Training in progress, step 48000

adapter_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1462088913437dad86ad0338ce743c5fe033920125cb93487967374e6f4a70ef
+oid sha256:0b1b8dd34210954de9184030e20cd78b268cce955c25bebb508877d570510dc8
 size 500897101

{checkpoint-44000 → checkpoint-47000/adapter_model}/adapter_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:514c6190fe2d77ecd2f858e923b2c7bec20b75928cee4654bcf237afb090151e
+oid sha256:1462088913437dad86ad0338ce743c5fe033920125cb93487967374e6f4a70ef
 size 500897101

{checkpoint-44000/adapter_model → checkpoint-48000}/adapter_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:514c6190fe2d77ecd2f858e923b2c7bec20b75928cee4654bcf237afb090151e
+oid sha256:0b1b8dd34210954de9184030e20cd78b268cce955c25bebb508877d570510dc8
 size 500897101

{checkpoint-44000 → checkpoint-48000}/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c358d7d09699b43fe08d08ae6928e3d239f686bf90709539476817b4526e8156
+oid sha256:393e8e8fc0b5556322d804ca4a6de96e8280404a43aa59493475b670a51ef388
 size 1001723453

{checkpoint-44000 → checkpoint-48000}/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0edbb7285779af4f2b9fba78491956e43d69071f56ba58ccc546a425b3c47636
+oid sha256:f8eba8077364b81e6cb0367d7296e7b2436a03990fb56044fd20b3df4eb45e85
 size 14575

{checkpoint-44000 → checkpoint-48000}/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4a53b5ec62bc72de547a30cfd6edc79f3190c27ad9d0a8e71792bbf9d33ce9b8
+oid sha256:0fb30b1012e7113d92b74989b92e243fdad9b6199049250d25a88fcba15d630c
 size 627

