Training in progress, step 10, checkpoint

last-checkpoint/adapter_config.json

@@ -20,11 +20,11 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "k_proj",
     "up_proj",
-    "gate_proj",
+    "k_proj",
     "q_proj",
     "v_proj",
+    "gate_proj",
     "o_proj",
     "down_proj"
   ],

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d95098d63373bb83b8f55cc9f33fe4602b0d75bf930545ec1e8aae3b194d886a
+oid sha256:a23c20cfc598df910cc5ebadea054dd898377f8dde7f5a46200e93b49879b253
 size 27024

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:719cbdeae92fe8707cec988c306b1a04a61b4cddb4f92b5f539886c6519996d8
+oid sha256:e6f482b16f5c4440daba4c766ae695017ce7f00829948c9d9793d6c077ffa801
 size 63910

last-checkpoint/trainer_state.json

@@ -10,7 +10,7 @@
   "log_history": [
     {
       "epoch": 0.005479452054794521,
-      "grad_norm": 0.023885909467935562,
+      "grad_norm": 0.023315319791436195,
       "learning_rate": 2e-05,
       "loss": 10.3806,
       "step": 1
@@ -18,94 +18,94 @@
     {
       "epoch": 0.005479452054794521,
       "eval_loss": 10.376846313476562,
-      "eval_runtime": 0.3174,
-      "eval_samples_per_second": 122.881,
-      "eval_steps_per_second": 122.881,
+      "eval_runtime": 0.2809,
+      "eval_samples_per_second": 138.857,
+      "eval_steps_per_second": 138.857,
       "step": 1
     },
     {
       "epoch": 0.010958904109589041,
-      "grad_norm": 0.02708149142563343,
+      "grad_norm": 0.02591625414788723,
       "learning_rate": 4e-05,
       "loss": 10.3795,
       "step": 2
     },
     {
       "epoch": 0.01643835616438356,
-      "grad_norm": 0.016956843435764313,
+      "grad_norm": 0.01505037397146225,
       "learning_rate": 6e-05,
       "loss": 10.3782,
       "step": 3
     },
     {
       "epoch": 0.01643835616438356,
-      "eval_loss": 10.376829147338867,
-      "eval_runtime": 0.2781,
-      "eval_samples_per_second": 140.23,
-      "eval_steps_per_second": 140.23,
+      "eval_loss": 10.376832008361816,
+      "eval_runtime": 0.3506,
+      "eval_samples_per_second": 111.233,
+      "eval_steps_per_second": 111.233,
       "step": 3
     },
     {
       "epoch": 0.021917808219178082,
-      "grad_norm": 0.022694140672683716,
+      "grad_norm": 0.021121414378285408,
       "learning_rate": 8e-05,
       "loss": 10.3785,
       "step": 4
     },
     {
       "epoch": 0.0273972602739726,
-      "grad_norm": 0.023907892405986786,
+      "grad_norm": 0.025904158130288124,
       "learning_rate": 0.0001,
       "loss": 10.3697,
       "step": 5
     },
     {
       "epoch": 0.03287671232876712,
-      "grad_norm": 0.019220907241106033,
+      "grad_norm": 0.016733916476368904,
       "learning_rate": 0.00012,
       "loss": 10.3829,
       "step": 6
     },
     {
       "epoch": 0.03287671232876712,
-      "eval_loss": 10.376757621765137,
-      "eval_runtime": 0.2992,
-      "eval_samples_per_second": 130.333,
-      "eval_steps_per_second": 130.333,
+      "eval_loss": 10.376763343811035,
+      "eval_runtime": 0.3306,
+      "eval_samples_per_second": 117.969,
+      "eval_steps_per_second": 117.969,
       "step": 6
     },
     {
       "epoch": 0.038356164383561646,
-      "grad_norm": 0.02152101695537567,
+      "grad_norm": 0.020109890028834343,
       "learning_rate": 0.00014,
       "loss": 10.3818,
       "step": 7
     },
     {
       "epoch": 0.043835616438356165,
-      "grad_norm": 0.07344566285610199,
+      "grad_norm": 0.06246686354279518,
       "learning_rate": 0.00016,
       "loss": 10.38,
       "step": 8
     },
     {
       "epoch": 0.049315068493150684,
-      "grad_norm": 0.03503725305199623,
+      "grad_norm": 0.037607599049806595,
       "learning_rate": 0.00018,
       "loss": 10.3829,
       "step": 9
     },
     {
       "epoch": 0.049315068493150684,
-      "eval_loss": 10.376622200012207,
-      "eval_runtime": 0.3928,
-      "eval_samples_per_second": 99.297,
-      "eval_steps_per_second": 99.297,
+      "eval_loss": 10.3766450881958,
+      "eval_runtime": 0.3452,
+      "eval_samples_per_second": 112.976,
+      "eval_steps_per_second": 112.976,
       "step": 9
     },
     {
       "epoch": 0.0547945205479452,
-      "grad_norm": 0.03340437635779381,
+      "grad_norm": 0.03345707431435585,
       "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:f9e7b71b1ce7772469f28700f63d8a31b72c32cf43950aa8f0fc2a8925ca4941
+oid sha256:f643f58895fbad4cde21cf2c5b4f14092feb5cf32eaa9351111da7835384d13d
 size 6776