Training in progress, step 30, checkpoint

checkpoint-30/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-30/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6bd51facca2bc0a3c3f048cc3950e7c453c79081b9cab7b141714fca6bc568ae
+oid sha256:7696ce0eabde7eb93d1bd677680a947fd0923b0334870c81bca1f2af8380ff2e
 size 83945296

checkpoint-30/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1ffe4805e44cd17f39989972073b9c83d4bdc8e45254ed8edc0d77914c2ee7ee
+oid sha256:94ea4d145246659349040ad95bd6972a9e718f69bd087672f8328627f4166cd3
 size 168149074

checkpoint-30/trainer_state.json

@@ -10,292 +10,292 @@
   "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
     },
     {
       "epoch": 0.008060765772748411,
-      "grad_norm": 0.5669320225715637,
+      "grad_norm": 0.5520397424697876,
       "learning_rate": 1.3600000000000002e-05,
-      "loss": 0.991,
+      "loss": 0.992,
       "step": 13
     },
     {
       "epoch": 0.008680824678344443,
-      "grad_norm": 0.49686795473098755,
+      "grad_norm": 0.381074458360672,
       "learning_rate": 1.2800000000000001e-05,
-      "loss": 0.8497,
+      "loss": 0.8504,
       "step": 14
     },
     {
       "epoch": 0.009300883583940475,
-      "grad_norm": 0.5733391642570496,
+      "grad_norm": 0.41114744544029236,
       "learning_rate": 1.2e-05,
-      "loss": 0.8679,
+      "loss": 0.8685,
       "step": 15
     },
     {
       "epoch": 0.009300883583940475,
-      "eval_loss": 0.9997775554656982,
-      "eval_runtime": 26.7666,
-      "eval_samples_per_second": 3.736,
-      "eval_steps_per_second": 3.736,
+      "eval_loss": 1.0006165504455566,
+      "eval_runtime": 26.4086,
+      "eval_samples_per_second": 3.787,
+      "eval_steps_per_second": 3.787,
       "step": 15
     },
     {
       "epoch": 0.009920942489536505,
-      "grad_norm": 0.5733532309532166,
+      "grad_norm": 0.5197786688804626,
       "learning_rate": 1.1200000000000001e-05,
-      "loss": 0.8173,
+      "loss": 0.8176,
       "step": 16
     },
     {
       "epoch": 0.010541001395132537,
-      "grad_norm": 0.375442773103714,
+      "grad_norm": 0.3905779719352722,
       "learning_rate": 1.04e-05,
-      "loss": 0.9054,
+      "loss": 0.9055,
       "step": 17
     },
     {
       "epoch": 0.01116106030072857,
-      "grad_norm": 0.4431408643722534,
+      "grad_norm": 0.43444159626960754,
       "learning_rate": 9.600000000000001e-06,
-      "loss": 0.8451,
+      "loss": 0.8452,
       "step": 18
     },
     {
       "epoch": 0.01116106030072857,
-      "eval_loss": 0.986146092414856,
-      "eval_runtime": 26.7104,
-      "eval_samples_per_second": 3.744,
-      "eval_steps_per_second": 3.744,
+      "eval_loss": 0.9866559505462646,
+      "eval_runtime": 26.3639,
+      "eval_samples_per_second": 3.793,
+      "eval_steps_per_second": 3.793,
       "step": 18
     },
     {
       "epoch": 0.011781119206324601,
-      "grad_norm": 0.5397400259971619,
+      "grad_norm": 0.5130135416984558,
       "learning_rate": 8.8e-06,
-      "loss": 0.8274,
+      "loss": 0.8276,
       "step": 19
     },
     {
       "epoch": 0.012401178111920633,
-      "grad_norm": 0.3542424440383911,
+      "grad_norm": 0.35828444361686707,
       "learning_rate": 8.000000000000001e-06,
-      "loss": 0.6586,
+      "loss": 0.6587,
       "step": 20
     },
     {
       "epoch": 0.013021237017516665,
-      "grad_norm": 0.45993754267692566,
+      "grad_norm": 0.4369220435619354,
       "learning_rate": 7.2000000000000005e-06,
-      "loss": 0.937,
+      "loss": 0.9372,
       "step": 21
     },
     {
       "epoch": 0.013021237017516665,
-      "eval_loss": 0.9746847748756409,