{checkpoint-44000 → checkpoint-48000}/trainer_state.json

@@ -1,8 +1,8 @@
 {
-  "best_metric": 0.5285405516624451,
-  "best_model_checkpoint": "./qlora-out/checkpoint-44000",
-  "epoch": 1.6405055739905299,
-  "global_step": 44000,
+  "best_metric": 0.5149570107460022,
+  "best_model_checkpoint": "./qlora-out/checkpoint-48000",
+  "epoch": 1.7896424443533052,
+  "global_step": 48000,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -2998,11 +2998,283 @@
       "eval_samples_per_second": 0.426,
       "eval_steps_per_second": 0.426,
       "step": 44000
+    },
+    {
+      "epoch": 1.64,
+      "learning_rate": 8.497066196958097e-05,
+      "loss": 0.4684,
+      "step": 44100
+    },
+    {
+      "epoch": 1.65,
+      "learning_rate": 8.458472497126595e-05,
+      "loss": 0.4923,
+      "step": 44200
+    },
+    {
+      "epoch": 1.65,
+      "learning_rate": 8.41990230260518e-05,
+      "loss": 0.492,
+      "step": 44300
+    },
+    {
+      "epoch": 1.66,
+      "learning_rate": 8.38135620151465e-05,
+      "loss": 0.4814,
+      "step": 44400
+    },
+    {
+      "epoch": 1.66,
+      "learning_rate": 8.342834781608424e-05,
+      "loss": 0.4753,
+      "step": 44500
+    },
+    {
+      "epoch": 1.66,
+      "learning_rate": 8.304338630263579e-05,
+      "loss": 0.5032,
+      "step": 44600
+    },
+    {
+      "epoch": 1.67,
+      "learning_rate": 8.265868334471895e-05,
+      "loss": 0.4838,
+      "step": 44700
+    },
+    {
+      "epoch": 1.67,
+      "learning_rate": 8.227424480830907e-05,
+      "loss": 0.4463,
+      "step": 44800
+    },
+    {
+      "epoch": 1.67,
+      "learning_rate": 8.189007655534959e-05,
+      "loss": 0.458,
+      "step": 44900
+    },
+    {
+      "epoch": 1.68,
+      "learning_rate": 8.150618444366255e-05,
+      "loss": 0.4812,
+      "step": 45000
+    },
+    {
+      "epoch": 1.68,
+      "eval_loss": 0.5250784754753113,
+      "eval_runtime": 1265.3534,
+      "eval_samples_per_second": 0.428,
+      "eval_steps_per_second": 0.428,
+      "step": 45000
+    },
+    {
+      "epoch": 1.68,
+      "learning_rate": 8.112257432685958e-05,
+      "loss": 0.4743,
+      "step": 45100
+    },
+    {
+      "epoch": 1.69,
+      "learning_rate": 8.073925205425225e-05,
+      "loss": 0.4925,
+      "step": 45200
+    },
+    {
+      "epoch": 1.69,
+      "learning_rate": 8.035622347076312e-05,
+      "loss": 0.4748,
+      "step": 45300
+    },
+    {
+      "epoch": 1.69,
+      "learning_rate": 7.997349441683657e-05,
+      "loss": 0.4745,
+      "step": 45400
+    },
+    {
+      "epoch": 1.7,
+      "learning_rate": 7.959107072834971e-05,
+      "loss": 0.449,
+      "step": 45500
+    },
+    {
+      "epoch": 1.7,
+      "learning_rate": 7.920895823652346e-05,
+      "loss": 0.4753,
+      "step": 45600
+    },
+    {
+      "epoch": 1.7,
+      "learning_rate": 7.882716276783352e-05,
+      "loss": 0.457,
+      "step": 45700
+    },
+    {
+      "epoch": 1.71,
+      "learning_rate": 7.844569014392172e-05,
+      "loss": 0.488,
+      "step": 45800
+    },
+    {
+      "epoch": 1.71,
+      "learning_rate": 7.806454618150698e-05,
+      "loss": 0.4782,
+      "step": 45900
+    },
+    {
+      "epoch": 1.72,
+      "learning_rate": 7.768373669229688e-05,
+      "loss": 0.4751,
+      "step": 46000
+    },
+    {
+      "epoch": 1.72,
+      "eval_loss": 0.5225542187690735,
+      "eval_runtime": 1263.3928,
+      "eval_samples_per_second": 0.429,
+      "eval_steps_per_second": 0.429,
+      "step": 46000
+    },
+    {
+      "epoch": 1.72,
+      "learning_rate": 7.730326748289895e-05,
+      "loss": 0.4525,
+      "step": 46100
+    },
+    {
+      "epoch": 1.72,
+      "learning_rate": 7.6923144354732e-05,
+      "loss": 0.483,
+      "step": 46200
+    },
+    {
+      "epoch": 1.73,
+      "learning_rate": 7.654337310393787e-05,
+      "loss": 0.4941,
+      "step": 46300
+    },
+    {
+      "epoch": 1.73,
+      "learning_rate": 7.616395952129287e-05,
+      "loss": 0.4637,
+      "step": 46400
+    },
+    {
+      "epoch": 1.73,
+      "learning_rate": 7.578490939211965e-05,
+      "loss": 0.4874,
+      "step": 46500
+    },
+    {
+      "epoch": 1.74,
+      "learning_rate": 7.540622849619883e-05,
+      "loss": 0.4815,
+      "step": 46600
+    },
+    {
+      "epoch": 1.74,
+      "learning_rate": 7.5027922607681e-05,
+      "loss": 0.4973,
+      "step": 46700
+    },
+    {
+      "epoch": 1.74,
+      "learning_rate": 7.46499974949985e-05,
+      "loss": 0.5003,
+      "step": 46800
+    },
+    {
+      "epoch": 1.75,
+      "learning_rate": 7.427245892077775e-05,
+      "loss": 0.4331,
+      "step": 46900
+    },
+    {
+      "epoch": 1.75,
+      "learning_rate": 7.389531264175103e-05,
+      "loss": 0.4691,
+      "step": 47000
+    },
+    {
+      "epoch": 1.75,
+      "eval_loss": 0.5184645652770996,
+      "eval_runtime": 1306.5029,
+      "eval_samples_per_second": 0.415,
+      "eval_steps_per_second": 0.415,
+      "step": 47000
+    },
+    {
+      "epoch": 1.76,
+      "learning_rate": 7.351856440866895e-05,
+      "loss": 0.4875,
+      "step": 47100
+    },
+    {
+      "epoch": 1.76,
+      "learning_rate": 7.314221996621279e-05,
+      "loss": 0.4615,
+      "step": 47200
+    },
+    {
+      "epoch": 1.76,
+      "learning_rate": 7.276628505290663e-05,
+      "loss": 0.4522,
+      "step": 47300
+    },
+    {
+      "epoch": 1.77,
+      "learning_rate": 7.239076540103013e-05,
+      "loss": 0.4861,
+      "step": 47400
+    },
+    {
+      "epoch": 1.77,
+      "learning_rate": 7.2015666736531e-05,
+      "loss": 0.4676,
+      "step": 47500
+    },
+    {
+      "epoch": 1.77,
+      "learning_rate": 7.164099477893768e-05,
+      "loss": 0.4451,
+      "step": 47600
+    },
+    {
+      "epoch": 1.78,
+      "learning_rate": 7.126675524127217e-05,
+      "loss": 0.456,
+      "step": 47700
+    },
+    {
+      "epoch": 1.78,
+      "learning_rate": 7.089295382996294e-05,
+      "loss": 0.4676,
+      "step": 47800
+    },
+    {
+      "epoch": 1.79,
+      "learning_rate": 7.05195962447578e-05,
+      "loss": 0.4687,
+      "step": 47900
+    },
+    {
+      "epoch": 1.79,
+      "learning_rate": 7.014668817863719e-05,
+      "loss": 0.4415,
+      "step": 48000
+    },
+    {
+      "epoch": 1.79,
+      "eval_loss": 0.5149570107460022,
+      "eval_runtime": 1238.5109,
+      "eval_samples_per_second": 0.438,
+      "eval_steps_per_second": 0.438,
+      "step": 48000
     }
   ],
   "max_steps": 80463,
   "num_train_epochs": 3,
-  "total_flos": 1.2339087922998313e+19,
+  "total_flos": 1.3459285466101432e+19,
   "trial_name": null,
   "trial_params": null
 }