Training in progress, step 39000

adapter_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4cf723872c571317009cef0bd95ce4c0da7de441496aabcb6c8a6e48bbc7e73e
+oid sha256:4ac02c8b60ff2af2043c8bbd7012fcc6335dd99c02ab7c2e9f0bff405da9b9af
 size 500897101

{checkpoint-35000 → checkpoint-38000/adapter_model}/adapter_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3bcaf4cf9f9bc20718f7e19c06c7145a78a3a5c6c1ca1a845231e7f1741b7f0e
+oid sha256:4cf723872c571317009cef0bd95ce4c0da7de441496aabcb6c8a6e48bbc7e73e
 size 500897101

{checkpoint-35000/adapter_model → checkpoint-39000}/adapter_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3bcaf4cf9f9bc20718f7e19c06c7145a78a3a5c6c1ca1a845231e7f1741b7f0e
+oid sha256:4ac02c8b60ff2af2043c8bbd7012fcc6335dd99c02ab7c2e9f0bff405da9b9af
 size 500897101

{checkpoint-35000 → checkpoint-39000}/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:021299a59b6ef922c41c510f3f8b93b7333417d0f6d05edbc3721b651b76908a
+oid sha256:bc61c2163348cf1f5d8718ee8096100da2ac348ad64cc9a1eddf53f8f686ebc6
 size 1001723453

{checkpoint-35000 → checkpoint-39000}/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e07a8bcb98397480852e842446a8ddbd7b1c726fe43695a8f26d8722c832cf30
+oid sha256:e9a2ffdbf1de428093fdd3697cf4e6d99a44644c793a039982af26ede4025ab2
 size 14575

{checkpoint-35000 → checkpoint-39000}/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:70880cdf77014daa95aad41e2ce4c4dffd35eef0882e2600a10e5ef00bca6a86
+oid sha256:f53c19898f710d27205821fa8efbe0cfe03a9cf5ec3fc464b9123705a68749b6
 size 627

{checkpoint-35000 → checkpoint-39000}/trainer_state.json

@@ -1,8 +1,8 @@
 {
-  "best_metric": 0.562954843044281,
-  "best_model_checkpoint": "./qlora-out/checkpoint-35000",
-  "epoch": 1.3049476156742852,
-  "global_step": 35000,
+  "best_metric": 0.5486682057380676,
+  "best_model_checkpoint": "./qlora-out/checkpoint-39000",
+  "epoch": 1.4540844860370605,
+  "global_step": 39000,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -2386,11 +2386,283 @@
       "eval_samples_per_second": 0.433,
       "eval_steps_per_second": 0.433,
       "step": 35000
+    },
+    {
+      "epoch": 1.31,
+      "learning_rate": 0.00011992319858898077,
+      "loss": 0.4952,
+      "step": 35100
+    },
+    {
+      "epoch": 1.31,
+      "learning_rate": 0.0001195403880867545,
+      "loss": 0.5157,
+      "step": 35200
+    },
+    {
+      "epoch": 1.32,
+      "learning_rate": 0.00011915727963143922,
+      "loss": 0.4973,
+      "step": 35300
+    },
+    {
+      "epoch": 1.32,
+      "learning_rate": 0.00011877387906469721,
+      "loss": 0.4884,
+      "step": 35400
+    },
+    {
+      "epoch": 1.32,
+      "learning_rate": 0.00011839019223264489,
+      "loss": 0.5017,
+      "step": 35500
+    },
+    {
+      "epoch": 1.33,
+      "learning_rate": 0.00011800622498576363,
+      "loss": 0.5157,
+      "step": 35600
+    },
+    {
+      "epoch": 1.33,
+      "learning_rate": 0.00011762198317881059,
+      "loss": 0.4774,
+      "step": 35700
+    },
+    {
+      "epoch": 1.33,
+      "learning_rate": 0.0001172374726707295,
+      "loss": 0.4855,
+      "step": 35800
+    },
+    {
+      "epoch": 1.34,
+      "learning_rate": 0.00011685269932456115,
+      "loss": 0.5134,
+      "step": 35900
+    },
+    {
+      "epoch": 1.34,
+      "learning_rate": 0.00011646766900735422,
+      "loss": 0.5143,
+      "step": 36000
+    },
+    {
+      "epoch": 1.34,
+      "eval_loss": 0.5594063997268677,
+      "eval_runtime": 1270.0722,
+      "eval_samples_per_second": 0.427,
+      "eval_steps_per_second": 0.427,
+      "step": 36000
+    },
+    {
+      "epoch": 1.35,
+      "learning_rate": 0.00011608238759007561,
+      "loss": 0.5268,
+      "step": 36100
+    },
+    {
+      "epoch": 1.35,
+      "learning_rate": 0.00011569686094752101,
+      "loss": 0.5179,
+      "step": 36200
+    },
+    {
+      "epoch": 1.35,
+      "learning_rate": 0.00011531109495822545,
+      "loss": 0.5236,
+      "step": 36300
+    },
+    {
+      "epoch": 1.36,
+      "learning_rate": 0.00011492509550437339,
+      "loss": 0.5197,
+      "step": 36400
+    },
+    {
+      "epoch": 1.36,
+      "learning_rate": 0.0001145388684717092,
+      "loss": 0.5109,
+      "step": 36500
+    },
+    {
+      "epoch": 1.36,
+      "learning_rate": 0.00011415241974944744,
+      "loss": 0.5126,
+      "step": 36600
+    },
+    {
+      "epoch": 1.37,
+      "learning_rate": 0.00011376575523018296,
+      "loss": 0.501,
+      "step": 36700
+    },
+    {
+      "epoch": 1.37,
+      "learning_rate": 0.00011337888080980115,
+      "loss": 0.4888,
+      "step": 36800
+    },
+    {
+      "epoch": 1.38,
+      "learning_rate": 0.00011299180238738789,
+      "loss": 0.5324,
+      "step": 36900
+    },
+    {
+      "epoch": 1.38,
+      "learning_rate": 0.00011260452586513981,
+      "loss": 0.5053,
+      "step": 37000
+    },
+    {
+      "epoch": 1.38,
+      "eval_loss": 0.5555862188339233,
+      "eval_runtime": 1324.5375,
+      "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
     }
   ],
   "max_steps": 80463,
   "num_train_epochs": 3,
-  "total_flos": 9.815104047716352e+18,
+  "total_flos": 1.0937258471303578e+19,
   "trial_name": null,
   "trial_params": null
 }