Training in progress, step 300, checkpoint

last-checkpoint/adapter_config.json

@@ -22,8 +22,8 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "k_proj",
-    "q_proj"
+    "q_proj",
+    "k_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:dffb58ead3a8261ad808ac56947c3cd3660dde80949934a70ee4007ecf573495
+oid sha256:6dfef42af6edf8fb1e0e66951afb0e090b6c26af8a9b470267578100c76b2fc5
 size 527048968

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b6d510ad6a7aadf1acfd0a7799fb32b8ccdd1d51e237076309c4a4f363e0b0c8
+oid sha256:9a5617aae1346c4823faf0f4babbace95896dff1ebbb74839af4ffbfe6b73e71
 size 1054135994

last-checkpoint/trainer_state.json

@@ -1,5 +1,5 @@
 {
-  "best_metric": 0.7312111854553223,
+  "best_metric": 0.7317642569541931,
   "best_model_checkpoint": "./output/checkpoint-300",
   "epoch": 13.043478260869565,
   "eval_steps": 150,
@@ -10,228 +10,228 @@
   "log_history": [
     {
       "epoch": 0.43478260869565216,
-      "grad_norm": 1.5021647214889526,
+      "grad_norm": 1.5021412372589111,
       "learning_rate": 3e-06,
-      "loss": 0.9064,
+      "loss": 0.9061,
       "step": 10
     },
     {
       "epoch": 0.8695652173913043,
-      "grad_norm": 1.6793277263641357,
+      "grad_norm": 1.687071442604065,
       "learning_rate": 6e-06,
-      "loss": 0.9024,
+      "loss": 0.9026,
       "step": 20
     },
     {
       "epoch": 1.3043478260869565,
-      "grad_norm": 1.7296056747436523,
+      "grad_norm": 1.7296651601791382,
       "learning_rate": 9e-06,
-      "loss": 0.9,
+      "loss": 0.9003,
       "step": 30
     },
     {
       "epoch": 1.7391304347826086,
-      "grad_norm": 1.4458602666854858,
+      "grad_norm": 1.4536513090133667,
       "learning_rate": 1.2e-05,
-      "loss": 0.9086,
+      "loss": 0.9093,
       "step": 40
     },
     {
       "epoch": 2.1739130434782608,
-      "grad_norm": 1.3516699075698853,
+      "grad_norm": 1.351776361465454,
       "learning_rate": 1.5e-05,
-      "loss": 0.8359,
+      "loss": 0.8358,
       "step": 50
     },
     {
       "epoch": 2.608695652173913,
-      "grad_norm": 2.03114914894104,
+      "grad_norm": 2.031055212020874,
       "learning_rate": 1.8e-05,
-      "loss": 0.8892,
+      "loss": 0.8894,
       "step": 60
     },
     {
       "epoch": 3.0434782608695654,
-      "grad_norm": 1.4920729398727417,
+      "grad_norm": 1.4920436143875122,
       "learning_rate": 2.1e-05,
-      "loss": 0.8913,
+      "loss": 0.8915,
       "step": 70
     },
     {
       "epoch": 3.4782608695652173,
-      "grad_norm": 1.7293280363082886,
+      "grad_norm": 1.729347586631775,
       "learning_rate": 2.4e-05,
-      "loss": 0.8234,
+      "loss": 0.8237,
       "step": 80
     },
     {
       "epoch": 3.9130434782608696,
-      "grad_norm": 1.4240705966949463,
+      "grad_norm": 1.4314346313476562,
       "learning_rate": 2.7000000000000002e-05,
       "loss": 0.8529,
       "step": 90
     },
     {
       "epoch": 4.3478260869565215,
-      "grad_norm": 1.3656291961669922,
+      "grad_norm": 1.3657989501953125,
       "learning_rate": 3e-05,
       "loss": 0.8649,
       "step": 100
     },
     {
       "epoch": 4.782608695652174,
-      "grad_norm": 2.198136806488037,
+      "grad_norm": 2.1989877223968506,
       "learning_rate": 2.999999702723963e-05,
       "loss": 0.8227,
       "step": 110
     },
     {
       "epoch": 5.217391304347826,
-      "grad_norm": 1.0726988315582275,
+      "grad_norm": 1.0724585056304932,
       "learning_rate": 2.9999988108959687e-05,
-      "loss": 0.765,
+      "loss": 0.7655,
       "step": 120
     },
     {
       "epoch": 5.6521739130434785,
-      "grad_norm": 1.5672078132629395,
+      "grad_norm": 1.5675039291381836,
       "learning_rate": 2.9999973245163716e-05,
       "loss": 0.7413,
       "step": 130
     },
     {
       "epoch": 6.086956521739131,
-      "grad_norm": 1.9071016311645508,
+      "grad_norm": 1.9068934917449951,
       "learning_rate": 2.99999524358576e-05,
-      "loss": 0.7655,
+      "loss": 0.7653,
       "step": 140
     },
     {
       "epoch": 6.521739130434782,
-      "grad_norm": 1.1219961643218994,
+      "grad_norm": 1.121341586112976,
       "learning_rate": 2.9999925681049593e-05,
       "loss": 0.7855,
       "step": 150
     },
     {
       "epoch": 6.521739130434782,
-      "eval_loss": 0.7960985898971558,
-      "eval_runtime": 0.5137,
