Training in progress, step 41000

adapter_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8b58822fe1e8262161f44bc61bc0ad0a9e7afd47e4892189dfa46e0a57e6dee1
+oid sha256:bc2e66ae2124d056bbe13c91ab39f1f815d18e8ea9fb247d1e4d278f8c33f9b5
 size 500897101

{checkpoint-37000 → checkpoint-40000/adapter_model}/adapter_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:dd9546c32a7c69f32e36c612415bdd464231e08b43267c0c48a31c7f40c0d408
+oid sha256:8b58822fe1e8262161f44bc61bc0ad0a9e7afd47e4892189dfa46e0a57e6dee1
 size 500897101

{checkpoint-37000/adapter_model → checkpoint-41000}/adapter_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:dd9546c32a7c69f32e36c612415bdd464231e08b43267c0c48a31c7f40c0d408
+oid sha256:bc2e66ae2124d056bbe13c91ab39f1f815d18e8ea9fb247d1e4d278f8c33f9b5
 size 500897101

{checkpoint-37000 → checkpoint-41000}/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:66752d28f3b7ca12e1b483bbb5bd48fa49998d34e026c908100c9cd122ff144f
+oid sha256:880772dedf09c7dcdad380b983a49bcfa2b0629343a9422152a901d98876bcf0
 size 1001723453

{checkpoint-37000 → checkpoint-41000}/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:12a1e84087c196e7e618617b884516dfa94b6d7a7e668c1d280c6b4c8239f845
+oid sha256:ab7da1c5092196900ff656e5706f0c4cdbd895ffbf8492b429fc2abdd89c40e0
 size 14575

{checkpoint-37000 → checkpoint-41000}/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7d7046cc5cdc134adf08950434c6180a09920d2a73dfbc0c6da7d0543135abf2
+oid sha256:a6d79d87ff5c6682a567ef4f654bba9bc3e905f13772c597a0ad097e7782c969
 size 627

