Training in progress, step 24000, checkpoint

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:47666e73d822697d81bbfccb0535c0d56782b13d9471c108cfb5f708ab05a025
+oid sha256:06b80a3e5e81ea8262619b59b411d50915cb7d91c532f396674e608aa19a7a5b
 size 36730224

last-checkpoint/ar_diffusion_info.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8b33800467cd4e0ea352bddd3d6bc949226fd682d012e035e3cd5842672b679a
+oid sha256:a812b933b5aac13b81e3ac023d4a0348fc701c5936c2577872dc75732f6799cc
 size 1544

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f1890e293af670b81b954d87d565c793549dd2a93009af624045ee0011125001
+oid sha256:641d51a5462cae27befb5debf367c14d6a2059451b08b406d1b63c2c76383d3a
 size 73588346

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9cb49390cd54c8aca2233c64b9e33d7396d51837a032de9f6d8f9419982ba2ef
+oid sha256:5a529729fbc0762ea5dabee8322aee007fe27ff660b1781623be0f3c2d691e1e
 size 14244

last-checkpoint/scaler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b4b01d3ddac56b25619c1ead241e352b13330923fdc417898657168ee04f3af4
+oid sha256:dc0f357a937dd3e56f37c213e3e98cc6fe77ca9c6c9815a27d739cddf90093de
 size 988

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e8a75b227d1139c1950381f2862e7e3504b8ee20dbb6bf732f2c5e660ca929c9
+oid sha256:7aecc1d321ec8fbf0f368c1b01443b517c9ca803ca3b79c3588dd96e5e6ade76
 size 1064

last-checkpoint/trainer_state.json

@@ -1,10 +1,10 @@
 {
-  "best_global_step": 22250,
-  "best_metric": 1.4596961736679077,
+  "best_global_step": 23750,
+  "best_metric": 1.4570631980895996,
   "best_model_checkpoint": "./ar-diffusion-checkpoints-fixed/checkpoint-21000",
-  "epoch": 1.8075532651334512,
+  "epoch": 1.8460118452426735,
   "eval_steps": 250,
-  "global_step": 23500,
+  "global_step": 24000,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -4050,6 +4050,92 @@
       "eval_samples_per_second": 55.623,
       "eval_steps_per_second": 13.906,
       "step": 23500
+    },
+    {
+      "epoch": 1.8113991231443736,
+      "grad_norm": 1.8125578165054321,
+      "learning_rate": 8.033140274783783e-05,
+      "loss": 1.4318,
+      "step": 23550
+    },
+    {
+      "epoch": 1.8152449811552958,
+      "grad_norm": 1.941846489906311,
+      "learning_rate": 8.007168272602136e-05,
+      "loss": 1.4569,
+      "step": 23600
+    },
+    {
+      "epoch": 1.819090839166218,
+      "grad_norm": 2.054161787033081,
+      "learning_rate": 7.981196270420487e-05,
+      "loss": 1.4426,
+      "step": 23650
+    },
+    {
+      "epoch": 1.8229366971771404,
+      "grad_norm": 1.6102700233459473,
+      "learning_rate": 7.955224268238839e-05,
+      "loss": 1.4407,
+      "step": 23700
+    },
+    {
+      "epoch": 1.8267825551880623,
+      "grad_norm": 1.7302616834640503,
+      "learning_rate": 7.929252266057191e-05,
+      "loss": 1.4432,
+      "step": 23750
+    },
+    {
+      "epoch": 1.8267825551880623,
+      "eval_loss": 1.4570631980895996,
+      "eval_runtime": 17.8893,
+      "eval_samples_per_second": 55.899,
+      "eval_steps_per_second": 13.975,
+      "step": 23750
+    },
+    {
+      "epoch": 1.8306284131989847,
+      "grad_norm": 9.521604537963867,
+      "learning_rate": 7.903280263875543e-05,
+      "loss": 1.3995,
+      "step": 23800
+    },
+    {
+      "epoch": 1.834474271209907,
+      "grad_norm": 1.9546847343444824,
+      "learning_rate": 7.877308261693894e-05,
+      "loss": 1.4957,
+      "step": 23850
+    },
+    {
+      "epoch": 1.838320129220829,
+      "grad_norm": 1.173722505569458,
+      "learning_rate": 7.851336259512246e-05,
+      "loss": 1.4157,
+      "step": 23900
+    },
+    {
+      "epoch": 1.8421659872317515,
+      "grad_norm": 1.5322128534317017,
+      "learning_rate": 7.825364257330599e-05,
+      "loss": 1.4453,
+      "step": 23950
+    },
+    {
+      "epoch": 1.8460118452426735,
+      "grad_norm": 1.0376055240631104,
+      "learning_rate": 7.799392255148949e-05,
+      "loss": 1.4646,
+      "step": 24000
+    },
+    {
+      "epoch": 1.8460118452426735,
+      "eval_loss": 1.4612687826156616,
+      "eval_runtime": 17.984,
+      "eval_samples_per_second": 55.605,
+      "eval_steps_per_second": 13.901,
+      "step": 24000
     }
   ],
   "logging_steps": 50,