-      "eval_samples_per_second": 19.465,
-      "eval_steps_per_second": 19.465,
+      "eval_loss": 0.7960079908370972,
+      "eval_runtime": 0.4677,
+      "eval_samples_per_second": 21.382,
+      "eval_steps_per_second": 21.382,
       "step": 150
     },
     {
       "epoch": 6.956521739130435,
-      "grad_norm": 1.5404783487319946,
+      "grad_norm": 1.5398368835449219,
       "learning_rate": 2.9999892980750297e-05,
-      "loss": 0.6588,
+      "loss": 0.6587,
       "step": 160
     },
     {
       "epoch": 7.391304347826087,
-      "grad_norm": 1.3464210033416748,
+      "grad_norm": 1.346014142036438,
       "learning_rate": 2.9999854334972675e-05,
       "loss": 0.739,
       "step": 170
     },
     {
       "epoch": 7.826086956521739,
-      "grad_norm": 1.7264870405197144,
+      "grad_norm": 1.7263554334640503,
       "learning_rate": 2.999980974373204e-05,
-      "loss": 0.7292,
+      "loss": 0.729,
       "step": 180
     },
     {
       "epoch": 8.26086956521739,
-      "grad_norm": 1.5393368005752563,
+      "grad_norm": 1.5398695468902588,
       "learning_rate": 2.9999759207046075e-05,
-      "loss": 0.6244,
+      "loss": 0.6246,
       "step": 190
     },
     {
       "epoch": 8.695652173913043,
-      "grad_norm": 1.7342770099639893,
+      "grad_norm": 1.7423065900802612,
       "learning_rate": 2.9999702724934804e-05,
-      "loss": 0.6764,
+      "loss": 0.6762,
       "step": 200
     },
     {
       "epoch": 9.130434782608695,
-      "grad_norm": 1.0416548252105713,
+      "grad_norm": 1.0418452024459839,
       "learning_rate": 2.999964029742062e-05,
-      "loss": 0.6521,
+      "loss": 0.6522,
       "step": 210
     },
     {
       "epoch": 9.565217391304348,
-      "grad_norm": 1.2193585634231567,
+      "grad_norm": 1.2121878862380981,
       "learning_rate": 2.9999571924528263e-05,
-      "loss": 0.5592,
+      "loss": 0.559,
       "step": 220
     },
     {
       "epoch": 10.0,
-      "grad_norm": 1.5286610126495361,
+      "grad_norm": 1.5282926559448242,
       "learning_rate": 2.9999497606284837e-05,
-      "loss": 0.756,
+      "loss": 0.7558,
       "step": 230
     },
     {
       "epoch": 10.434782608695652,
-      "grad_norm": 1.427451252937317,
+      "grad_norm": 1.4222936630249023,
       "learning_rate": 2.9999417342719796e-05,
-      "loss": 0.7118,
+      "loss": 0.7119,
       "step": 240
     },
     {
       "epoch": 10.869565217391305,
-      "grad_norm": 0.9794074296951294,
+      "grad_norm": 0.9751431941986084,
       "learning_rate": 2.9999331133864956e-05,
-      "loss": 0.5897,
+      "loss": 0.59,
       "step": 250
     },
     {
       "epoch": 11.304347826086957,
-      "grad_norm": 1.194082260131836,
+      "grad_norm": 1.194357991218567,
       "learning_rate": 2.9999238979754485e-05,
-      "loss": 0.6548,
+      "loss": 0.6547,
       "step": 260
     },
     {
       "epoch": 11.73913043478261,
-      "grad_norm": 1.0489437580108643,
+      "grad_norm": 1.049717903137207,
       "learning_rate": 2.999914088042492e-05,
       "loss": 0.6474,
       "step": 270
     },
     {
       "epoch": 12.173913043478262,
-      "grad_norm": 1.3105570077896118,
+      "grad_norm": 1.3123618364334106,
       "learning_rate": 2.9999036835915132e-05,
       "loss": 0.5936,
       "step": 280
     },
     {
       "epoch": 12.608695652173914,
-      "grad_norm": 1.0830979347229004,
+      "grad_norm": 1.0834040641784668,
       "learning_rate": 2.9998926846266365e-05,
       "loss": 0.6328,
       "step": 290
     },
     {
       "epoch": 13.043478260869565,
-      "grad_norm": 1.3943493366241455,
+      "grad_norm": 1.3875536918640137,
       "learning_rate": 2.9998810911522213e-05,
-      "loss": 0.5806,
+      "loss": 0.5807,
       "step": 300
     },
     {
       "epoch": 13.043478260869565,
-      "eval_loss": 0.7312111854553223,
-      "eval_runtime": 0.6239,
-      "eval_samples_per_second": 16.028,
-      "eval_steps_per_second": 16.028,
+      "eval_loss": 0.7317642569541931,
+      "eval_runtime": 0.6288,
+      "eval_samples_per_second": 15.902,
+      "eval_steps_per_second": 15.902,
       "step": 300
     }
   ],

last-checkpoint/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0c84fd1341bca21f546fb34f0984c2737ed54e8dbbe2ce01666781675b951db1
+oid sha256:a4e26a07b48ab9f4eeb67e7d951dfa1f71b1fc6e6c955372cec1a3c50fa8b9a2
 size 5368