Training in progress, step 25000, checkpoint

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:864e393a661e445ab791d8ddc8820ce0124af5e25c563025e5274fb46871c164
+oid sha256:31864ac4bb9726b9569409ab5161fad26e94cbf3e7dd95686402bacb97d569e6
 size 36730224

last-checkpoint/ar_diffusion_info.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:69b19c7b3ab8d24c43cb3d97f2b5942e5102d4a46144bd92361e16d8ebb3d307
+oid sha256:7d52ac292e08ab1865944b2cee1c9d3b2be89e70ca0ae0fa8ad99b662c99bf03
 size 1736

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ae59df9b85342e49ccc0406baf5315bd2fd8f6e695f8cf5a8b8edc0181516fc6
+oid sha256:306ba75cd660966ab767dfb4cbb174b1b61b5756f39410aac084e39a57af127d
 size 73588346

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:741e3f6978068fd80c0a2938c07ae3df768f971aa6e2756e07360dec11f377ea
-size 14180
+oid sha256:f580d496f0048fe1cffdac35f5b2b7b4ede7ac0df721e48429fe4715fc745efa
+size 14244

last-checkpoint/scaler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8d31530c22f7fe8bcf00bdc50aaced04b559ddaa1caf679a3f57a45170ec249b
+oid sha256:9d0369b7df9f4f6057f54d12203a572c70823bfca8ac268e71fe1a019e47a330
 size 988

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:96928cc818781495e8d68be480c829211a700dd8771b47547694b6ed3ba139b1
+oid sha256:c9b07666b0c0a9f4236eb42880b6e9eedd3084d2bd56250a745ae04b4f3412ef
 size 1064

last-checkpoint/trainer_state.json

@@ -1,10 +1,10 @@
 {
-  "best_global_step": 24500,
-  "best_metric": 0.7536496520042419,
+  "best_global_step": 24750,
+  "best_metric": 0.7527515888214111,
   "best_model_checkpoint": "./ar-diffusion-checkpoints-fixed/checkpoint-24500",
-  "epoch": 1.884470425351896,
+  "epoch": 1.9229290054611183,
   "eval_steps": 250,
-  "global_step": 24500,
+  "global_step": 25000,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -4222,6 +4222,92 @@
       "eval_samples_per_second": 55.741,
       "eval_steps_per_second": 13.935,
       "step": 24500
+    },
+    {
+      "epoch": 1.8883162833628182,
+      "grad_norm": 0.8080185055732727,
+      "learning_rate": 7.512141911019921e-05,
+      "loss": 0.8067,
+      "step": 24550
+    },
+    {
+      "epoch": 1.8921621413737406,
+      "grad_norm": 0.5850221514701843,
+      "learning_rate": 7.486169908838274e-05,
+      "loss": 0.7455,
+      "step": 24600
+    },
+    {
+      "epoch": 1.8960079993846626,
+      "grad_norm": 0.7521384954452515,
+      "learning_rate": 7.460197906656624e-05,
+      "loss": 0.7604,
+      "step": 24650
+    },
+    {
+      "epoch": 1.899853857395585,
+      "grad_norm": 0.6376401782035828,
+      "learning_rate": 7.434225904474976e-05,
+      "loss": 0.7485,
+      "step": 24700
+    },
+    {
+      "epoch": 1.9036997154065072,
+      "grad_norm": 0.6305235624313354,
+      "learning_rate": 7.408253902293329e-05,
+      "loss": 0.7567,
+      "step": 24750
+    },
+    {
+      "epoch": 1.9036997154065072,
+      "eval_loss": 0.7527515888214111,
+      "eval_runtime": 17.967,
+      "eval_samples_per_second": 55.658,
+      "eval_steps_per_second": 13.914,
+      "step": 24750
+    },
+    {
+      "epoch": 1.9075455734174294,
+      "grad_norm": 0.5975210666656494,
+      "learning_rate": 7.382281900111679e-05,
+      "loss": 0.7855,
+      "step": 24800
+    },
+    {
+      "epoch": 1.9113914314283518,
+      "grad_norm": 0.41196370124816895,
+      "learning_rate": 7.356309897930031e-05,
+      "loss": 0.7711,
+      "step": 24850
+    },
+    {
+      "epoch": 1.9152372894392737,
+      "grad_norm": 1.1755207777023315,
+      "learning_rate": 7.330337895748384e-05,
+      "loss": 0.8114,
+      "step": 24900
+    },
+    {
+      "epoch": 1.9190831474501961,
+      "grad_norm": 0.37193945050239563,
+      "learning_rate": 7.304365893566735e-05,
+      "loss": 0.7437,
+      "step": 24950
+    },
+    {
+      "epoch": 1.9229290054611183,
+      "grad_norm": 0.6753848195075989,
+      "learning_rate": 7.278393891385087e-05,
+      "loss": 0.8289,
+      "step": 25000
+    },
+    {
+      "epoch": 1.9229290054611183,
+      "eval_loss": 0.7534742951393127,
+      "eval_runtime": 17.8978,
+      "eval_samples_per_second": 55.873,
+      "eval_steps_per_second": 13.968,
+      "step": 25000
     }
   ],
   "logging_steps": 50,