Training in progress, epoch 2, checkpoint

checkpoint-7198/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d1f99348a0f0ec6f75d89bc6c6673ecea0120e47af106bb0032284b951fead15
+oid sha256:24fac6fdccc64b54767936fe43dd399513a1b5d6f9a78b4e1ed7d0dd961eb734
 size 541459256

checkpoint-7198/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ac95698934c75384f39adaad1508b4f4d21cbb6590023114bd3151db45d31576
+oid sha256:44c94dfb515b3183a5c80c6f673e063664bcfbfd035b67e867307a30001a87b7
 size 33662074

checkpoint-7198/rng_state_0.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:429687bc10a78b0faddf5066a56b8a8ebacd4ecec2cf2b445861ba5877917b23
+oid sha256:272b45e4f0943c023882837870c5cc33fd0ca410bac72c542a1d8dcaeb473797
 size 15024

checkpoint-7198/rng_state_1.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:047c8237008b1fa7b8cbb7e59eb7a35a8332f967e20b68feb13d5bd1e5e9cb83
+oid sha256:a9af7f851bc0dbde2f5de344f420f3c32da9c08804725ae984e31caa6ab8ad9f
 size 15024

checkpoint-7198/rng_state_2.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ba2b901719d85858f975d6c9c852a1350cefa32b2e80a94cbada70119c445708
+oid sha256:4056c05f901e446416791497a93fd97cfb90fba06d3313d34db0775e2bef68ca
 size 15024

checkpoint-7198/rng_state_3.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9a8842a6023b395d14baeb0bdd9a3c7e17d0c2a1d56165776b0b9dc03d55bb7c
+oid sha256:8fd59dcd683e8bd44904f922d218a8863f30e73670df1a4899871cfbc08f31b6
 size 15024

