Training in progress, step 26000, checkpoint

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b2e333303611bdcfe53850b8382bc5c6096b913ec43665a75beb67494ee5bcb0
+oid sha256:7380e911b47a593b6e6e294e7827a8de2e4e3f7b1b83607bf62e1583c2d7f7a3
 size 36730224

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:66d78210b05e1c87c136a3c08a8840809f76d0b17ce40128a18c0fba7403a0f6
+oid sha256:7af7e30906067165a3669470320cb653b4ffd33b1d8a610f3716b94a2f716fce
 size 73588346

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:dc69f04c57ac233f68440607df3025c3f527f698c64e56350e0ea45b99be0781
+oid sha256:35c260873e53de579d314af7f23d76fd9a02c023129e1ca7c8f1d3d5f4fae1d0
 size 14244

last-checkpoint/scaler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:92ee59d15723c28ba4726ae8f5244e001c8225a696c80d61ce6d100eb1539722
+oid sha256:60e9205691ce5148ac9b5c101f7716ac13075c3c3735582239bfb45d9a860e75
 size 988

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:cfd4ecc27bf3b1377a8048f15a651b111fa15d35d5db24035804ad1d2f0064c3
+oid sha256:4e14edeb9123c5eec7e26cdb8d77f3aaa83054aadc39c2057a183c66b1f00314
 size 1064

last-checkpoint/trainer_state.json

@@ -2,9 +2,9 @@
   "best_global_step": 24500,
   "best_metric": 1.4431298971176147,
   "best_model_checkpoint": "./ar-diffusion-checkpoints/checkpoint-24500",
-  "epoch": 1.9613875855703409,
+  "epoch": 1.9998461656795632,
   "eval_steps": 250,
-  "global_step": 25500,
+  "global_step": 26000,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -4394,6 +4394,92 @@
       "eval_samples_per_second": 55.388,
       "eval_steps_per_second": 13.847,
       "step": 25500
+    },
+    {
+      "epoch": 1.9652334435812628,
+      "grad_norm": 1.8318700790405273,
+      "learning_rate": 9.079077975074445e-06,
+      "loss": 1.42,
+      "step": 25550
+    },
+    {
+      "epoch": 1.9690793015921852,
+      "grad_norm": 1.7966841459274292,
+      "learning_rate": 8.821734495055328e-06,
+      "loss": 1.3236,
+      "step": 25600
+    },
+    {
+      "epoch": 1.9729251596031074,
+      "grad_norm": 0.7579635977745056,
+      "learning_rate": 8.564391015036211e-06,
+      "loss": 1.3957,
+      "step": 25650
+    },
+    {
+      "epoch": 1.9767710176140296,
+      "grad_norm": 1.4515990018844604,
+      "learning_rate": 8.307047535017094e-06,
+      "loss": 1.3347,
+      "step": 25700
+    },
+    {
+      "epoch": 1.980616875624952,
+      "grad_norm": 1.5671380758285522,
+      "learning_rate": 8.049704054997977e-06,
+      "loss": 1.4624,
+      "step": 25750
+    },
+    {
+      "epoch": 1.980616875624952,
+      "eval_loss": 1.450337290763855,
+      "eval_runtime": 17.9548,
+      "eval_samples_per_second": 55.695,
+      "eval_steps_per_second": 13.924,
+      "step": 25750
+    },
+    {
+      "epoch": 1.984462733635874,
+      "grad_norm": 1.7020714282989502,
+      "learning_rate": 7.79236057497886e-06,
+      "loss": 1.3822,
+      "step": 25800
+    },
+    {
+      "epoch": 1.9883085916467964,
+      "grad_norm": 1.297658920288086,
+      "learning_rate": 7.535017094959743e-06,
+      "loss": 1.4008,
+      "step": 25850
+    },
+    {
+      "epoch": 1.9921544496577186,
+      "grad_norm": 1.8151623010635376,
+      "learning_rate": 7.277673614940627e-06,
+      "loss": 1.4408,
+      "step": 25900
+    },
+    {
+      "epoch": 1.9960003076686408,
+      "grad_norm": 0.8869682550430298,
+      "learning_rate": 7.02033013492151e-06,
+      "loss": 1.4767,
+      "step": 25950
+    },
+    {
+      "epoch": 1.9998461656795632,
+      "grad_norm": 1.898775339126587,
+      "learning_rate": 6.762986654902392e-06,
+      "loss": 1.5032,
+      "step": 26000
+    },
+    {
+      "epoch": 1.9998461656795632,
+      "eval_loss": 1.4542045593261719,
+      "eval_runtime": 18.0059,
+      "eval_samples_per_second": 55.537,
+      "eval_steps_per_second": 13.884,
+      "step": 26000
     }
   ],
   "logging_steps": 50,