Training in progress, step 28000, checkpoint

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:50babce35e19ba1c52c051d5771bd2ef5b1fb8837c7eb686a5f692232dc3476b
+oid sha256:327ebc57ff4a2fce5b1ff51eacb255f502d310697ef3cac70837eee01705b907
 size 36730224

last-checkpoint/ar_diffusion_info.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a09ffdd971dbd61d394723ea366d9cae6c80cffca01f518b073b9aee24418ca7
+oid sha256:82ce8b4d70a4e5ffdc955b54b4e61ae42e280ad3043bd6c657a5a4f94d25406f
 size 1736

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5beb8c7d993dfee78065768c05d693af80563384c89dc9e510ad6471657c1e2b
+oid sha256:cefd1fadeee425e44de130ce43706a90f27357be2970c4c0e99ec3480bec2b93
 size 73588346

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:fbdead63124c1fc84f329ca037df0e53909d6ff2384119198ddaf577d7e61369
-size 14308
+oid sha256:675e8eec5b71db621439753580d23e328c8c5e2da6333850e0095f28c2ae9e4d
+size 14244

last-checkpoint/scaler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6914f08d038d2fd5583fe9d1dfb74316a740319f66641c1d86781bc3fc4b8381
+oid sha256:df1b3ae13e2a70e8f07dd9d7c74ac5ab9a7fad8c5d805908d310a8c82d08cad1
 size 988

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a5dbc7bedb58edd6ffc364b01517a26604f569a354c2be3f565e10596a8fd705
+oid sha256:14c5f190b4eb0a7cb5ff6d1c912de09299ca659cbcc34c660c71e8207c2fc146
 size 1064

last-checkpoint/trainer_state.json

@@ -1,10 +1,10 @@
 {
-  "best_global_step": 27500,
-  "best_metric": -30.462617874145508,
+  "best_global_step": 27750,
+  "best_metric": -30.463788986206055,
   "best_model_checkpoint": "./ar-diffusion-checkpoints-progressive-attention/checkpoint-27500",
-  "epoch": 2.11522190600723,
+  "epoch": 2.1536804861164525,
   "eval_steps": 250,
-  "global_step": 27500,
+  "global_step": 28000,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -4738,6 +4738,92 @@
       "eval_samples_per_second": 59.24,
       "eval_steps_per_second": 14.81,
       "step": 27500
+    },
+    {
+      "epoch": 2.1190677640181526,
+      "grad_norm": 1.9954910278320312,
+      "learning_rate": 5.957457860426461e-05,
+      "loss": -30.458,
+      "step": 27550
+    },
+    {
+      "epoch": 2.1229136220290745,
+      "grad_norm": 1.2620997428894043,
+      "learning_rate": 5.931485858244812e-05,
+      "loss": -30.4578,
+      "step": 27600
+    },
+    {
+      "epoch": 2.126759480039997,
+      "grad_norm": 1.8231252431869507,
+      "learning_rate": 5.905513856063164e-05,
+      "loss": -30.4571,
+      "step": 27650
+    },
+    {
+      "epoch": 2.1306053380509193,
+      "grad_norm": 2.8753602504730225,
+      "learning_rate": 5.879541853881516e-05,
+      "loss": -30.4576,
+      "step": 27700
+    },
+    {
+      "epoch": 2.1344511960618413,
+      "grad_norm": 2.1304397583007812,
+      "learning_rate": 5.8535698516998686e-05,
+      "loss": -30.4579,
+      "step": 27750
+    },
+    {
+      "epoch": 2.1344511960618413,
+      "eval_loss": -30.463788986206055,
+      "eval_runtime": 16.8747,
+      "eval_samples_per_second": 59.26,
+      "eval_steps_per_second": 14.815,
+      "step": 27750
+    },
+    {
+      "epoch": 2.1382970540727637,
+      "grad_norm": 2.8373219966888428,
+      "learning_rate": 5.827597849518219e-05,
+      "loss": -30.458,
+      "step": 27800
+    },
+    {
+      "epoch": 2.1421429120836857,
+      "grad_norm": 2.214996576309204,
+      "learning_rate": 5.801625847336571e-05,
+      "loss": -30.4577,
+      "step": 27850
+    },
+    {
+      "epoch": 2.145988770094608,
+      "grad_norm": 2.79176664352417,
+      "learning_rate": 5.775653845154924e-05,
+      "loss": -30.4586,
+      "step": 27900
+    },
+    {
+      "epoch": 2.1498346281055305,
+      "grad_norm": 3.1659934520721436,
+      "learning_rate": 5.7496818429732744e-05,
+      "loss": -30.4584,
+      "step": 27950
+    },
+    {
+      "epoch": 2.1536804861164525,
+      "grad_norm": 3.1562278270721436,
+      "learning_rate": 5.723709840791627e-05,
+      "loss": -30.4578,
+      "step": 28000
+    },
+    {
+      "epoch": 2.1536804861164525,
+      "eval_loss": -30.461944580078125,
+      "eval_runtime": 16.8316,
+      "eval_samples_per_second": 59.412,
+      "eval_steps_per_second": 14.853,
+      "step": 28000
     }
   ],
   "logging_steps": 50,