Training in progress, step 33500, checkpoint

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9d5249d2e4a0e2c62be6d244a83b7f32c320808fa3501e119a143dce10a98ac6
+oid sha256:904612ad214ac2bafb26cde50b740c606861f3c1bcf698b2941883099d09b11d
 size 36730224

last-checkpoint/ar_diffusion_info.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8b531ed3efb53cfd6dee27fde322525aaa5f91dc2456c502d4cf407cfcbaaa10
+oid sha256:6a26cd8e1f7baf12e2d1fc2f7a2f9179b7d0afae267d9121646e8e982b844161
 size 1736

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c2b82d1107bd8d7ac9c326ad60110f1cc45f063c1b8edd0eba9468e959b141ca
+oid sha256:e16ad9d11e8833ab43804cc0637e6f5002bf05a55979e0072622efd1a8bc9af9
 size 73588346

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4c4264cf843d78a367918bf61d4e354d4af750891260d149d7cd76850e5ac4fd
-size 14180
+oid sha256:5194e877da156b53363274f298068e5ae6dd5345f3a74ab2afc311cad0046d09
+size 14244

last-checkpoint/scaler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5645d84b16674bcc3d2b3fbe7e60bdcce54ed7ae3b2520547c5c2900c994328f
+oid sha256:7402b49777be73c34eb03204101175d67711cc065bdfdb2c545471bf86e49076
 size 988

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d635fd5979249ab7d18e83b4f9185707744230ac67b531762b1707f47ef5b551
+oid sha256:b7725f3433039e7c4ae6500e89a735ec0b49b1c3ae917ad92fdc3b730b7bd351
 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": 2.5382662872086765,
+  "epoch": 2.576724867317899,
   "eval_steps": 250,
-  "global_step": 33000,
+  "global_step": 33500,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -5684,6 +5684,92 @@
       "eval_samples_per_second": 55.797,
       "eval_steps_per_second": 13.949,
       "step": 33000
+    },
+    {
+      "epoch": 2.5421121452195985,
+      "grad_norm": 0.7762789726257324,
+      "learning_rate": 3.098459860270628e-05,
+      "loss": 0.7279,
+      "step": 33050
+    },
+    {
+      "epoch": 2.5459580032305205,
+      "grad_norm": 0.8807786703109741,
+      "learning_rate": 3.07248785808898e-05,
+      "loss": 0.6957,
+      "step": 33100
+    },
+    {
+      "epoch": 2.549803861241443,
+      "grad_norm": 0.8823468089103699,
+      "learning_rate": 3.046515855907332e-05,
+      "loss": 0.7584,
+      "step": 33150
+    },
+    {
+      "epoch": 2.5536497192523653,
+      "grad_norm": 0.6461008191108704,
+      "learning_rate": 3.0205438537256835e-05,
+      "loss": 0.7231,
+      "step": 33200
+    },
+    {
+      "epoch": 2.5574955772632872,
+      "grad_norm": 0.9959568977355957,
+      "learning_rate": 2.9945718515440358e-05,
+      "loss": 0.7097,
+      "step": 33250
+    },
+    {
+      "epoch": 2.5574955772632872,
+      "eval_loss": 0.7804549336433411,
+      "eval_runtime": 17.9765,
+      "eval_samples_per_second": 55.628,
+      "eval_steps_per_second": 13.907,
+      "step": 33250
+    },
+    {
+      "epoch": 2.5613414352742097,
+      "grad_norm": 1.0902256965637207,
+      "learning_rate": 2.9685998493623874e-05,
+      "loss": 0.7259,
+      "step": 33300
+    },
+    {
+      "epoch": 2.565187293285132,
+      "grad_norm": 0.8527780771255493,
+      "learning_rate": 2.9426278471807394e-05,
+      "loss": 0.7138,
+      "step": 33350
+    },
+    {
+      "epoch": 2.569033151296054,
+      "grad_norm": 0.7497609257698059,
+      "learning_rate": 2.916655844999091e-05,
+      "loss": 0.6676,
+      "step": 33400
+    },
+    {
+      "epoch": 2.5728790093069764,
+      "grad_norm": 1.252274751663208,
+      "learning_rate": 2.8906838428174433e-05,
+      "loss": 0.7545,
+      "step": 33450
+    },
+    {
+      "epoch": 2.576724867317899,
+      "grad_norm": 0.8742374777793884,
+      "learning_rate": 2.864711840635795e-05,
+      "loss": 0.6727,
+      "step": 33500
+    },
+    {
+      "epoch": 2.576724867317899,
+      "eval_loss": 0.7791668176651001,
+      "eval_runtime": 17.8738,
+      "eval_samples_per_second": 55.948,
+      "eval_steps_per_second": 13.987,
+      "step": 33500
     }
   ],
   "logging_steps": 50,