4349649250507355,
       "learning_rate": 1.8400000000000003e-05,
-      "loss": 0.9073,
+      "loss": 0.9076,
       "step": 7
     },
     {
       "epoch": 0.004960471244768253,
-      "grad_norm": 0.5576643943786621,
+      "grad_norm": 0.5443339943885803,
       "learning_rate": 1.76e-05,
-      "loss": 0.9823,
+      "loss": 0.9829,
       "step": 8
     },
     {
       "epoch": 0.005580530150364285,
-      "grad_norm": 0.4029313325881958,
+      "grad_norm": 0.3990451395511627,
       "learning_rate": 1.6800000000000002e-05,
       "loss": 0.9045,
       "step": 9
     },
     {
       "epoch": 0.005580530150364285,
-      "eval_loss": 1.0290312767028809,
-      "eval_runtime": 26.3678,
-      "eval_samples_per_second": 3.793,
-      "eval_steps_per_second": 3.793,
+      "eval_loss": 1.0292819738388062,
+      "eval_runtime": 26.3477,
+      "eval_samples_per_second": 3.795,
+      "eval_steps_per_second": 3.795,
       "step": 9
     },
     {
       "epoch": 0.0062005890559603165,
-      "grad_norm": 0.4460880160331726,
+      "grad_norm": 0.44750896096229553,
       "learning_rate": 1.6000000000000003e-05,
-      "loss": 1.0763,
+      "loss": 1.0768,
       "step": 10
     },
     {
       "epoch": 0.0068206479615563476,
-      "grad_norm": 0.7060808539390564,
+      "grad_norm": 0.6865904331207275,
       "learning_rate": 1.5200000000000002e-05,
-      "loss": 0.8713,
+      "loss": 0.8717,
       "step": 11
     },
     {
       "epoch": 0.0074407068671523795,
-      "grad_norm": 0.6190696954727173,
+      "grad_norm": 0.548491895198822,
       "learning_rate": 1.4400000000000001e-05,
-      "loss": 1.1753,
+      "loss": 1.1758,
       "step": 12
     },
     {
       "epoch": 0.0074407068671523795,
-      "eval_loss": 1.0143623352050781,
-      "eval_runtime": 26.7829,
-      "eval_samples_per_second": 3.734,
-      "eval_steps_per_second": 3.734,
+      "eval_loss": 1.0153621435165405,
+      "eval_runtime": 26.5025,
+      "eval_samples_per_second": 3.773,
+      "eval_steps_per_second": 3.773,
       "step": 12
     }
   ],

checkpoint-12/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7c433cc5009675ccadca19513c8ae659f957c78885071da55c2a8d32a98d517a
+oid sha256:cbe2fb31cf4c04f893e5299faa77267af76b8447f922d82c2d6c0200e1fe225f
 size 5688