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:dce2cfe7ef78c1b081a422071300fead0d75106053b5f049a42ada26761ae754
+oid sha256:4cb6c8a0b2c106d46167c80ad940e05faf7be6e8c488347878b12d9c1613eaaa
 size 36730224

last-checkpoint/ar_diffusion_info.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a31437fd444dcaf0935d9c64d6e3d1702f2bb5519ddb595a0202c4f2bd2a612f
+oid sha256:3c93f6a5a30f422904bef55476365b3cddae7d20bf919682601832e7adc836a6
 size 1736

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:763237292b314fde98770ac7b874a48945d2f1abca3b2c84598219bbffc7efd9
+oid sha256:3791734d0548551bbd8fb3d5c2f3391fdc29d0cbe0edc356708c62363c6ae0c1
 size 73588346

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e8bfd573f2544c60f45f4769693f9ca807bef468213a4a6127137e9e334ee28a
+oid sha256:7830bd2b40308208f3fc50f0359995bd60a796cf38e987d5a42e513619b75942
 size 14244

last-checkpoint/scaler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:38e8022c7f3002679e6fa3f97f4fe89fb5c796e6e8d168c42ccf78f18e8bd32d
+oid sha256:db57d5e262103dec37d91f40b9f442834db40beb8e2908aae10c8b555c4b4e03
 size 988

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a609d4c8b7c9c250baf5aaeedbae25a95f3a4513a558852d1de4e86b5f18ff63
+oid sha256:7257b05f4dbf8d80135c409b68234e541f9d914e30b27a9ad576dc6da07856df
 size 1064

last-checkpoint/trainer_state.json

@@ -1,10 +1,10 @@
 {
-  "best_global_step": 32500,
-  "best_metric": -30.467578887939453,
+  "best_global_step": 33250,
+  "best_metric": -30.468294143676758,
   "best_model_checkpoint": "./ar-diffusion-checkpoints-progressive-attention/checkpoint-32500",
-  "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": 59.237,
       "eval_steps_per_second": 14.809,
       "step": 33000
+    },
+    {
+      "epoch": 2.5421121452195985,
+      "grad_norm": 1.2125709056854248,
+      "learning_rate": 3.10053762044516e-05,
+      "loss": -30.4629,
+      "step": 33050
+    },
+    {
+      "epoch": 2.5459580032305205,
+      "grad_norm": 1.4404875040054321,
+      "learning_rate": 3.0745656182635115e-05,
+      "loss": -30.464,
+      "step": 33100
+    },
+    {
+      "epoch": 2.549803861241443,
+      "grad_norm": 1.889853835105896,
+      "learning_rate": 3.048593616081864e-05,
+      "loss": -30.4633,
+      "step": 33150
+    },
+    {
+      "epoch": 2.5536497192523653,
+      "grad_norm": 1.402807593345642,
+      "learning_rate": 3.0226216139002155e-05,
+      "loss": -30.4639,
+      "step": 33200
+    },
+    {
+      "epoch": 2.5574955772632872,
+      "grad_norm": 1.044944405555725,
+      "learning_rate": 2.9966496117185678e-05,
+      "loss": -30.4639,
+      "step": 33250
+    },
+    {
+      "epoch": 2.5574955772632872,
+      "eval_loss": -30.468294143676758,
+      "eval_runtime": 16.9528,
+      "eval_samples_per_second": 58.987,
+      "eval_steps_per_second": 14.747,
+      "step": 33250
+    },
+    {
+      "epoch": 2.5613414352742097,
+      "grad_norm": 1.4782401323318481,
+      "learning_rate": 2.9706776095369194e-05,
+      "loss": -30.4639,
+      "step": 33300
+    },
+    {
+      "epoch": 2.565187293285132,
+      "grad_norm": 1.0335558652877808,
+      "learning_rate": 2.944705607355271e-05,
+      "loss": -30.4635,
+      "step": 33350
+    },
+    {
+      "epoch": 2.569033151296054,
+      "grad_norm": 1.1205657720565796,
+      "learning_rate": 2.918733605173623e-05,
+      "loss": -30.4643,
+      "step": 33400
+    },
+    {
+      "epoch": 2.5728790093069764,
+      "grad_norm": 1.6154303550720215,
+      "learning_rate": 2.8927616029919746e-05,
+      "loss": -30.4641,
+      "step": 33450
+    },
+    {
+      "epoch": 2.576724867317899,
+      "grad_norm": 1.305467128753662,
+      "learning_rate": 2.866789600810327e-05,
+      "loss": -30.4642,
+      "step": 33500
+    },
+    {
+      "epoch": 2.576724867317899,
+      "eval_loss": -30.468229293823242,
+      "eval_runtime": 16.8335,
+      "eval_samples_per_second": 59.405,
+      "eval_steps_per_second": 14.851,
+      "step": 33500
     }
   ],
   "logging_steps": 50,