Training in progress, step 8500, checkpoint

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f8cf14ec041a3170fc93bd13962e58c3433da275c26e3ddcd248f44fe9261656
+oid sha256:64d306725c80312e5129c700db5f1b446afe8e7ae7dd6316ba0e260f92b175cc
 size 36730224

last-checkpoint/ar_diffusion_info.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c5afd0a70049fd01ffc5b0b2af6dbbfcaccc76963457c541c72cb41caeec8cf4
+oid sha256:aded19b16b58d2665657ed9436c8f8202f2f226989fa9f44064f3004c4c9f352
 size 1736

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1383466fc8ec131ed5aea99415fe444f7a641621908ee8efda1f95cd39e73d3d
+oid sha256:6ef627c205c49617bbd8eff17e12d329ee9da18b36aa3d369121d14fed0afb2b
 size 73588346

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:565e9a1f20b4e27434c54601d243feb5ec67c27b2d167d2e9e70202e4cc1bb10
+oid sha256:ca8ab81e621162f8a9202e1abd3bdf1fc40e3d6f45f592533b7f8b7c3cb0c2df
 size 14180

last-checkpoint/scaler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f4875a6ebf2ebe9f7eec596109836a4abaa1f225d707484c4b3129f5845c484e
+oid sha256:54672d09f1aec146185491fd461111d38d14cab878b7fb3a8d83bbaa34824c7e
 size 988

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:763a8e91c3a700b449d0501649e8a0112790ca7b7dbb67a7e5e5dcfd39db55d8
+oid sha256:0d578571b1ae2f1a6a8668f7ff332c5420175cf78129d97a69b053a775064e46
 size 1064

last-checkpoint/trainer_state.json

@@ -1,10 +1,10 @@
 {
-  "best_global_step": 8000,
-  "best_metric": 1.410463809967041,
-  "best_model_checkpoint": "./ar-diffusion-checkpoints-progressive-attention/checkpoint-8000",
-  "epoch": 0.6153372817475579,
+  "best_global_step": 8500,
+  "best_metric": 1.4011365175247192,
+  "best_model_checkpoint": "./ar-diffusion-checkpoints-progressive-attention/checkpoint-8500",
+  "epoch": 0.6537958618567803,
   "eval_steps": 250,
-  "global_step": 8000,
+  "global_step": 8500,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -1384,6 +1384,92 @@
       "eval_samples_per_second": 59.34,
       "eval_steps_per_second": 14.835,
       "step": 8000
+    },
+    {
+      "epoch": 0.6191831397584802,
+      "grad_norm": 1.6925851106643677,
+      "learning_rate": 0.0001608134431083292,
+      "loss": 1.3827,
+      "step": 8050
+    },
+    {
+      "epoch": 0.6230289977694023,
+      "grad_norm": 1.419701337814331,
+      "learning_rate": 0.00016055372308651275,
+      "loss": 1.4598,
+      "step": 8100
+    },
+    {
+      "epoch": 0.6268748557803245,
+      "grad_norm": 1.6113284826278687,
+      "learning_rate": 0.00016029400306469626,
+      "loss": 1.4369,
+      "step": 8150
+    },
+    {
+      "epoch": 0.6307207137912468,
+      "grad_norm": 1.8418887853622437,
+      "learning_rate": 0.00016003428304287978,
+      "loss": 1.405,
+      "step": 8200
+    },
+    {
+      "epoch": 0.634566571802169,
+      "grad_norm": 0.9828066825866699,
+      "learning_rate": 0.00015977456302106332,
+      "loss": 1.3865,
+      "step": 8250
+    },
+    {
+      "epoch": 0.634566571802169,
+      "eval_loss": 1.4110686779022217,
+      "eval_runtime": 16.91,
+      "eval_samples_per_second": 59.137,
+      "eval_steps_per_second": 14.784,
+      "step": 8250
+    },
+    {
+      "epoch": 0.6384124298130913,
+      "grad_norm": 1.8815706968307495,
+      "learning_rate": 0.00015951484299924683,
+      "loss": 1.3965,
+      "step": 8300
+    },
+    {
+      "epoch": 0.6422582878240135,
+      "grad_norm": 0.975824773311615,
+      "learning_rate": 0.00015925512297743034,
+      "loss": 1.4273,
+      "step": 8350
+    },
+    {
+      "epoch": 0.6461041458349358,
+      "grad_norm": 1.3958059549331665,
+      "learning_rate": 0.00015899540295561385,
+      "loss": 1.4204,
+      "step": 8400
+    },
+    {
+      "epoch": 0.649950003845858,
+      "grad_norm": 1.059217929840088,
+      "learning_rate": 0.00015873568293379737,
+      "loss": 1.4439,
+      "step": 8450
+    },
+    {
+      "epoch": 0.6537958618567803,
+      "grad_norm": 1.491314172744751,
+      "learning_rate": 0.0001584759629119809,
+      "loss": 1.5128,
+      "step": 8500
+    },
+    {
+      "epoch": 0.6537958618567803,
+      "eval_loss": 1.4011365175247192,
+      "eval_runtime": 16.9623,
+      "eval_samples_per_second": 58.954,
+      "eval_steps_per_second": 14.739,
+      "step": 8500
     }
   ],
   "logging_steps": 50,