vvduc03/lora-llava-3b

adapter_config.json

@@ -19,9 +19,9 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "q_proj",
     "k_proj",
-    "v_proj"
+    "v_proj",
+    "q_proj"
   ],
   "task_type": null
 }

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:943d0f3883083f3d9820d0c1abd7af634249a9b793eae7189fe2599276553575
+oid sha256:aaf64a0c26cb48b3c7c523d0cf1b5a68debca5533ad7777755e0a6f735296173
 size 47237856

trainer_state.json

@@ -10,90 +10,90 @@
   "log_history": [
     {
       "epoch": 20.0,
-      "grad_norm": 0.38967031240463257,
+      "grad_norm": 0.38066378235816956,
       "learning_rate": 8e-05,
-      "loss": 0.1783,
+      "loss": 0.1782,
       "step": 20
     },
     {
       "epoch": 40.0,
-      "grad_norm": 0.009602731093764305,
+      "grad_norm": 0.008850287646055222,
       "learning_rate": 0.00016,
-      "loss": 0.0093,
+      "loss": 0.0092,
       "step": 40
     },
     {
       "epoch": 60.0,
-      "grad_norm": 0.000580370076932013,
+      "grad_norm": 0.0006731236935593188,
       "learning_rate": 0.0001866666666666667,
       "loss": 0.0,
       "step": 60
     },
     {
       "epoch": 80.0,
-      "grad_norm": 0.00020515041251201183,
+      "grad_norm": 0.00021651879069395363,
       "learning_rate": 0.00016,
       "loss": 0.0,
       "step": 80
     },
     {
       "epoch": 100.0,
-      "grad_norm": 0.00012300981325097382,
+      "grad_norm": 0.00013040985504630953,
       "learning_rate": 0.00013333333333333334,
       "loss": 0.0,
       "step": 100
     },
     {
       "epoch": 120.0,
-      "grad_norm": 9.383377619087696e-05,
+      "grad_norm": 9.823329310165718e-05,
       "learning_rate": 0.00010666666666666667,
       "loss": 0.0,
       "step": 120
     },
     {
       "epoch": 140.0,
-      "grad_norm": 8.079586405074224e-05,
+      "grad_norm": 8.462095138384029e-05,
       "learning_rate": 8e-05,
       "loss": 0.0,
       "step": 140
     },
     {
       "epoch": 160.0,
-      "grad_norm": 7.359922165051103e-05,
+      "grad_norm": 7.46678197174333e-05,
       "learning_rate": 5.333333333333333e-05,
       "loss": 0.0,
       "step": 160
     },
     {
       "epoch": 180.0,
-      "grad_norm": 6.797131209168583e-05,
+      "grad_norm": 6.831328937551007e-05,
       "learning_rate": 2.6666666666666667e-05,
       "loss": 0.0,
       "step": 180
     },
     {
       "epoch": 200.0,
-      "grad_norm": 6.64097024127841e-05,
+      "grad_norm": 6.943791231606156e-05,
       "learning_rate": 0.0,
       "loss": 0.0,
       "step": 200
     },
     {
       "epoch": 200.0,
-      "eval_loss": 1.2874310414190404e-05,
-      "eval_runtime": 0.4025,
-      "eval_samples_per_second": 2.485,
-      "eval_steps_per_second": 2.485,
+      "eval_loss": 1.2972288459422998e-05,
+      "eval_runtime": 0.4099,
+      "eval_samples_per_second": 2.439,
+      "eval_steps_per_second": 2.439,
       "step": 200
     },
     {
       "epoch": 200.0,
       "step": 200,
       "total_flos": 1572670596710400.0,
-      "train_loss": 0.018764802380228502,
-      "train_runtime": 147.4566,
-      "train_samples_per_second": 21.701,
-      "train_steps_per_second": 1.356
+      "train_loss": 0.018746382653935145,
+      "train_runtime": 146.7988,
+      "train_samples_per_second": 21.799,
+      "train_steps_per_second": 1.362
     }
   ],
   "logging_steps": 20,

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:97a0369021c64d53315ac38195fdc982e14c3b117e8662c6734eec42e120a343
+oid sha256:de8523b3104361a991d2d52bf0069c4f84e685c4be6c6aada009e81b1c325515
 size 5112