Model save

adapter_config.json

@@ -21,12 +21,12 @@
   "revision": null,
   "target_modules": [
     "v_proj",
-    "down_proj",
     "q_proj",
     "o_proj",
     "gate_proj",
-    "k_proj",
-    "up_proj"
+    "down_proj",
+    "up_proj",
+    "k_proj"
   ],
   "task_type": "CAUSAL_LM",
   "use_dora": false,

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4d64c009abce4cc3de0aa6e85c7c185cf85bf8290a592f3027bf922d1e6da631
+oid sha256:1a8fdd8395758335ef51503713f048169d0710ae9390f11b401a0ea5f30f3828
 size 1103312056

trainer_state.json

@@ -10,67 +10,67 @@
   "log_history": [
     {
       "epoch": 0.4854368932038835,
-      "grad_norm": 0.5449720621109009,
+      "grad_norm": 0.763418436050415,
       "learning_rate": 8.786407766990292e-05,
-      "loss": 0.9459,
+      "loss": 0.9158,
       "step": 250
     },
     {
       "epoch": 0.970873786407767,
-      "grad_norm": 0.8299338817596436,
+      "grad_norm": 0.7192116975784302,
       "learning_rate": 7.572815533980583e-05,
-      "loss": 0.6303,
+      "loss": 0.4766,
       "step": 500
     },
     {
       "epoch": 1.4563106796116505,
-      "grad_norm": 0.7776351571083069,
+      "grad_norm": 0.612208902835846,
       "learning_rate": 6.359223300970875e-05,
-      "loss": 0.5388,
+      "loss": 0.4012,
       "step": 750
     },
     {
       "epoch": 1.941747572815534,
-      "grad_norm": 0.6382055878639221,
+      "grad_norm": 0.5454747080802917,
       "learning_rate": 5.145631067961165e-05,
-      "loss": 0.4846,
+      "loss": 0.3761,
       "step": 1000
     },
     {
       "epoch": 2.4271844660194173,
-      "grad_norm": 0.5591705441474915,
+      "grad_norm": 0.5297148823738098,
       "learning_rate": 3.9320388349514564e-05,
-      "loss": 0.4519,
+      "loss": 0.3648,
       "step": 1250
     },
     {
       "epoch": 2.912621359223301,
-      "grad_norm": 0.5643532872200012,
+      "grad_norm": 0.4939157962799072,
       "learning_rate": 2.7184466019417475e-05,
-      "loss": 0.4332,
+      "loss": 0.3596,
       "step": 1500
     },
     {
       "epoch": 3.3980582524271843,
-      "grad_norm": 0.430412620306015,
+      "grad_norm": 0.4684547483921051,
       "learning_rate": 1.5048543689320387e-05,
-      "loss": 0.4187,
+      "loss": 0.3529,
       "step": 1750
     },
     {
       "epoch": 3.883495145631068,
-      "grad_norm": 0.42239910364151,
+      "grad_norm": 0.4473694860935211,
       "learning_rate": 2.912621359223301e-06,
-      "loss": 0.4126,
+      "loss": 0.3498,
       "step": 2000
     },
     {
       "epoch": 4.0,
       "step": 2060,
       "total_flos": 5.02561395295531e+17,
-      "train_loss": 0.535736940439465,
-      "train_runtime": 7753.8986,
-      "train_samples_per_second": 68.011,
+      "train_loss": 0.4466323500698053,
+      "train_runtime": 7752.7447,
+      "train_samples_per_second": 68.021,
       "train_steps_per_second": 0.266
     }
   ],

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:dc47040be00cb4db9e2393b4c6323aa5309ada0c79926b9c8a5fba62f2c7773a
+oid sha256:99394c592324947bc9fc413a5d6d91863bead634191d7f3c9f1bd2639219d431
 size 5240