Training in progress, step 750, checkpoint

checkpoint-750/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-750/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:af5be3d4407b508f7bbd71e63f8f85437f1a211cb86149dc2c94bcb2257953b0
+oid sha256:d0fea320deeae03449ceb42b94b9a610cbdb03c76d94e3ccbcce59b5fc3e01de
 size 400616360

checkpoint-750/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5c97ebb9c71cb917584a654f918e0d6bbaa56a2f25c0ed415f263d53cea2f131
+oid sha256:897125d3b508e12716cf0e0b98fa57a13bf9a2901506004c4326870aa7d9aed5
 size 203945684

checkpoint-750/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:fbaa3dfcb09925315e426f0caa9dc7c0debd4699fcb7a9f78c14885fcec107fa
+oid sha256:ac34fb399b008853e821c1b660c6db3bfd79068b5dd7e6b5706f701bd2d9ddb7
 size 14308

checkpoint-750/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1a6c20427749d2ad2e92b7bff20374e4a6d0785ca9f30521693441afaa0f1715
+oid sha256:903ae037e72717b21e8330e7e2b503be1e4eaf79929b7c6ba9768e98c9028124
 size 1064

checkpoint-750/trainer_state.json

@@ -10,47 +10,47 @@
   "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
     }
   ],
@@ -71,7 +71,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 1.91166786561024e+16,
+  "total_flos": 6.169714429149389e+16,
   "train_batch_size": 1,
   "trial_name": null,
   "trial_params": null