Training in progress, step 5500, checkpoint

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ccc114460bdc62c626ba50c920e272e7f575d8aeb0e2543e39fc6ee05fa17062
+oid sha256:4eb8910307bec61db122da2594e6b3f530d868dc95f89d2ca698471ef1bf7e40
 size 36730224

last-checkpoint/ar_diffusion_info.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:50d17eabb10066129b6714ff713807dcb69661d9ef6abc8937da238409918435
+oid sha256:7f7c3bd2e2b53a0436de47612a1d9d71e57d2e478df8b52acc7aec65dc000186
 size 1736

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f41eb4686c9fcfd89bdeef4e240aa2cba7d92d7862f3e09543657417b8a31b1f
+oid sha256:42be51fbf276340c9331b95a10d361ad5f54ac5fddd8d3003ab5a2b3205b1bf3
 size 73588346

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b3f90304776c4344263775ad85fa55a101daf353f68d13a5872b9fc01a14e2b3
-size 14180
+oid sha256:edba3e2e11b49613ce2a377ec116d91655a77025fcd35863ca859fcc236283bd
+size 14244

last-checkpoint/scaler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:429536711957c0df9399d6c91b39d43aa4faf1c40a0c6bae8287d373fb9daa66
+oid sha256:ee231d5124e1faa72bb59e0d3351580448c9875eebe2a58142e26f5eeab5354c
 size 988

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:17a4cf673b793da03b6c2fcc104b7095156a3747182923bcd50f7b981d959c40
+oid sha256:c1402f998a531b2d9e639df3a0d0bab3a51ee695db29df819edf5370824d49f4
 size 1064

last-checkpoint/trainer_state.json

@@ -1,10 +1,10 @@
 {
-  "best_global_step": 5000,
-  "best_metric": 1.4847265481948853,
+  "best_global_step": 5250,
+  "best_metric": 1.4721767902374268,
   "best_model_checkpoint": "./ar-diffusion-checkpoints-progressive-attention/checkpoint-5000",
-  "epoch": 0.3845858010922237,
+  "epoch": 0.42304438120144605,
   "eval_steps": 250,
-  "global_step": 5000,
+  "global_step": 5500,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -868,6 +868,92 @@
       "eval_samples_per_second": 59.163,
       "eval_steps_per_second": 14.791,
       "step": 5000
+    },
+    {
+      "epoch": 0.3884316591031459,
+      "grad_norm": 1.0815047025680542,
+      "learning_rate": 0.00017639145001688183,
+      "loss": 1.5059,
+      "step": 5050
+    },
+    {
+      "epoch": 0.39227751711406816,
+      "grad_norm": 0.8886498808860779,
+      "learning_rate": 0.00017613172999506534,
+      "loss": 1.5056,
+      "step": 5100
+    },
+    {
+      "epoch": 0.3961233751249904,
+      "grad_norm": 1.647912859916687,
+      "learning_rate": 0.00017587200997324883,
+      "loss": 1.4471,
+      "step": 5150
+    },
+    {
+      "epoch": 0.39996923313591265,
+      "grad_norm": 1.0519336462020874,
+      "learning_rate": 0.00017561228995143237,
+      "loss": 1.5272,
+      "step": 5200
+    },
+    {
+      "epoch": 0.40381509114683484,
+      "grad_norm": 0.7143021821975708,
+      "learning_rate": 0.00017535256992961588,
+      "loss": 1.5315,
+      "step": 5250
+    },
+    {
+      "epoch": 0.40381509114683484,
+      "eval_loss": 1.4721767902374268,
+      "eval_runtime": 16.9221,
+      "eval_samples_per_second": 59.094,
+      "eval_steps_per_second": 14.774,
+      "step": 5250
+    },
+    {
+      "epoch": 0.4076609491577571,
+      "grad_norm": 0.9163336157798767,
+      "learning_rate": 0.0001750928499077994,
+      "loss": 1.4824,
+      "step": 5300
+    },
+    {
+      "epoch": 0.4115068071686793,
+      "grad_norm": 1.6825785636901855,
+      "learning_rate": 0.00017483312988598293,
+      "loss": 1.5034,
+      "step": 5350
+    },
+    {
+      "epoch": 0.41535266517960157,
+      "grad_norm": 1.1994304656982422,
+      "learning_rate": 0.00017457340986416645,
+      "loss": 1.4712,
+      "step": 5400
+    },
+    {
+      "epoch": 0.4191985231905238,
+      "grad_norm": 2.0379066467285156,
+      "learning_rate": 0.00017431368984234996,
+      "loss": 1.409,
+      "step": 5450
+    },
+    {
+      "epoch": 0.42304438120144605,
+      "grad_norm": 1.5934576988220215,
+      "learning_rate": 0.00017405396982053347,
+      "loss": 1.4791,
+      "step": 5500
+    },
+    {
+      "epoch": 0.42304438120144605,
+      "eval_loss": 1.4770058393478394,
+      "eval_runtime": 16.8532,
+      "eval_samples_per_second": 59.336,
+      "eval_steps_per_second": 14.834,
+      "step": 5500
     }
   ],
   "logging_steps": 50,