Training in progress, step 4000, checkpoint

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:11a062f0d5247b9505627696a295b4feca04c6c45dd688fcbc2c07d7828e414c
+oid sha256:b445c42b37f9090e63ec9e650d67af120bdec1120aa2b99d075b10217d2f4041
 size 36730224

last-checkpoint/ar_diffusion_info.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1ce273710431703c604c9d640c23434ce3f2b038664d7b3df1e4a13e933764d0
+oid sha256:4ae940e2a8fce62cfe8e7b2911818f4f043d9298216ee2524750b9c2bb4195bb
 size 1736

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:eb347b9f0c8f6a0d9a0e6b08949ae99d2459f1f662e3a7dd16f39b8ba0ddf69c
+oid sha256:dacaa8970d573f7be27d82942d32e193ea6f961af3a7a55fdd87f418730a6ce8
 size 73588346

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3cf21f101c997b0643b73ca8f669be23f0abb9e5a2c6fedb07e17298d9cdd268
+oid sha256:1698c9cb87c2e75a65bbfdd5b5bac56ca042377422d15083ff3eeeb34095ca85
 size 14244

last-checkpoint/scaler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:028b7d42ce0dbafb2a0126ad830dd1957166a4ca85043c6644487cae8315bfe9
+oid sha256:1b33c54ce26acce0407557fdf2626c004064a12b74672d7c82d6602cfe737b3d
 size 988

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d16b13e1d35412f9085034fa2481674486c7b966c61cae81848c36587b827f3d
+oid sha256:cd43e4afdd546fe9e425b45803ea4071bedf22fd593f1e2dc9ea450f413acbe3
 size 1064

last-checkpoint/trainer_state.json

@@ -1,10 +1,10 @@
 {
-  "best_global_step": 3500,
-  "best_metric": 1.5354665517807007,
-  "best_model_checkpoint": "./ar-diffusion-checkpoints-progressive-attention/checkpoint-3500",
-  "epoch": 0.26921006076455656,
+  "best_global_step": 4000,
+  "best_metric": 1.5196877717971802,
+  "best_model_checkpoint": "./ar-diffusion-checkpoints-progressive-attention/checkpoint-4000",
+  "epoch": 0.30766864087377893,
   "eval_steps": 250,
-  "global_step": 3500,
+  "global_step": 4000,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -610,6 +610,92 @@
       "eval_samples_per_second": 59.184,
       "eval_steps_per_second": 14.796,
       "step": 3500
+    },
+    {
+      "epoch": 0.2730559187754788,
+      "grad_norm": 1.5297306776046753,
+      "learning_rate": 0.00018418305067137626,
+      "loss": 1.5385,
+      "step": 3550
+    },
+    {
+      "epoch": 0.27690177678640104,
+      "grad_norm": 1.2498586177825928,
+      "learning_rate": 0.00018392333064955977,
+      "loss": 1.5591,
+      "step": 3600
+    },
+    {
+      "epoch": 0.2807476347973233,
+      "grad_norm": 1.5092873573303223,
+      "learning_rate": 0.0001836636106277433,
+      "loss": 1.554,
+      "step": 3650
+    },
+    {
+      "epoch": 0.28459349280824553,
+      "grad_norm": 1.4669116735458374,
+      "learning_rate": 0.00018340389060592682,
+      "loss": 1.5062,
+      "step": 3700
+    },
+    {
+      "epoch": 0.2884393508191678,
+      "grad_norm": 1.5152875185012817,
+      "learning_rate": 0.00018314417058411033,
+      "loss": 1.5277,
+      "step": 3750
+    },
+    {
+      "epoch": 0.2884393508191678,
+      "eval_loss": 1.5374407768249512,
+      "eval_runtime": 16.9404,
+      "eval_samples_per_second": 59.03,
+      "eval_steps_per_second": 14.758,
+      "step": 3750
+    },
+    {
+      "epoch": 0.29228520883009,
+      "grad_norm": 2.025120735168457,
+      "learning_rate": 0.00018288445056229387,
+      "loss": 1.534,
+      "step": 3800
+    },
+    {
+      "epoch": 0.2961310668410122,
+      "grad_norm": 1.34319269657135,
+      "learning_rate": 0.00018262473054047736,
+      "loss": 1.4707,
+      "step": 3850
+    },
+    {
+      "epoch": 0.29997692485193445,
+      "grad_norm": 1.9989622831344604,
+      "learning_rate": 0.00018236501051866087,
+      "loss": 1.567,
+      "step": 3900
+    },
+    {
+      "epoch": 0.3038227828628567,
+      "grad_norm": 1.2157036066055298,
+      "learning_rate": 0.0001821052904968444,
+      "loss": 1.507,
+      "step": 3950
+    },
+    {
+      "epoch": 0.30766864087377893,
+      "grad_norm": 0.9292582869529724,
+      "learning_rate": 0.00018184557047502793,
+      "loss": 1.4934,
+      "step": 4000
+    },
+    {
+      "epoch": 0.30766864087377893,
+      "eval_loss": 1.5196877717971802,
+      "eval_runtime": 16.7517,
+      "eval_samples_per_second": 59.695,
+      "eval_steps_per_second": 14.924,
+      "step": 4000
     }
   ],
   "logging_steps": 50,