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:cfa682c656512f0db38279360e472eed38f6e68c384179fb8ebc379ae0bbd931
+oid sha256:7b1420e5f67ba76bfd4febfb8166681a03c08ca555de4aa6aed11752038beb7c
 size 36730224

last-checkpoint/ar_diffusion_info.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2baaf7d1ff2607c2b51f6cfbe05d40604cadc3c0c83d92adf1fa1aa8878e8345
+oid sha256:631f5d8920185c1643d4b91793959905c6ffb5cfdb99578fa1feeb9561f6cc11
 size 1544

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9b3473e19f33d061d72e874bb7ead12bc6195f3a6d1ef6cb2617c3cd645bff56
+oid sha256:b56e183e85858df2561ffd8210747bca3130019aa445c6931bdecb14ee1c1463
 size 73588346

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:32c4abfc978e57dde0e9122ce6e0a211a14cdb3e9d7e1f8e79d82dbf8734a5a1
+oid sha256:6ba374ceeab3c18ecca1a51999f0124017113854b9f36714acb5b3cbd2a2d7ce
 size 14244

last-checkpoint/scaler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:56883ca08ba7d3883d8223a68a16de2347a2e5eee3a5f6cfa3b0ef70d1be127c
+oid sha256:56099c2856f5340cf7eae161f2a31dbe463e29390c14dc50fb171114345187df
 size 988

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:336a48abe3f0221158899799d17e11e8d7fd80ac42ba2794eb638e218cd1c8a0
+oid sha256:97c85fac85989286832186f4fae10d665a033afecb7b12aeb8bf5a4a199aa56c
 size 1064

last-checkpoint/trainer_state.json

@@ -2,9 +2,9 @@
   "best_global_step": 3500,
   "best_metric": 4.520543098449707,
   "best_model_checkpoint": "./ar-diffusion-checkpoints-fixed/checkpoint-3500",
-  "epoch": 0.26921006076455656,
+  "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": 53.444,
       "eval_steps_per_second": 13.361,
       "step": 3500
+    },
+    {
+      "epoch": 0.2730559187754788,
+      "grad_norm": 3.639690399169922,
+      "learning_rate": 0.0001841934394722489,
+      "loss": 4.5127,
+      "step": 3550
+    },
+    {
+      "epoch": 0.27690177678640104,
+      "grad_norm": 3.864473342895508,
+      "learning_rate": 0.00018393371945043245,
+      "loss": 4.599,
+      "step": 3600
+    },
+    {
+      "epoch": 0.2807476347973233,
+      "grad_norm": 4.662705421447754,
+      "learning_rate": 0.00018367399942861596,
+      "loss": 4.5951,
+      "step": 3650
+    },
+    {
+      "epoch": 0.28459349280824553,
+      "grad_norm": 3.066333532333374,
+      "learning_rate": 0.00018341427940679947,
+      "loss": 4.5583,
+      "step": 3700
+    },
+    {
+      "epoch": 0.2884393508191678,
+      "grad_norm": 4.054274082183838,
+      "learning_rate": 0.000183154559384983,
+      "loss": 4.5732,
+      "step": 3750
+    },
+    {
+      "epoch": 0.2884393508191678,
+      "eval_loss": 4.569947242736816,
+      "eval_runtime": 18.5242,
+      "eval_samples_per_second": 53.983,
+      "eval_steps_per_second": 13.496,
+      "step": 3750
+    },
+    {
+      "epoch": 0.29228520883009,
+      "grad_norm": 6.440130710601807,
+      "learning_rate": 0.00018289483936316652,
+      "loss": 4.6025,
+      "step": 3800
+    },
+    {
+      "epoch": 0.2961310668410122,
+      "grad_norm": 5.849060535430908,
+      "learning_rate": 0.00018263511934135004,
+      "loss": 4.4932,
+      "step": 3850
+    },
+    {
+      "epoch": 0.29997692485193445,
+      "grad_norm": 7.0537800788879395,
+      "learning_rate": 0.00018237539931953355,
+      "loss": 4.6986,
+      "step": 3900
+    },
+    {
+      "epoch": 0.3038227828628567,
+      "grad_norm": 5.134806156158447,
+      "learning_rate": 0.00018211567929771706,
+      "loss": 4.5074,
+      "step": 3950
+    },
+    {
+      "epoch": 0.30766864087377893,
+      "grad_norm": 2.872307777404785,
+      "learning_rate": 0.00018185595927590058,
+      "loss": 4.547,
+      "step": 4000
+    },
+    {
+      "epoch": 0.30766864087377893,
+      "eval_loss": 4.5459160804748535,
+      "eval_runtime": 18.4232,
+      "eval_samples_per_second": 54.279,
+      "eval_steps_per_second": 13.57,
+      "step": 4000
     }
   ],
   "logging_steps": 50,