Training in progress, step 34000, checkpoint

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:904612ad214ac2bafb26cde50b740c606861f3c1bcf698b2941883099d09b11d
+oid sha256:e0f8571f67cab7e7355d0a835675f10db403d911582b504d4460e834ec066221
 size 36730224

last-checkpoint/ar_diffusion_info.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6a26cd8e1f7baf12e2d1fc2f7a2f9179b7d0afae267d9121646e8e982b844161
+oid sha256:2e3e943eb7488886ee9103c8f80a97c673eb370cf20aa3e483dd2b7de7e0250c
 size 1736

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e16ad9d11e8833ab43804cc0637e6f5002bf05a55979e0072622efd1a8bc9af9
+oid sha256:3eb6181ab2211380e8f9d27636b01f02a9625cef49eec6124c5f793a9f83748a
 size 73588346

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5194e877da156b53363274f298068e5ae6dd5345f3a74ab2afc311cad0046d09
+oid sha256:841b8722635c3f632de2d0434a474170c723a52d8927b56279dc71208dcd5fdc
 size 14244

last-checkpoint/scaler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7402b49777be73c34eb03204101175d67711cc065bdfdb2c545471bf86e49076
+oid sha256:6f7263086319458c667c4ec9de24949320c079dc02cd6366f7b321d3562069b8
 size 988

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b7725f3433039e7c4ae6500e89a735ec0b49b1c3ae917ad92fdc3b730b7bd351
+oid sha256:94f277b3f91e9aeea511c74ffaa8a849317c65464dc1f9282b6b85fc9d3e68b8
 size 1064

last-checkpoint/trainer_state.json

@@ -2,9 +2,9 @@
   "best_global_step": 31000,
   "best_metric": 0.7226839661598206,
   "best_model_checkpoint": "./ar-diffusion-checkpoints-fixed/checkpoint-31000",
-  "epoch": 2.576724867317899,
+  "epoch": 2.615183447427121,
   "eval_steps": 250,
-  "global_step": 33500,
+  "global_step": 34000,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -5770,6 +5770,92 @@
       "eval_samples_per_second": 55.948,
       "eval_steps_per_second": 13.987,
       "step": 33500
+    },
+    {
+      "epoch": 2.580570725328821,
+      "grad_norm": 0.5088424682617188,
+      "learning_rate": 2.8387398384541465e-05,
+      "loss": 0.7113,
+      "step": 33550
+    },
+    {
+      "epoch": 2.5844165833397432,
+      "grad_norm": 0.6116564273834229,
+      "learning_rate": 2.8127678362724985e-05,
+      "loss": 0.7523,
+      "step": 33600
+    },
+    {
+      "epoch": 2.588262441350665,
+      "grad_norm": 0.6378856301307678,
+      "learning_rate": 2.78679583409085e-05,
+      "loss": 0.6924,
+      "step": 33650
+    },
+    {
+      "epoch": 2.5921082993615876,
+      "grad_norm": 1.1341512203216553,
+      "learning_rate": 2.760823831909202e-05,
+      "loss": 0.7089,
+      "step": 33700
+    },
+    {
+      "epoch": 2.5959541573725096,
+      "grad_norm": 0.8231232762336731,
+      "learning_rate": 2.7348518297275537e-05,
+      "loss": 0.7309,
+      "step": 33750
+    },
+    {
+      "epoch": 2.5959541573725096,
+      "eval_loss": 0.7776284217834473,
+      "eval_runtime": 17.9143,
+      "eval_samples_per_second": 55.821,
+      "eval_steps_per_second": 13.955,
+      "step": 33750
+    },
+    {
+      "epoch": 2.599800015383432,
+      "grad_norm": 0.7154203653335571,
+      "learning_rate": 2.7088798275459053e-05,
+      "loss": 0.7273,
+      "step": 33800
+    },
+    {
+      "epoch": 2.6036458733943544,
+      "grad_norm": 0.9213638305664062,
+      "learning_rate": 2.6829078253642576e-05,
+      "loss": 0.7628,
+      "step": 33850
+    },
+    {
+      "epoch": 2.6074917314052763,
+      "grad_norm": 1.260438084602356,
+      "learning_rate": 2.6569358231826092e-05,
+      "loss": 0.7004,
+      "step": 33900
+    },
+    {
+      "epoch": 2.6113375894161988,
+      "grad_norm": 0.9463502764701843,
+      "learning_rate": 2.6309638210009612e-05,
+      "loss": 0.6771,
+      "step": 33950
+    },
+    {
+      "epoch": 2.615183447427121,
+      "grad_norm": 0.7610837817192078,
+      "learning_rate": 2.604991818819313e-05,
+      "loss": 0.7217,
+      "step": 34000
+    },
+    {
+      "epoch": 2.615183447427121,
+      "eval_loss": 0.7775171995162964,
+      "eval_runtime": 17.9057,
+      "eval_samples_per_second": 55.848,
+      "eval_steps_per_second": 13.962,
+      "step": 34000
     }
   ],
   "logging_steps": 50,