Training in progress, step 10, checkpoint

last-checkpoint/adapter_config.json

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

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c20330890176aec2153715f7803e60009e132390f930c1102693f1f5eeb06d6d
+oid sha256:d95098d63373bb83b8f55cc9f33fe4602b0d75bf930545ec1e8aae3b194d886a
 size 27024

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8a28bc35af8f2f4a76667ce0fad7ca97312d4d373acfa90661c6af290a9c9eb7
+oid sha256:719cbdeae92fe8707cec988c306b1a04a61b4cddb4f92b5f539886c6519996d8
 size 63910

last-checkpoint/trainer_state.json

@@ -10,7 +10,7 @@
   "log_history": [
     {
       "epoch": 0.005479452054794521,
-      "grad_norm": 0.02559257671236992,
+      "grad_norm": 0.023885909467935562,
       "learning_rate": 2e-05,
       "loss": 10.3806,
       "step": 1
@@ -18,94 +18,94 @@
     {
       "epoch": 0.005479452054794521,
       "eval_loss": 10.376846313476562,
-      "eval_runtime": 0.2941,
-      "eval_samples_per_second": 132.597,
-      "eval_steps_per_second": 132.597,
+      "eval_runtime": 0.3174,
+      "eval_samples_per_second": 122.881,
+      "eval_steps_per_second": 122.881,
       "step": 1
     },
     {
       "epoch": 0.010958904109589041,
-      "grad_norm": 0.0281706303358078,
+      "grad_norm": 0.02708149142563343,
       "learning_rate": 4e-05,
       "loss": 10.3795,
       "step": 2
     },
     {
       "epoch": 0.01643835616438356,
-      "grad_norm": 0.01565679721534252,
+      "grad_norm": 0.016956843435764313,
       "learning_rate": 6e-05,
       "loss": 10.3782,
       "step": 3
     },
     {
       "epoch": 0.01643835616438356,
-      "eval_loss": 10.376832962036133,
-      "eval_runtime": 0.3989,
-      "eval_samples_per_second": 97.77,
-      "eval_steps_per_second": 97.77,
+      "eval_loss": 10.376829147338867,
+      "eval_runtime": 0.2781,
+      "eval_samples_per_second": 140.23,
+      "eval_steps_per_second": 140.23,
       "step": 3
     },
     {
       "epoch": 0.021917808219178082,
-      "grad_norm": 0.023026714101433754,
+      "grad_norm": 0.022694140672683716,
       "learning_rate": 8e-05,
       "loss": 10.3785,
       "step": 4
     },
     {
       "epoch": 0.0273972602739726,
-      "grad_norm": 0.027846675366163254,
+      "grad_norm": 0.023907892405986786,
       "learning_rate": 0.0001,
       "loss": 10.3697,
       "step": 5
     },
     {
       "epoch": 0.03287671232876712,
-      "grad_norm": 0.015995651483535767,
+      "grad_norm": 0.019220907241106033,
       "learning_rate": 0.00012,
       "loss": 10.3829,
       "step": 6
     },
     {
       "epoch": 0.03287671232876712,
-      "eval_loss": 10.376774787902832,
-      "eval_runtime": 0.3327,
-      "eval_samples_per_second": 117.209,
-      "eval_steps_per_second": 117.209,
+      "eval_loss": 10.376757621765137,
+      "eval_runtime": 0.2992,
+      "eval_samples_per_second": 130.333,
+      "eval_steps_per_second": 130.333,
       "step": 6
     },
     {
       "epoch": 0.038356164383561646,
-      "grad_norm": 0.02356347069144249,
+      "grad_norm": 0.02152101695537567,
       "learning_rate": 0.00014,
       "loss": 10.3818,
       "step": 7
     },
     {
       "epoch": 0.043835616438356165,
-      "grad_norm": 0.06583186239004135,
+      "grad_norm": 0.07344566285610199,
       "learning_rate": 0.00016,
       "loss": 10.38,
       "step": 8
     },
     {
       "epoch": 0.049315068493150684,
-      "grad_norm": 0.03688575327396393,
+      "grad_norm": 0.03503725305199623,
       "learning_rate": 0.00018,
-      "loss": 10.383,
+      "loss": 10.3829,
       "step": 9
     },
     {
       "epoch": 0.049315068493150684,
-      "eval_loss": 10.376665115356445,
-      "eval_runtime": 0.338,
-      "eval_samples_per_second": 115.399,
-      "eval_steps_per_second": 115.399,
+      "eval_loss": 10.376622200012207,
+      "eval_runtime": 0.3928,
+      "eval_samples_per_second": 99.297,
+      "eval_steps_per_second": 99.297,
       "step": 9
     },
     {
       "epoch": 0.0547945205479452,
-      "grad_norm": 0.0334353931248188,
+      "grad_norm": 0.03340437635779381,
       "learning_rate": 0.0002,
       "loss": 10.3759,
       "step": 10

last-checkpoint/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6821faa402816dc870cc3e028c580686f878eb32e4f18e75741a9277f08b1912
+oid sha256:f9e7b71b1ce7772469f28700f63d8a31b72c32cf43950aa8f0fc2a8925ca4941
 size 6776