Training in progress, step 4500, checkpoint

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b445c42b37f9090e63ec9e650d67af120bdec1120aa2b99d075b10217d2f4041
+oid sha256:a2da0677fa12d4198b9a2aefbc499f61536ce08274e2b77ec9742768adbc966b
 size 36730224

last-checkpoint/ar_diffusion_info.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4ae940e2a8fce62cfe8e7b2911818f4f043d9298216ee2524750b9c2bb4195bb
+oid sha256:6d5cb22e083fb9b6ead9a6b9754983dfef6d4fd0258d21fc899979d4540b8ab1
 size 1736

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:dacaa8970d573f7be27d82942d32e193ea6f961af3a7a55fdd87f418730a6ce8
+oid sha256:6c840a4d894a9e893236ba3eb39bb9d761fb13f08794f23f4bb8cb0fa462367c
 size 73588346

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1698c9cb87c2e75a65bbfdd5b5bac56ca042377422d15083ff3eeeb34095ca85
+oid sha256:770b1304c731c8aa940b69b50f7998667dba8e089ec052ddb586fa531ecef79d
 size 14244

last-checkpoint/scaler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1b33c54ce26acce0407557fdf2626c004064a12b74672d7c82d6602cfe737b3d
+oid sha256:bcd3dca35f0088299a2a2e0ac814a4cd8b9d9681c16fef5ada156579adebbcb1
 size 988

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:cd43e4afdd546fe9e425b45803ea4071bedf22fd593f1e2dc9ea450f413acbe3
+oid sha256:5d1c26a223e581bb15f1e95b6c820cdf95e07905cfe02588473944a7fc241ee1
 size 1064

last-checkpoint/trainer_state.json

@@ -1,10 +1,10 @@
 {
-  "best_global_step": 4000,
-  "best_metric": 1.5196877717971802,
-  "best_model_checkpoint": "./ar-diffusion-checkpoints-progressive-attention/checkpoint-4000",
-  "epoch": 0.30766864087377893,
+  "best_global_step": 4500,
+  "best_metric": 1.4980955123901367,
+  "best_model_checkpoint": "./ar-diffusion-checkpoints-progressive-attention/checkpoint-4500",
+  "epoch": 0.3461272209830013,
   "eval_steps": 250,
-  "global_step": 4000,
+  "global_step": 4500,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -696,6 +696,92 @@
       "eval_samples_per_second": 59.695,
       "eval_steps_per_second": 14.924,
       "step": 4000
+    },
+    {
+      "epoch": 0.3115144988847012,
+      "grad_norm": 0.9612170457839966,
+      "learning_rate": 0.00018158585045321144,
+      "loss": 1.5134,
+      "step": 4050
+    },
+    {
+      "epoch": 0.3153603568956234,
+      "grad_norm": 1.4119675159454346,
+      "learning_rate": 0.00018132613043139498,
+      "loss": 1.5019,
+      "step": 4100
+    },
+    {
+      "epoch": 0.31920621490654566,
+      "grad_norm": 1.5474581718444824,
+      "learning_rate": 0.0001810664104095785,
+      "loss": 1.5243,
+      "step": 4150
+    },
+    {
+      "epoch": 0.3230520729174679,
+      "grad_norm": 1.5670969486236572,
+      "learning_rate": 0.000180806690387762,
+      "loss": 1.5482,
+      "step": 4200
+    },
+    {
+      "epoch": 0.32689793092839015,
+      "grad_norm": 1.5624980926513672,
+      "learning_rate": 0.00018054697036594552,
+      "loss": 1.5429,
+      "step": 4250
+    },
+    {
+      "epoch": 0.32689793092839015,
+      "eval_loss": 1.5110443830490112,
+      "eval_runtime": 17.0856,
+      "eval_samples_per_second": 58.529,
+      "eval_steps_per_second": 14.632,
+      "step": 4250
+    },
+    {
+      "epoch": 0.33074378893931233,
+      "grad_norm": 1.0080711841583252,
+      "learning_rate": 0.00018028725034412903,
+      "loss": 1.5022,
+      "step": 4300
+    },
+    {
+      "epoch": 0.3345896469502346,
+      "grad_norm": 0.9390202760696411,
+      "learning_rate": 0.00018002753032231257,
+      "loss": 1.482,
+      "step": 4350
+    },
+    {
+      "epoch": 0.3384355049611568,
+      "grad_norm": 1.0671314001083374,
+      "learning_rate": 0.00017976781030049608,
+      "loss": 1.5271,
+      "step": 4400
+    },
+    {
+      "epoch": 0.34228136297207906,
+      "grad_norm": 1.290907382965088,
+      "learning_rate": 0.0001795080902786796,
+      "loss": 1.4667,
+      "step": 4450
+    },
+    {
+      "epoch": 0.3461272209830013,
+      "grad_norm": 0.6162307858467102,
+      "learning_rate": 0.0001792483702568631,
+      "loss": 1.4435,
+      "step": 4500
+    },
+    {
+      "epoch": 0.3461272209830013,
+      "eval_loss": 1.4980955123901367,
+      "eval_runtime": 16.8193,
+      "eval_samples_per_second": 59.456,
+      "eval_steps_per_second": 14.864,
+      "step": 4500
     }
   ],
   "logging_steps": 50,