Training in progress, step 29000, checkpoint

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:fe6e2fa1318dd3ad1205203e4fff88f82e693270053626bb1d002f9b23c9b24c
+oid sha256:afb1f994dba522602e482c6617e68b2ec24254fe04ad07399654bbd51f70ee28
 size 36730224

last-checkpoint/ar_diffusion_info.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:599432259819e93dad17855a204a171f8d5bf2703ad47dd88ce1d465a13b2f0b
+oid sha256:8076f6cd20d7f4c66417fcba7f62699920c2ac2bc88d4cc342541ab812336f0a
 size 1544

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:48da092ff21d2489fb67e59a49519237044b07edc1b34a941910a561850d4e26
+oid sha256:3531171f249076e214e0e84ae3500c8a6ba16b1c8f7d014a2f1b1fe43cc09834
 size 73588346

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1e10c23934142af0d6a52cc080658216fed6f296b2eaa84b6cfd6a75c3f4683b
+oid sha256:f5f1da1040925b2052aee2dca0c3fa273b607e957cbf18770059a4080553de3f
 size 14244

last-checkpoint/scaler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:575e45baa8bffe406e93f42f00b83861120a81ba67710e20edf9cb739fb34380
+oid sha256:8ac450af05edcacc287976d7cce0a697b2e2b1346a00fb8804e2fa8c2427dff3
 size 988

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:841075f91d905b0f965b453df9b09a92889e8e8880ee6a19f7c088b54a9e11c6
+oid sha256:01b7c9c1894540a3bb43365814662cd05e754a3caa17602ed4c796362bf454cf
 size 1064

last-checkpoint/trainer_state.json

@@ -1,10 +1,10 @@
 {
-  "best_global_step": 28500,
-  "best_metric": 1.0052642822265625,
-  "best_model_checkpoint": "./ar-diffusion-checkpoints-fixed/checkpoint-28500",
-  "epoch": 2.192139066225675,
+  "best_global_step": 29000,
+  "best_metric": 0.9999537467956543,
+  "best_model_checkpoint": "./ar-diffusion-checkpoints-fixed/checkpoint-29000",
+  "epoch": 2.230597646334897,
   "eval_steps": 250,
-  "global_step": 28500,
+  "global_step": 29000,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -4910,6 +4910,92 @@
       "eval_samples_per_second": 57.12,
       "eval_steps_per_second": 14.28,
       "step": 28500
+    },
+    {
+      "epoch": 2.195984924236597,
+      "grad_norm": 1.00751793384552,
+      "learning_rate": 5.436978936706231e-05,
+      "loss": 0.9479,
+      "step": 28550
+    },
+    {
+      "epoch": 2.1998307822475196,
+      "grad_norm": 0.9239784479141235,
+      "learning_rate": 5.411006934524583e-05,
+      "loss": 0.9896,
+      "step": 28600
+    },
+    {
+      "epoch": 2.2036766402584416,
+      "grad_norm": 0.9650816917419434,
+      "learning_rate": 5.385034932342935e-05,
+      "loss": 1.0081,
+      "step": 28650
+    },
+    {
+      "epoch": 2.207522498269364,
+      "grad_norm": 1.455723524093628,
+      "learning_rate": 5.359062930161286e-05,
+      "loss": 0.9717,
+      "step": 28700
+    },
+    {
+      "epoch": 2.211368356280286,
+      "grad_norm": 0.609380304813385,
+      "learning_rate": 5.333090927979638e-05,
+      "loss": 0.9774,
+      "step": 28750
+    },
+    {
+      "epoch": 2.211368356280286,
+      "eval_loss": 1.0018259286880493,
+      "eval_runtime": 17.3475,
+      "eval_samples_per_second": 57.645,
+      "eval_steps_per_second": 14.411,
+      "step": 28750
+    },
+    {
+      "epoch": 2.2152142142912084,
+      "grad_norm": 0.9981700778007507,
+      "learning_rate": 5.30711892579799e-05,
+      "loss": 0.9873,
+      "step": 28800
+    },
+    {
+      "epoch": 2.2190600723021308,
+      "grad_norm": 0.5452422499656677,
+      "learning_rate": 5.2811469236163414e-05,
+      "loss": 1.0078,
+      "step": 28850
+    },
+    {
+      "epoch": 2.2229059303130527,
+      "grad_norm": 1.4739840030670166,
+      "learning_rate": 5.255174921434693e-05,
+      "loss": 0.9739,
+      "step": 28900
+    },
+    {
+      "epoch": 2.226751788323975,
+      "grad_norm": 0.9250359535217285,
+      "learning_rate": 5.229202919253045e-05,
+      "loss": 0.9723,
+      "step": 28950
+    },
+    {
+      "epoch": 2.230597646334897,
+      "grad_norm": 1.0843122005462646,
+      "learning_rate": 5.203230917071398e-05,
+      "loss": 0.989,
+      "step": 29000
+    },
+    {
+      "epoch": 2.230597646334897,
+      "eval_loss": 0.9999537467956543,
+      "eval_runtime": 17.2417,
+      "eval_samples_per_second": 57.999,
+      "eval_steps_per_second": 14.5,
+      "step": 29000
     }
   ],
   "logging_steps": 50,