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:f805305c4f16d7ddb14094596e5c66389a92152130a42019b341941cd773ed8a
+oid sha256:90bc8826980d08db241b6f672fc9752d24e222935a88afa18ce47256402e7b0a
 size 80013120

checkpoint-3/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:48752c1935aac4f4802139db3c6d92cb6e6403eec986b0e00c823e8593f71cbe
+oid sha256:8894f6792437069f76df20e8215504c52c687ad0fb69f68f39e17779af478644
 size 160284754

checkpoint-3/trainer_state.json

@@ -10,31 +10,31 @@
   "log_history": [
     {
       "epoch": 0.0006200589055960316,
-      "grad_norm": 0.18599237501621246,
+      "grad_norm": 0.16878783702850342,
       "learning_rate": 4.000000000000001e-06,
-      "loss": 0.9319,
+      "loss": 0.9729,
       "step": 1
     },
     {
       "epoch": 0.0012401178111920632,
-      "grad_norm": 0.09086523950099945,
+      "grad_norm": 0.0865061804652214,
       "learning_rate": 8.000000000000001e-06,
-      "loss": 0.6463,
+      "loss": 0.6661,
       "step": 2
     },
     {
       "epoch": 0.0018601767167880949,
-      "grad_norm": 0.09325356036424637,
+      "grad_norm": 0.0920981764793396,
       "learning_rate": 1.2e-05,
-      "loss": 0.575,
+      "loss": 0.5954,
       "step": 3
     },
     {
       "epoch": 0.0018601767167880949,
-      "eval_loss": 0.7272388935089111,
-      "eval_runtime": 33.9288,
-      "eval_samples_per_second": 2.947,
-      "eval_steps_per_second": 2.947,
+      "eval_loss": 0.762992262840271,
+      "eval_runtime": 33.3972,
+      "eval_samples_per_second": 2.994,
+      "eval_steps_per_second": 2.994,
       "step": 3
     }
   ],
@@ -55,7 +55,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 4362073592610816.0,
+  "total_flos": 4400247402971136.0,
   "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:bd165a72088a81e5441087b0ae01dd72f7f14c5dad6a2e407eebc961586b0fc2
+oid sha256:1fba20505ee5f941c5085404aed8036910f9919d2bc7b2ba5ba4186849388b0c
 size 5752