-      "eval_runtime": 26.6243,
-      "eval_samples_per_second": 3.756,
-      "eval_steps_per_second": 3.756,
+      "eval_loss": 0.9746317267417908,
+      "eval_runtime": 26.2985,
+      "eval_samples_per_second": 3.803,
+      "eval_steps_per_second": 3.803,
       "step": 21
     },
     {
       "epoch": 0.013641295923112695,
-      "grad_norm": 0.49625083804130554,
+      "grad_norm": 0.4957813024520874,
       "learning_rate": 6.4000000000000006e-06,
-      "loss": 0.9156,
+      "loss": 0.9153,
       "step": 22
     },
     {
       "epoch": 0.014261354828708727,
-      "grad_norm": 0.5724066495895386,
+      "grad_norm": 0.5806369185447693,
       "learning_rate": 5.600000000000001e-06,
-      "loss": 1.0178,
+      "loss": 1.0175,
       "step": 23
     },
     {
       "epoch": 0.014881413734304759,
-      "grad_norm": 0.43846315145492554,
+      "grad_norm": 0.42675939202308655,
       "learning_rate": 4.800000000000001e-06,
-      "loss": 0.8426,
+      "loss": 0.8421,
       "step": 24
     },
     {
       "epoch": 0.014881413734304759,
-      "eval_loss": 0.9658841490745544,
-      "eval_runtime": 26.6745,
-      "eval_samples_per_second": 3.749,
-      "eval_steps_per_second": 3.749,
+      "eval_loss": 0.9651661515235901,
+      "eval_runtime": 26.4248,
+      "eval_samples_per_second": 3.784,
+      "eval_steps_per_second": 3.784,
       "step": 24
     },
     {
       "epoch": 0.01550147263990079,
-      "grad_norm": 0.48291683197021484,
+      "grad_norm": 0.554572343826294,
       "learning_rate": 4.000000000000001e-06,
-      "loss": 0.8982,
+      "loss": 0.8981,
       "step": 25
     },
     {
       "epoch": 0.016121531545496823,
-      "grad_norm": 0.5055978894233704,
+      "grad_norm": 0.4890541136264801,
       "learning_rate": 3.2000000000000003e-06,
-      "loss": 0.7746,
+      "loss": 0.7742,
       "step": 26
     },
     {
       "epoch": 0.016741590451092855,
-      "grad_norm": 0.5127044916152954,
+      "grad_norm": 0.4652097821235657,
       "learning_rate": 2.4000000000000003e-06,
-      "loss": 0.7894,
+      "loss": 0.7888,
       "step": 27
     },
     {
       "epoch": 0.016741590451092855,
-      "eval_loss": 0.9593952298164368,
-      "eval_runtime": 26.6276,
-      "eval_samples_per_second": 3.756,
-      "eval_steps_per_second": 3.756,
+      "eval_loss": 0.9591427445411682,
+      "eval_runtime": 26.4791,
+      "eval_samples_per_second": 3.777,
+      "eval_steps_per_second": 3.777,
       "step": 27
     },
     {
       "epoch": 0.017361649356688887,
-      "grad_norm": 0.3996654450893402,
+      "grad_norm": 0.3710659444332123,
       "learning_rate": 1.6000000000000001e-06,
       "loss": 0.6674,
       "step": 28
     },
     {
       "epoch": 0.01798170826228492,
-      "grad_norm": 0.5467767119407654,
+      "grad_norm": 0.5322249531745911,
       "learning_rate": 8.000000000000001e-07,
-      "loss": 0.9628,
+      "loss": 0.9623,
       "step": 29
     },
     {
       "epoch": 0.01860176716788095,
-      "grad_norm": 0.6179224252700806,
+      "grad_norm": 0.6200012564659119,
       "learning_rate": 0.0,
-      "loss": 0.6993,
+      "loss": 0.699,
       "step": 30
     },
     {
       "epoch": 0.01860176716788095,
-      "eval_loss": 0.9566577672958374,
-      "eval_runtime": 26.7338,
-      "eval_samples_per_second": 3.741,
-      "eval_steps_per_second": 3.741,
+      "eval_loss": 0.9562991261482239,
+      "eval_runtime": 26.5068,
+      "eval_samples_per_second": 3.773,
+      "eval_steps_per_second": 3.773,
       "step": 30
     }
   ],

checkpoint-30/training_args.bin

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