Training in progress, step 9, checkpoint

checkpoint-9/adapter_config.json

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

checkpoint-9/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c501a209690f505d0379fb8312538ff1da9af8f604308a68366ce0ea3b868cb8
+oid sha256:4b542a78b261576c241682fa3c5694cbeb3e1aa40ce3b5c6217867ff77d270be
 size 137714904

checkpoint-9/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:68b9ac5e0f99074a979b2c7c70d3e1528465cb4f5863cef6d23bff02c6b20c44
+oid sha256:ea3a6146425559d7dd9e85cce037eea41951b83be6dfe1854d1dbd3f22e9bcf2
 size 275817586

checkpoint-9/trainer_state.json

@@ -10,89 +10,89 @@
   "log_history": [
     {
       "epoch": 0.0006200589055960316,
-      "grad_norm": 0.599152684211731,
+      "grad_norm": 0.6006524562835693,
       "learning_rate": 4.000000000000001e-06,
-      "loss": 1.4569,
+      "loss": 1.4363,
       "step": 1
     },
     {
       "epoch": 0.0012401178111920632,
-      "grad_norm": 0.41296300292015076,
+      "grad_norm": 0.4452102482318878,
       "learning_rate": 8.000000000000001e-06,
-      "loss": 1.1609,
+      "loss": 1.1641,
       "step": 2
     },
     {
       "epoch": 0.0018601767167880949,
-      "grad_norm": 0.3926275372505188,
+      "grad_norm": 0.418393075466156,
       "learning_rate": 1.2e-05,
-      "loss": 0.9938,
+      "loss": 1.0005,
       "step": 3
     },
     {
       "epoch": 0.0018601767167880949,
-      "eval_loss": 1.2600003480911255,
-      "eval_runtime": 49.3684,
-      "eval_samples_per_second": 2.026,
-      "eval_steps_per_second": 2.026,
+      "eval_loss": 1.2559138536453247,
+      "eval_runtime": 50.2632,
+      "eval_samples_per_second": 1.99,
+      "eval_steps_per_second": 1.99,
       "step": 3
     },
     {
       "epoch": 0.0024802356223841263,
-      "grad_norm": 0.4559924006462097,
+      "grad_norm": 0.4599858820438385,
       "learning_rate": 1.6000000000000003e-05,
-      "loss": 1.1897,
+      "loss": 1.1938,
       "step": 4
     },
     {
       "epoch": 0.0031002945279801583,
-      "grad_norm": 0.40210390090942383,
+      "grad_norm": 0.4313387870788574,
       "learning_rate": 2e-05,
-      "loss": 0.9835,
+      "loss": 0.9897,
       "step": 5
     },
     {
       "epoch": 0.0037203534335761897,
-      "grad_norm": 0.45071661472320557,
+      "grad_norm": 0.4715091586112976,
       "learning_rate": 1.9200000000000003e-05,
-      "loss": 1.112,
+      "loss": 1.1105,
       "step": 6
     },
     {
       "epoch": 0.0037203534335761897,
-      "eval_loss": 1.2538120746612549,
-      "eval_runtime": 49.623,
-      "eval_samples_per_second": 2.015,
-      "eval_steps_per_second": 2.015,
+      "eval_loss": 1.2492942810058594,
+      "eval_runtime": 49.6589,
+      "eval_samples_per_second": 2.014,
+      "eval_steps_per_second": 2.014,
       "step": 6
     },
     {
       "epoch": 0.004340412339172222,
-      "grad_norm": 0.44490283727645874,
+      "grad_norm": 0.4590495824813843,
       "learning_rate": 1.8400000000000003e-05,
-      "loss": 1.0809,
+      "loss": 1.0824,
       "step": 7
     },
     {
       "epoch": 0.004960471244768253,
-      "grad_norm": 0.6036495566368103,
+      "grad_norm": 0.6232957243919373,
       "learning_rate": 1.76e-05,
-      "loss": 1.2205,
+      "loss": 1.2176,
       "step": 8
     },
     {
       "epoch": 0.005580530150364285,
-      "grad_norm": 0.4536096751689911,
+      "grad_norm": 0.4820755422115326,
       "learning_rate": 1.6800000000000002e-05,
-      "loss": 1.0715,
+      "loss": 1.0774,
       "step": 9
     },
     {
       "epoch": 0.005580530150364285,
-      "eval_loss": 1.2402353286743164,
-      "eval_runtime": 49.777,
-      "eval_samples_per_second": 2.009,
-      "eval_steps_per_second": 2.009,
+      "eval_loss": 1.2356064319610596,
+      "eval_runtime": 49.9029,
+      "eval_samples_per_second": 2.004,
+      "eval_steps_per_second": 2.004,
       "step": 9
     }
   ],
@@ -113,7 +113,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 2.0250610651508736e+16,
+  "total_flos": 2.052529333627699e+16,
   "train_batch_size": 2,
   "trial_name": null,
   "trial_params": null

checkpoint-9/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:bf98514c81a51106969007e868d7ab8fb036a6bedc2987ddadf4a06b2a43f04f
+oid sha256:522b553cd97c4d175d86850f43dd8230595c851fb1c776adff1402d853af53e9
 size 5752