Training in progress, step 3, checkpoint

checkpoint-3/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-3/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9f7ab3a3fdfb383ae1d90482a19663f46ebe01a1d6b2363d9397a19420333fa8
+oid sha256:b9f54ee7de946e3c9849a8b8e69fe49844816d1eca4adaba52e0c1777e5ed629
 size 131146352

checkpoint-3/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:aed0c02e70bb67fd812d5484f061680fe48f74788f8bdd655d93b188acb79c27
+oid sha256:436f71cf0b57dfb1d67bd300344350cff996c9ea25b9b6b44c4bb2049e9066d6
 size 262615650

checkpoint-3/trainer_state.json

@@ -10,31 +10,31 @@
   "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
     }
   ],
@@ -55,7 +55,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 6959134812119040.0,
+  "total_flos": 7052712733655040.0,
   "train_batch_size": 2,
   "trial_name": null,
   "trial_params": null

checkpoint-3/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:31008ae8f5fa3525c3b5855165b7a47a0cdb47e438b843022563680f47a27521
+oid sha256:b7726c9626d0a377a638b6674780a9e8d32aed5a1fc017243f1823b069fc15ac
 size 5688