Training in progress, step 4000, checkpoint

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:24082cd1025a9c16d6bc6449cbc8ef1f1d242da73316dafd4f2689fc8c860917
+oid sha256:75410142e7d88776abfc3caa1d65963cd8d85da0ff4f4ef2a8275d75bab13fc7
 size 36730224

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a4d61ee4cf28e305ccbd9c0cd63b04a590ac2c163f4c12096fd42f05674ed054
+oid sha256:543307a68c41ab7eab4206e5219ee92650f34cf85fdde0ae16f2041a7d666f3f
 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 sha256:835fd47c5d5c5f39bf487b0cc415ae7b687f6d1d1601a7df5275be58f8fd603c
+oid sha256:51fee8eb233312b3e7cebdb2d20739865391ab32cb8dc65d3edc7a1f54a4611d
 size 988

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3b0c041eadc91e24d174b435fcb1ec46a0c56f36b09b7dbbf49009145cd7fa33
+oid sha256:c31d829461e38fcd04f37e8cd6d881ff82a519c1416dbfe781c31b6217218abc
 size 1064

last-checkpoint/trainer_state.json

@@ -2,9 +2,9 @@
   "best_global_step": 3500,
   "best_metric": 4.488556861877441,
   "best_model_checkpoint": "./ar-diffusion-checkpoints/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": 52.851,
       "eval_steps_per_second": 13.213,
       "step": 3500
+    },
+    {
+      "epoch": 0.2730559187754788,
+      "grad_norm": 3.794307231903076,
+      "learning_rate": 0.00012227418109628322,
+      "loss": 4.4994,
+      "step": 3550
+    },
+    {
+      "epoch": 0.27690177678640104,
+      "grad_norm": 3.5770812034606934,
+      "learning_rate": 0.00012201683761626409,
+      "loss": 4.5888,
+      "step": 3600
+    },
+    {
+      "epoch": 0.2807476347973233,
+      "grad_norm": 4.770874500274658,
+      "learning_rate": 0.00012175949413624498,
+      "loss": 4.5644,
+      "step": 3650
+    },
+    {
+      "epoch": 0.28459349280824553,
+      "grad_norm": 3.4447147846221924,
+      "learning_rate": 0.00012150215065622586,
+      "loss": 4.5301,
+      "step": 3700
+    },
+    {
+      "epoch": 0.2884393508191678,
+      "grad_norm": 4.76978063583374,
+      "learning_rate": 0.00012124480717620675,
+      "loss": 4.5563,
+      "step": 3750
+    },
+    {
+      "epoch": 0.2884393508191678,
+      "eval_loss": 4.53049898147583,
+      "eval_runtime": 18.9074,
+      "eval_samples_per_second": 52.889,
+      "eval_steps_per_second": 13.222,
+      "step": 3750
+    },
+    {
+      "epoch": 0.29228520883009,
+      "grad_norm": 5.7456512451171875,
+      "learning_rate": 0.00012098746369618763,
+      "loss": 4.5612,
+      "step": 3800
+    },
+    {
+      "epoch": 0.2961310668410122,
+      "grad_norm": 5.577849864959717,
+      "learning_rate": 0.00012073012021616851,
+      "loss": 4.4629,
+      "step": 3850
+    },
+    {
+      "epoch": 0.29997692485193445,
+      "grad_norm": 4.432284832000732,
+      "learning_rate": 0.00012047277673614939,
+      "loss": 4.6661,
+      "step": 3900
+    },
+    {
+      "epoch": 0.3038227828628567,
+      "grad_norm": 5.174475193023682,
+      "learning_rate": 0.00012021543325613027,
+      "loss": 4.4835,
+      "step": 3950
+    },
+    {
+      "epoch": 0.30766864087377893,
+      "grad_norm": 3.5657413005828857,
+      "learning_rate": 0.00011995808977611117,
+      "loss": 4.4894,
+      "step": 4000
+    },
+    {
+      "epoch": 0.30766864087377893,
+      "eval_loss": 4.5068535804748535,
+      "eval_runtime": 18.7951,
+      "eval_samples_per_second": 53.205,
+      "eval_steps_per_second": 13.301,
+      "step": 4000
     }
   ],
   "logging_steps": 50,