checkpoint-7198/trainer_state.json

@@ -10,172 +10,172 @@
   "log_history": [
     {
       "epoch": 0.13892747985551543,
-      "grad_norm": 0.6994414925575256,
-      "learning_rate": 0.00039537834583680656,
-      "loss": 0.2943,
+      "grad_norm": 0.4299773573875427,
+      "learning_rate": 0.0003953968695007873,
+      "loss": 0.2967,
       "step": 500
     },
     {
       "epoch": 0.2000555709919422,
-      "eval_loss": 0.21401312947273254,
-      "eval_runtime": 16.5317,
-      "eval_samples_per_second": 30.245,
-      "eval_steps_per_second": 3.811,
+      "eval_loss": 0.1983761489391327,
+      "eval_runtime": 16.4512,
+      "eval_samples_per_second": 30.393,
+      "eval_steps_per_second": 3.83,
       "step": 720
     },
     {
       "epoch": 0.27785495971103086,
-      "grad_norm": 0.5945971608161926,
-      "learning_rate": 0.00039075669167361305,
-      "loss": 0.1986,
+      "grad_norm": 0.5706949830055237,
+      "learning_rate": 0.0003907659535056034,
+      "loss": 0.1959,
       "step": 1000
     },
     {
       "epoch": 0.4001111419838844,
-      "eval_loss": 0.17625106871128082,
-      "eval_runtime": 16.4823,
-      "eval_samples_per_second": 30.336,
+      "eval_loss": 0.17383727431297302,
+      "eval_runtime": 16.4842,
+      "eval_samples_per_second": 30.332,
       "eval_steps_per_second": 3.822,
       "step": 1440
     },
     {
       "epoch": 0.41678243956654626,
-      "grad_norm": 0.3619824945926666,
-      "learning_rate": 0.0003861257756784292,
-      "loss": 0.1802,
+      "grad_norm": 0.5392869710922241,
+      "learning_rate": 0.0003861350375104196,
+      "loss": 0.1723,
       "step": 1500
     },
     {
       "epoch": 0.5557099194220617,
-      "grad_norm": 0.40362638235092163,
-      "learning_rate": 0.00038149485968324534,
-      "loss": 0.1686,
+      "grad_norm": 0.5005412697792053,
+      "learning_rate": 0.0003815041215152357,
+      "loss": 0.1636,
       "step": 2000
     },
     {
       "epoch": 0.6001667129758266,
-      "eval_loss": 0.16156445443630219,
-      "eval_runtime": 16.3535,
-      "eval_samples_per_second": 30.575,
-      "eval_steps_per_second": 3.852,
+      "eval_loss": 0.16229495406150818,
+      "eval_runtime": 16.4963,
+      "eval_samples_per_second": 30.31,
+      "eval_steps_per_second": 3.819,
       "step": 2160
     },
     {
       "epoch": 0.6946373992775771,
-      "grad_norm": 0.4962793290615082,
-      "learning_rate": 0.0003768639436880615,
-      "loss": 0.1577,
+      "grad_norm": 0.370914489030838,
+      "learning_rate": 0.0003768732055200519,
+      "loss": 0.1539,
       "step": 2500
     },
     {
       "epoch": 0.8002222839677688,
-      "eval_loss": 0.15043224394321442,
-      "eval_runtime": 16.3325,
-      "eval_samples_per_second": 30.614,
-      "eval_steps_per_second": 3.857,
+      "eval_loss": 0.1537286937236786,
+      "eval_runtime": 16.4662,
+      "eval_samples_per_second": 30.365,
+      "eval_steps_per_second": 3.826,
       "step": 2880
     },
     {
       "epoch": 0.8335648791330925,
-      "grad_norm": 0.41315382719039917,
-      "learning_rate": 0.0003722330276928777,
-      "loss": 0.1493,
+      "grad_norm": 0.41378697752952576,
+      "learning_rate": 0.000372242289524868,
+      "loss": 0.1445,
       "step": 3000
     },
     {
       "epoch": 0.972492358988608,
-      "grad_norm": 0.37953054904937744,
+      "grad_norm": 0.4000154137611389,
       "learning_rate": 0.0003676113735296842,
-      "loss": 0.1477,
+      "loss": 0.1384,
       "step": 3500
     },
     {
       "epoch": 1.000277854959711,
-      "eval_loss": 0.14538082480430603,
-      "eval_runtime": 16.4466,
-      "eval_samples_per_second": 30.401,
-      "eval_steps_per_second": 3.831,
+      "eval_loss": 0.15115150809288025,
+      "eval_runtime": 16.8021,
+      "eval_samples_per_second": 29.758,
+      "eval_steps_per_second": 3.75,
       "step": 3600
     },
     {
       "epoch": 1.1114198388441234,
-      "grad_norm": 0.4146048128604889,
+      "grad_norm": 0.4774770140647888,
       "learning_rate": 0.00036298045753450036,
-      "loss": 0.1318,
+      "loss": 0.1317,
       "step": 4000
     },
     {
       "epoch": 1.2003334259516532,
-      "eval_loss": 0.13961325585842133,
-      "eval_runtime": 16.5835,
-      "eval_samples_per_second": 30.15,
-      "eval_steps_per_second": 3.799,
+      "eval_loss": 0.14330309629440308,
+      "eval_runtime": 16.4814,
+      "eval_samples_per_second": 30.337,
+      "eval_steps_per_second": 3.822,
       "step": 4320
     },
     {
       "epoch": 1.2503473186996388,
-      "grad_norm": 0.6914293766021729,
+      "grad_norm": 0.4988621175289154,
       "learning_rate": 0.0003583495415393165,
-      "loss": 0.1303,
+      "loss": 0.1283,
       "step": 4500
     },
     {
       "epoch": 1.3892747985551543,
-      "grad_norm": 0.39365154504776,
-      "learning_rate": 0.00035371862554413265,
-      "loss": 0.1269,
+      "grad_norm": 0.5688238739967346,
+      "learning_rate": 0.000353727887376123,
+      "loss": 0.1237,
       "step": 5000
     },
     {
       "epoch": 1.4003889969435954,
-      "eval_loss": 0.14090509712696075,
-      "eval_runtime": 16.4413,
-      "eval_samples_per_second": 30.411,
-      "eval_steps_per_second": 3.832,
+      "eval_loss": 0.13683326542377472,
+      "eval_runtime": 16.5065,
+      "eval_samples_per_second": 30.291,
+      "eval_steps_per_second": 3.817,
       "step": 5040
     },
     {
       "epoch": 1.5282022784106695,
-      "grad_norm": 0.5620034337043762,
+      "grad_norm": 0.5433902740478516,
       "learning_rate": 0.00034909697138093914,
-      "loss": 0.1263,
+      "loss": 0.1174,
       "step": 5500
     },
     {
       "epoch": 1.6004445679355377,
-      "eval_loss": 0.13119302690029144,
-      "eval_runtime": 16.5618,
-      "eval_samples_per_second": 30.19,
-      "eval_steps_per_second": 3.804,
+      "eval_loss": 0.13955478370189667,
+      "eval_runtime": 16.4989,
+      "eval_samples_per_second": 30.305,
+      "eval_steps_per_second": 3.818,
       "step": 5760
     },
     {
       "epoch": 1.667129758266185,
-      "grad_norm": 1.5726152658462524,
+      "grad_norm": 0.5300644040107727,
       "learning_rate": 0.0003444660553857553,
-      "loss": 0.1258,
+      "loss": 0.1178,
       "step": 6000
     },
     {
       "epoch": 1.8005001389274797,
-      "eval_loss": 0.12919241189956665,
-      "eval_runtime": 16.4031,
-      "eval_samples_per_second": 30.482,
-      "eval_steps_per_second": 3.841,
+      "eval_loss": 0.13158197700977325,
+      "eval_runtime": 16.533,
+      "eval_samples_per_second": 30.243,
+      "eval_steps_per_second": 3.811,
       "step": 6480
     },
     {
       "epoch": 1.8060572381217006,
-      "grad_norm": 0.9997938871383667,
+      "grad_norm": 0.3573897182941437,
       "learning_rate": 0.0003398351393905715,
-      "loss": 0.1212,
+      "loss": 0.113,
       "step": 6500
     },
     {
       "epoch": 1.9449847179772157,
-      "grad_norm": 0.4134954810142517,
+      "grad_norm": 0.5019258260726929,
       "learning_rate": 0.00033520422339538766,
-      "loss": 0.1201,
+      "loss": 0.1134,
       "step": 7000
     }
   ],
@@ -196,7 +196,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 1.3464410121463398e+18,
+  "total_flos": 1.3492710519970202e+18,
   "train_batch_size": 4,
   "trial_name": null,
   "trial_params": null

checkpoint-7198/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:58052fbf3a2b07f2a6024b5cc28db88f1f0e48109a11483aa716d00657e9906e
+oid sha256:b2784cc9ebb113293b9d7c5af564dbf5463d67b520bf149c8840105fec4706ec
 size 5496