Training in progress, step 32000, checkpoint

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:26fe5fbd72f8309a5966565e1f05d03455fe898aaf3f6a593408a231e92fa1ae
+oid sha256:5592650a84fed0e53ff657190c7416551adf3f79e86876a254dd6c273b0bccdc
 size 36730224

last-checkpoint/ar_diffusion_info.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d7702730e2360690456ce5fb2de003ceedb407e6abc67a9dd87efa0ca97a4f6f
+oid sha256:32d27943f70edaf81f6f4c1789ed21c721bcc0f5827fef85bdd08cb9092a87bd
 size 1736

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6e8c967a1bb650a282dfca97a5bf89cc59b84a455fbf7c176eaf708aaf67147b
+oid sha256:c3d55653bad9ee7c1db558a9b36a74c2617a2024075b12282a849a401a70d6bf
 size 73588346

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b81eede41db1b2112b377e40f5bb217fe7088a7db51c27fb1ff5966c6047f590
+oid sha256:c1f687f79c35c0845312b8176cbb48a154ba5067da083afab2edee21bf2e08ac
 size 14244

last-checkpoint/scaler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1c3b2474e60ad886726c989e3871ac343e5eba2ea2c8e3456763929416e60cc5
+oid sha256:66e856574eb8a1f2a02d63513aff48279087ce622f3e8211116a357d66952614
 size 988

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:26b6028446ce95055f0aef8145b820eaeff69cebdc860cf1504a88dbd4f2d412
+oid sha256:a9ed56cf408beb5289ea76a3c22da837df362efd2635153f560e4bbed768681d
 size 1064

last-checkpoint/trainer_state.json

@@ -2,9 +2,9 @@
   "best_global_step": 31000,
   "best_metric": 0.7226839661598206,
   "best_model_checkpoint": "./ar-diffusion-checkpoints-fixed/checkpoint-31000",
-  "epoch": 2.422890546881009,
+  "epoch": 2.4613491269902315,
   "eval_steps": 250,
-  "global_step": 31500,
+  "global_step": 32000,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -5426,6 +5426,92 @@
       "eval_samples_per_second": 55.616,
       "eval_steps_per_second": 13.904,
       "step": 31500
+    },
+    {
+      "epoch": 2.4267364048919315,
+      "grad_norm": 0.9221381545066833,
+      "learning_rate": 3.877619925720074e-05,
+      "loss": 0.7077,
+      "step": 31550
+    },
+    {
+      "epoch": 2.4305822629028535,
+      "grad_norm": 1.1155864000320435,
+      "learning_rate": 3.8516479235384255e-05,
+      "loss": 0.7358,
+      "step": 31600
+    },
+    {
+      "epoch": 2.434428120913776,
+      "grad_norm": 0.949946939945221,
+      "learning_rate": 3.825675921356778e-05,
+      "loss": 0.7314,
+      "step": 31650
+    },
+    {
+      "epoch": 2.4382739789246983,
+      "grad_norm": 0.7200281023979187,
+      "learning_rate": 3.7997039191751294e-05,
+      "loss": 0.6781,
+      "step": 31700
+    },
+    {
+      "epoch": 2.4421198369356203,
+      "grad_norm": 1.1924189329147339,
+      "learning_rate": 3.773731916993481e-05,
+      "loss": 0.7116,
+      "step": 31750
+    },
+    {
+      "epoch": 2.4421198369356203,
+      "eval_loss": 0.7858553528785706,
+      "eval_runtime": 18.0342,
+      "eval_samples_per_second": 55.45,
+      "eval_steps_per_second": 13.863,
+      "step": 31750
+    },
+    {
+      "epoch": 2.4459656949465427,
+      "grad_norm": 0.7993971109390259,
+      "learning_rate": 3.747759914811833e-05,
+      "loss": 0.7068,
+      "step": 31800
+    },
+    {
+      "epoch": 2.4498115529574647,
+      "grad_norm": 0.6277671456336975,
+      "learning_rate": 3.7217879126301846e-05,
+      "loss": 0.7273,
+      "step": 31850
+    },
+    {
+      "epoch": 2.453657410968387,
+      "grad_norm": 0.8524878621101379,
+      "learning_rate": 3.6958159104485365e-05,
+      "loss": 0.7537,
+      "step": 31900
+    },
+    {
+      "epoch": 2.457503268979309,
+      "grad_norm": 0.9068925976753235,
+      "learning_rate": 3.6698439082668885e-05,
+      "loss": 0.7192,
+      "step": 31950
+    },
+    {
+      "epoch": 2.4613491269902315,
+      "grad_norm": 0.866385817527771,
+      "learning_rate": 3.6438719060852405e-05,
+      "loss": 0.7145,
+      "step": 32000
+    },
+    {
+      "epoch": 2.4613491269902315,
+      "eval_loss": 0.7852405905723572,
+      "eval_runtime": 18.0021,
+      "eval_samples_per_second": 55.549,
+      "eval_steps_per_second": 13.887,
+      "step": 32000
     }
   ],
   "logging_steps": 50,