Training in progress, step 250, checkpoint

checkpoint-250/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-250/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:cba09c7bde7a8ee13136dd21b9f64cf27c44724ed98afa07b847517580e60b68
+oid sha256:424fe34e2eff3706533d8cf803ecc170ce6ae033df1f91fdbc222ac978a76c3e
 size 400616360

checkpoint-250/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:170428c3b8e7754c3caa3015ebc991493d953edabe8ae33a84f0fdfdebd2ff6b
+oid sha256:f20d63ce27206830c8cb81522ce0c398d26bb96373c25e08234d74f6018cbe6d
 size 203945236

checkpoint-250/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:844958c05a65bd2f49ca79e6572d1d1b004147bd5fa3620e068a5ce4977bbc53
+oid sha256:51cc08418e322d2d5b4eec8c6fe4753c8de541b5877c5de90f297cf126c6971d
 size 14308

checkpoint-250/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:22707378ff6f25857dc145524fb9eae1c5e6a1f8599cf7963ffb925061e78f66
+oid sha256:bf08db5cf832e66ff85068fec2fe95dd9459a1a97d5c1be76f3e9f92898f48a3
 size 1064

checkpoint-250/trainer_state.json

@@ -10,17 +10,17 @@
   "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
     }
   ],
@@ -41,7 +41,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 6411802582659072.0,
+  "total_flos": 2.069533399612416e+16,
   "train_batch_size": 1,
   "trial_name": null,
   "trial_params": null