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:d25f7767596f8bd173a4f4b4764752c3d12de9d88e2faf2fb805da5a858efe28
+oid sha256:6250382341285c6f4fa6cd2e06e72e75037937e27f3016f4706a165ea168d540
 size 131146352

checkpoint-9/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:96a5d63184d63d2dc2b29e0d74076461e2c763f3b80a320ff88377129b0125b6
+oid sha256:83552538889a59d7a2f2686ed2b0e8bcb7dfdc2cca1fa20f9be8cb61e873a118
 size 262615650

checkpoint-9/trainer_state.json

@@ -10,89 +10,89 @@
   "log_history": [
     {
       "epoch": 0.0006200589055960316,
-      "grad_norm": 0.39027148485183716,
+      "grad_norm": 0.5875879526138306,
       "learning_rate": 4.000000000000001e-06,
-      "loss": 1.1401,
+      "loss": 1.179,
       "step": 1
     },
     {
       "epoch": 0.0012401178111920632,
-      "grad_norm": 0.1772935390472412,
+      "grad_norm": 0.2559925317764282,
       "learning_rate": 8.000000000000001e-06,
-      "loss": 0.9133,
+      "loss": 0.9296,
       "step": 2
     },
     {
       "epoch": 0.0018601767167880949,
-      "grad_norm": 0.192855104804039,
+      "grad_norm": 0.27975356578826904,
       "learning_rate": 1.2e-05,
-      "loss": 0.8036,
+      "loss": 0.8212,
       "step": 3
     },
     {
       "epoch": 0.0018601767167880949,
-      "eval_loss": 0.993510365486145,
-      "eval_runtime": 46.5106,
-      "eval_samples_per_second": 2.15,
-      "eval_steps_per_second": 2.15,
+      "eval_loss": 1.0227885246276855,
+      "eval_runtime": 47.9985,
+      "eval_samples_per_second": 2.083,
+      "eval_steps_per_second": 2.083,
       "step": 3
     },
     {
       "epoch": 0.0024802356223841263,
-      "grad_norm": 0.23705194890499115,
+      "grad_norm": 0.4077853858470917,
       "learning_rate": 1.6000000000000003e-05,
-      "loss": 0.9389,
+      "loss": 0.9553,
       "step": 4
     },
     {
       "epoch": 0.0031002945279801583,
-      "grad_norm": 0.22702816128730774,
+      "grad_norm": 0.3481099307537079,
       "learning_rate": 2e-05,
-      "loss": 0.7836,
+      "loss": 0.8071,
       "step": 5
     },
     {
       "epoch": 0.0037203534335761897,
-      "grad_norm": 0.22573529183864594,
+      "grad_norm": 0.33538541197776794,
       "learning_rate": 1.9200000000000003e-05,
-      "loss": 0.8874,
+      "loss": 0.9046,
       "step": 6
     },
     {
       "epoch": 0.0037203534335761897,
-      "eval_loss": 0.9887735843658447,
-      "eval_runtime": 47.2769,
-      "eval_samples_per_second": 2.115,
-      "eval_steps_per_second": 2.115,
+      "eval_loss": 1.0122885704040527,
+      "eval_runtime": 48.1202,
+      "eval_samples_per_second": 2.078,
+      "eval_steps_per_second": 2.078,
       "step": 6
     },
     {
       "epoch": 0.004340412339172222,
-      "grad_norm": 0.231398344039917,
+      "grad_norm": 0.2848486602306366,
       "learning_rate": 1.8400000000000003e-05,
-      "loss": 0.8755,
+      "loss": 0.8963,
       "step": 7
     },
     {
       "epoch": 0.004960471244768253,
-      "grad_norm": 0.28229379653930664,
+      "grad_norm": 0.29836800694465637,
       "learning_rate": 1.76e-05,
-      "loss": 0.971,
+      "loss": 0.9869,
       "step": 8
     },
     {
       "epoch": 0.005580530150364285,
-      "grad_norm": 0.21991951763629913,
+      "grad_norm": 0.22438423335552216,
       "learning_rate": 1.6800000000000002e-05,
-      "loss": 0.8546,
+      "loss": 0.8704,
       "step": 9
     },
     {
       "epoch": 0.005580530150364285,
-      "eval_loss": 0.9774181842803955,
-      "eval_runtime": 47.3982,
-      "eval_samples_per_second": 2.11,
-      "eval_steps_per_second": 2.11,
+      "eval_loss": 0.996995210647583,
+      "eval_runtime": 48.1511,
+      "eval_samples_per_second": 2.077,
+      "eval_steps_per_second": 2.077,
       "step": 9
     }
   ],
@@ -113,7 +113,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 2.031134308190208e+16,
+  "total_flos": 2.059258727153664e+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 sha256:31008ae8f5fa3525c3b5855165b7a47a0cdb47e438b843022563680f47a27521
+oid sha256:b7726c9626d0a377a638b6674780a9e8d32aed5a1fc017243f1823b069fc15ac
 size 5688