Training in progress, step 500, checkpoint

checkpoint-500/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-500/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0c54647e44a0c65575625c756d0a394d1c95b45c74e97e0b6d6a4f1fb698c13a
+oid sha256:45453cf612ef7709ce6ab10b1fab712afbdfdd15e1462c20d1858598d61c1f4c
 size 400616360

checkpoint-500/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ddd4ac1a2d0913a7777abd18abef198bdf8103ffb6f16efa06a5f8647629f8ae
+oid sha256:4ea3e83dbe7c7019005f8e83f523a12dd492473b8494caf0a90d93a84da74861
 size 203945684

checkpoint-500/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a32ee4e2dc58afe5c7370a0a1cf1bc47d54c05c7348dc090a9726ac58a52d0ee
+oid sha256:e89d8fea670842b9c6e3146abbd17aa8f0b3453d238cdec664895a8f1032c91d
 size 14308

checkpoint-500/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:fe28e0fe7a8a43e4bb60cdeed101757578d7e994b7ece71c009f09ade5ac4f25
+oid sha256:e4714d9f3df1f7decd443af2696a47d3b78e31d3e2e5e7e021b05b678f04f782
 size 1064

checkpoint-500/trainer_state.json

@@ -10,32 +10,32 @@
   "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
     }
   ],
@@ -56,7 +56,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 1.2758795549171712e+16,
+  "total_flos": 4.124302026529997e+16,
   "train_batch_size": 1,
   "trial_name": null,
   "trial_params": null