Training in progress, step 19500, checkpoint

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:31847bb52936d686e74359d3478c14923c1ebe168be3dabe55eeb97e094e4e4d
+oid sha256:1fd7698db7af5ca1d96c523abf1ce3ed39dad6f1ef364885f5698168d5072eab
 size 36730224

last-checkpoint/ar_diffusion_info.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a28811b9b6f21f7c3eb8dd1560ffdbc347e72ba020959bb8f5bc088d847bdc83
+oid sha256:d356319b5cb8614b773ad954915afabc933d54cfb874f3f02d0e30cd4ce665d3
 size 1544

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:40965b88e5eb22bf8d3caa10525eb8f13ed0f07bce8d46631079916b2bb29747
+oid sha256:3a96cf13c25e993c5a75039856a7a5ef57f09af02fd14f8120fb00ff3c830655
 size 73588346

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3c57dd490464340c9b3aa212b9d3844b50cfd1f6b44323e804e87459d3ff333e
+oid sha256:aa7c19bbb017976e63ac5e9fec3eea0eda0cf5868e0275e6b00fde078e2c850d
 size 14244

last-checkpoint/scaler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:dbc432474a75821e9b2dae49e65dbd5211090e2d95ceeb48957adde037f27b78
+oid sha256:0f8f3437cd7c6266ef3d6b26d1f23a26258ce02afba475c3dabc69b494e62bdf
 size 988

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:68cdaa34d552da65c658445f52e286a15f93a5093fe3daa82953914a2ad847cc
+oid sha256:51597cfa5b705f0eeeef37eec7f4e933d1fd3032aa1f165769969345026a8d82
 size 1064

last-checkpoint/trainer_state.json

@@ -2,9 +2,9 @@
   "best_global_step": 18750,
   "best_metric": 1.4757392406463623,
   "best_model_checkpoint": "./ar-diffusion-checkpoints-fixed/checkpoint-15500",
-  "epoch": 1.4614260441504499,
+  "epoch": 1.4998846242596724,
   "eval_steps": 250,
-  "global_step": 19000,
+  "global_step": 19500,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -3276,6 +3276,92 @@
       "eval_samples_per_second": 56.258,
       "eval_steps_per_second": 14.065,
       "step": 19000
+    },
+    {
+      "epoch": 1.4652719021613723,
+      "grad_norm": 1.2875189781188965,
+      "learning_rate": 0.00010370101031088487,
+      "loss": 1.4826,
+      "step": 19050
+    },
+    {
+      "epoch": 1.4691177601722945,
+      "grad_norm": 1.7036223411560059,
+      "learning_rate": 0.00010344129028906839,
+      "loss": 1.4392,
+      "step": 19100
+    },
+    {
+      "epoch": 1.4729636181832166,
+      "grad_norm": 1.537514328956604,
+      "learning_rate": 0.0001031815702672519,
+      "loss": 1.4846,
+      "step": 19150
+    },
+    {
+      "epoch": 1.476809476194139,
+      "grad_norm": 0.9159242510795593,
+      "learning_rate": 0.00010292185024543543,
+      "loss": 1.4668,
+      "step": 19200
+    },
+    {
+      "epoch": 1.480655334205061,
+      "grad_norm": 3.47868013381958,
+      "learning_rate": 0.00010266213022361894,
+      "loss": 1.493,
+      "step": 19250
+    },
+    {
+      "epoch": 1.480655334205061,
+      "eval_loss": 1.4858986139297485,
+      "eval_runtime": 17.7738,
+      "eval_samples_per_second": 56.263,
+      "eval_steps_per_second": 14.066,
+      "step": 19250
+    },
+    {
+      "epoch": 1.4845011922159834,
+      "grad_norm": 1.467437505722046,
+      "learning_rate": 0.00010240241020180246,
+      "loss": 1.4639,
+      "step": 19300
+    },
+    {
+      "epoch": 1.4883470502269056,
+      "grad_norm": 1.2710049152374268,
+      "learning_rate": 0.00010214269017998598,
+      "loss": 1.4244,
+      "step": 19350
+    },
+    {
+      "epoch": 1.4921929082378278,
+      "grad_norm": 2.0059661865234375,
+      "learning_rate": 0.0001018829701581695,
+      "loss": 1.4258,
+      "step": 19400
+    },
+    {
+      "epoch": 1.4960387662487502,
+      "grad_norm": 1.7536308765411377,
+      "learning_rate": 0.00010162325013635301,
+      "loss": 1.396,
+      "step": 19450
+    },
+    {
+      "epoch": 1.4998846242596724,
+      "grad_norm": 0.9684279561042786,
+      "learning_rate": 0.00010136353011453655,
+      "loss": 1.4598,
+      "step": 19500
+    },
+    {
+      "epoch": 1.4998846242596724,
+      "eval_loss": 1.4841110706329346,
+      "eval_runtime": 17.9151,
+      "eval_samples_per_second": 55.819,
+      "eval_steps_per_second": 13.955,
+      "step": 19500
     }
   ],
   "logging_steps": 50,