{checkpoint-37000 → checkpoint-41000}/trainer_state.json

@@ -1,8 +1,8 @@
 {
-  "best_metric": 0.5555862188339233,
-  "best_model_checkpoint": "./qlora-out/checkpoint-37000",
-  "epoch": 1.3795160508556727,
-  "global_step": 37000,
+  "best_metric": 0.539782702922821,
+  "best_model_checkpoint": "./qlora-out/checkpoint-41000",
+  "epoch": 1.5286529212184483,
+  "global_step": 41000,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -2522,11 +2522,283 @@
       "eval_samples_per_second": 0.409,
       "eval_steps_per_second": 0.409,
       "step": 37000
+    },
+    {
+      "epoch": 1.38,
+      "learning_rate": 0.00011221705714827408,
+      "loss": 0.4994,
+      "step": 37100
+    },
+    {
+      "epoch": 1.39,
+      "learning_rate": 0.00011182940214493858,
+      "loss": 0.4826,
+      "step": 37200
+    },
+    {
+      "epoch": 1.39,
+      "learning_rate": 0.0001114415667661215,
+      "loss": 0.5173,
+      "step": 37300
+    },
+    {
+      "epoch": 1.39,
+      "learning_rate": 0.00011105355692556165,
+      "loss": 0.4992,
+      "step": 37400
+    },
+    {
+      "epoch": 1.4,
+      "learning_rate": 0.00011066537853965788,
+      "loss": 0.5155,
+      "step": 37500
+    },
+    {
+      "epoch": 1.4,
+      "learning_rate": 0.00011027703752737913,
+      "loss": 0.5118,
+      "step": 37600
+    },
+    {
+      "epoch": 1.41,
+      "learning_rate": 0.00010988853981017393,
+      "loss": 0.5078,
+      "step": 37700
+    },
+    {
+      "epoch": 1.41,
+      "learning_rate": 0.00010949989131188043,
+      "loss": 0.506,
+      "step": 37800
+    },
+    {
+      "epoch": 1.41,
+      "learning_rate": 0.00010911109795863581,
+      "loss": 0.5074,
+      "step": 37900
+    },
+    {
+      "epoch": 1.42,
+      "learning_rate": 0.00010872216567878599,
+      "loss": 0.4837,
+      "step": 38000
+    },
+    {
+      "epoch": 1.42,
+      "eval_loss": 0.5529844164848328,
+      "eval_runtime": 1281.6978,
+      "eval_samples_per_second": 0.423,
+      "eval_steps_per_second": 0.423,
+      "step": 38000
+    },
+    {
+      "epoch": 1.42,
+      "learning_rate": 0.00010833310040279531,
+      "loss": 0.5161,
+      "step": 38100
+    },
+    {
+      "epoch": 1.42,
+      "learning_rate": 0.00010794390806315602,
+      "loss": 0.4929,
+      "step": 38200
+    },
+    {
+      "epoch": 1.43,
+      "learning_rate": 0.0001075545945942978,
+      "loss": 0.4819,
+      "step": 38300
+    },
+    {
+      "epoch": 1.43,
+      "learning_rate": 0.00010716516593249742,
+      "loss": 0.4594,
+      "step": 38400
+    },
+    {
+      "epoch": 1.44,
+      "learning_rate": 0.00010677562801578798,
+      "loss": 0.4802,
+      "step": 38500
+    },
+    {
+      "epoch": 1.44,
+      "learning_rate": 0.00010638598678386864,
+      "loss": 0.4793,
+      "step": 38600
+    },
+    {
+      "epoch": 1.44,
+      "learning_rate": 0.00010599624817801383,
+      "loss": 0.4912,
+      "step": 38700
+    },
+    {
+      "epoch": 1.45,
+      "learning_rate": 0.0001056064181409828,
+      "loss": 0.519,
+      "step": 38800
+    },
+    {
+      "epoch": 1.45,
+      "learning_rate": 0.00010521650261692886,
+      "loss": 0.4993,
+      "step": 38900
+    },
+    {
+      "epoch": 1.45,
+      "learning_rate": 0.00010482650755130898,
+      "loss": 0.4688,
+      "step": 39000
+    },
+    {
+      "epoch": 1.45,
+      "eval_loss": 0.5486682057380676,
+      "eval_runtime": 1265.4476,
+      "eval_samples_per_second": 0.428,
+      "eval_steps_per_second": 0.428,
+      "step": 39000
+    },
+    {
+      "epoch": 1.46,
+      "learning_rate": 0.00010443643889079282,
+      "loss": 0.4901,
+      "step": 39100
+    },
+    {
+      "epoch": 1.46,
+      "learning_rate": 0.00010404630258317236,
+      "loss": 0.4783,
+      "step": 39200
+    },
+    {
+      "epoch": 1.47,
+      "learning_rate": 0.00010365610457727095,
+      "loss": 0.5026,
+      "step": 39300
+    },
+    {
+      "epoch": 1.47,
+      "learning_rate": 0.00010326585082285279,
+      "loss": 0.4602,
+      "step": 39400
+    },
+    {
+      "epoch": 1.47,
+      "learning_rate": 0.00010287554727053215,
+      "loss": 0.5299,
+      "step": 39500
+    },
+    {
+      "epoch": 1.48,
+      "learning_rate": 0.00010248519987168252,
+      "loss": 0.5135,
+      "step": 39600
+    },
+    {
+      "epoch": 1.48,
+      "learning_rate": 0.00010209481457834616,
+      "loss": 0.4792,
+      "step": 39700
+    },
+    {
+      "epoch": 1.48,
+      "learning_rate": 0.0001017043973431429,
+      "loss": 0.5004,
+      "step": 39800
+    },
+    {
+      "epoch": 1.49,
+      "learning_rate": 0.00010131395411917979,
+      "loss": 0.5013,
+      "step": 39900
+    },
+    {
+      "epoch": 1.49,
+      "learning_rate": 0.00010092349085996011,
+      "loss": 0.501,
+      "step": 40000
+    },
+    {
+      "epoch": 1.49,
+      "eval_loss": 0.5438262820243835,
+      "eval_runtime": 1262.0322,
+      "eval_samples_per_second": 0.429,
+      "eval_steps_per_second": 0.429,
+      "step": 40000
+    },
+    {
+      "epoch": 1.5,
+      "learning_rate": 0.0001005330135192927,
+      "loss": 0.5043,
+      "step": 40100
+    },
+    {
+      "epoch": 1.5,
+      "learning_rate": 0.000100142528051201,
+      "loss": 0.4888,
+      "step": 40200
+    },
+    {
+      "epoch": 1.5,
+      "learning_rate": 9.975204040983244e-05,
+      "loss": 0.4866,
+      "step": 40300
+    },
+    {
+      "epoch": 1.51,
+      "learning_rate": 9.936155654936761e-05,
+      "loss": 0.4845,
+      "step": 40400
+    },
+    {
+      "epoch": 1.51,
+      "learning_rate": 9.897108242392937e-05,
+      "loss": 0.5027,
+      "step": 40500
+    },
+    {
+      "epoch": 1.51,
+      "learning_rate": 9.858062398749225e-05,
+      "loss": 0.5048,
+      "step": 40600
+    },
+    {
+      "epoch": 1.52,
+      "learning_rate": 9.819018719379143e-05,
+      "loss": 0.4985,
+      "step": 40700
+    },
+    {
+      "epoch": 1.52,
+      "learning_rate": 9.77997779962322e-05,
+      "loss": 0.4861,
+      "step": 40800
+    },
+    {
+      "epoch": 1.52,
+      "learning_rate": 9.740940234779903e-05,
+      "loss": 0.4743,
+      "step": 40900
+    },
+    {
+      "epoch": 1.53,
+      "learning_rate": 9.701906620096474e-05,
+      "loss": 0.5036,
+      "step": 41000
+    },
+    {
+      "epoch": 1.53,
+      "eval_loss": 0.539782702922821,
+      "eval_runtime": 1250.1524,
+      "eval_samples_per_second": 0.434,
+      "eval_steps_per_second": 0.434,
+      "step": 41000
     }
   ],
   "max_steps": 80463,
   "num_train_epochs": 3,
-  "total_flos": 1.0375131020284969e+19,
+  "total_flos": 1.1498079849470607e+19,
   "trial_name": null,
   "trial_params": null
 }