Training in progress, step 1071, checkpoint

checkpoint-1071/adapter_config.json

@@ -24,12 +24,12 @@
   "revision": null,
   "target_modules": [
     "up_proj",
-    "down_proj",
     "v_proj",
     "q_proj",
     "k_proj",
+    "o_proj",
     "gate_proj",
-    "o_proj"
+    "down_proj"
   ],
   "task_type": null,
   "use_dora": false,

checkpoint-1071/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ab5551299479ba911ecd5411343414b7492d9c3f8906f23f91ed7c13f47f8e92
+oid sha256:6032c9bfd3dd3cfbd399f6df89e6745b953a0cce46e3ac8f37f69ea4dd08c949
 size 400616360

checkpoint-1071/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:48a08f8a4cce2dcd93e566a4154fff311744329e01ebc7648885f66c8d4603ea
+oid sha256:a53b37eb7ae0024a6122f993df44e0b804cc99c0d1d097f4b3cd413395d27c0a
 size 203945684

checkpoint-1071/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1e977bc2b30023a922fb3707d1a2e138d9823618c90421fc6115a8c8b6143f91
+oid sha256:a70ea75b4dbc4e7ea7d24e89c83e9fa40a49068c80c4665bd494e08830072168
 size 14308

checkpoint-1071/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0e5aa052a9cf776ef4705a882d4a844e05e8bf7d99da3c7d5257cacbf49be4a3
+oid sha256:8a08e80ee90076e5bbfa1d9e624782bc839ca768f4eac3252d59d8cf20555f20
 size 1064

checkpoint-1071/trainer_state.json

@@ -10,62 +10,62 @@
   "log_history": [
     {
       "epoch": 0.2333994631812347,
-      "grad_norm": 5.120560169219971,
-      "learning_rate": 8.987402315498223e-05,
-      "loss": 0.4681,
+      "grad_norm": 3.4083237648010254,
+      "learning_rate": 8.996514402527381e-05,
+      "loss": 0.2768,
       "step": 250
     },
     {
       "epoch": 0.2333994631812347,
-      "eval_loss": 0.3557259440422058,
-      "eval_runtime": 26.5914,
-      "eval_samples_per_second": 3.272,
-      "eval_steps_per_second": 3.272,
+      "eval_loss": 0.15054132044315338,
+      "eval_runtime": 39.385,
+      "eval_samples_per_second": 2.209,
+      "eval_steps_per_second": 2.209,
       "step": 250
     },
     {
       "epoch": 0.4667989263624694,
-      "grad_norm": 3.340378999710083,
-      "learning_rate": 5.828471682626175e-05,
-      "loss": 0.3022,
+      "grad_norm": 7.858141899108887,
+      "learning_rate": 5.843391598090062e-05,
+      "loss": 0.1092,
       "step": 500
     },
     {
       "epoch": 0.4667989263624694,
-      "eval_loss": 0.3018253445625305,
-      "eval_runtime": 26.6712,
-      "eval_samples_per_second": 3.262,
-      "eval_steps_per_second": 3.262,
+      "eval_loss": 0.11498060077428818,
+      "eval_runtime": 39.2923,
+      "eval_samples_per_second": 2.214,
+      "eval_steps_per_second": 2.214,
       "step": 500
     },
     {
       "epoch": 0.7001983895437041,
-      "grad_norm": 3.158240556716919,
-      "learning_rate": 2.2174321662025427e-05,
-      "loss": 0.2736,
+      "grad_norm": 2.502495765686035,
+      "learning_rate": 2.2300178981718183e-05,
+      "loss": 0.0945,
       "step": 750
     },
     {
       "epoch": 0.7001983895437041,
-      "eval_loss": 0.2686610519886017,
-      "eval_runtime": 26.6022,
-      "eval_samples_per_second": 3.27,
-      "eval_steps_per_second": 3.27,
+      "eval_loss": 0.10474376380443573,
+      "eval_runtime": 39.2665,
+      "eval_samples_per_second": 2.216,
+      "eval_steps_per_second": 2.216,
       "step": 750
     },
     {
       "epoch": 0.9335978527249388,
-      "grad_norm": 3.1761715412139893,
-      "learning_rate": 1.2487983905362933e-06,
-      "loss": 0.2525,
+      "grad_norm": 2.4208877086639404,
+      "learning_rate": 1.2826317302122581e-06,
+      "loss": 0.0862,
       "step": 1000
     },
     {
       "epoch": 0.9335978527249388,
-      "eval_loss": 0.259181410074234,
-      "eval_runtime": 26.6606,
-      "eval_samples_per_second": 3.263,
-      "eval_steps_per_second": 3.263,
+      "eval_loss": 0.10126293450593948,
+      "eval_runtime": 39.3834,
+      "eval_samples_per_second": 2.209,
+      "eval_steps_per_second": 2.209,
       "step": 1000
     }
   ],
@@ -86,7 +86,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 2.7233286238900224e+16,
+  "total_flos": 8.796557404984934e+16,
   "train_batch_size": 1,
   "trial_name": null,
   "trial_params": null