Training in progress, step 39000, checkpoint

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d9fccd6e8cf77b198a4f4b87f312750517824202a02166eb04e56577ecd2ba00
+oid sha256:aec93bf3af378bc69e279d3da1f22a2356cd072cc873629676a0de486c684213
 size 36730224

last-checkpoint/ar_diffusion_info.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:66d49ec8e5680b4ea0cac8d1930f19a010f9ae0bd4d5ca4b158ebc77b3d54acd
+oid sha256:b1e17b4b57a9838b7215832af6318f536b3b6acee7c730845278b1a8f5d0cb17
 size 1736

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:38f23e9c9a2a443780a64b0cf18b93c2bd78a442ad08f1e37293ea92f7c88993
+oid sha256:5e8f0bbc447de59f2033199d6477b7b4db91aeaa9c681dfd30f7ca1df276b9b1
 size 73588346

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b1b4df89a65698a9335c43463fa8aa3cc49c1acdfae5f7a285af000ba78e95b1
-size 14244
+oid sha256:60919181f100a38a6bc8dcdeb135786c172525306422517244d9c17e263e8748
+size 14180

last-checkpoint/scaler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1a1d56165e0890bb9432c217b765bdd79a8b4d93dcde80b4671874513824409b
+oid sha256:5cd25c48749969bd5bf972a3edb2b91874acf34f950f422e77399a6589ee1816
 size 988

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f3dea87c40748698e240c93bc3eea5482b3f627ab3a1ed391f6cbe2fe5d7da89
+oid sha256:06d9f9352b3f9aa43ec304f953eb511fb6c42be93c910bdb749b28823ddd5892
 size 1064

last-checkpoint/trainer_state.json

@@ -2,9 +2,9 @@
   "best_global_step": 38250,
   "best_metric": -30.471132278442383,
   "best_model_checkpoint": "./ar-diffusion-checkpoints-progressive-attention/checkpoint-37500",
-  "epoch": 2.961310668410122,
+  "epoch": 2.999769248519345,
   "eval_steps": 250,
-  "global_step": 38500,
+  "global_step": 39000,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -6630,6 +6630,92 @@
       "eval_samples_per_second": 59.287,
       "eval_steps_per_second": 14.822,
       "step": 38500
+    },
+    {
+      "epoch": 2.9651565264210444,
+      "grad_norm": 0.5108075737953186,
+      "learning_rate": 2.4361738046385997e-06,
+      "loss": -30.467,
+      "step": 38550
+    },
+    {
+      "epoch": 2.969002384431967,
+      "grad_norm": 0.608711838722229,
+      "learning_rate": 2.1764537828221176e-06,
+      "loss": -30.467,
+      "step": 38600
+    },
+    {
+      "epoch": 2.972848242442889,
+      "grad_norm": 0.9723970890045166,
+      "learning_rate": 1.916733761005636e-06,
+      "loss": -30.4672,
+      "step": 38650
+    },
+    {
+      "epoch": 2.976694100453811,
+      "grad_norm": 0.4361710846424103,
+      "learning_rate": 1.6570137391891543e-06,
+      "loss": -30.467,
+      "step": 38700
+    },
+    {
+      "epoch": 2.9805399584647336,
+      "grad_norm": 1.1543878316879272,
+      "learning_rate": 1.3972937173726722e-06,
+      "loss": -30.4669,
+      "step": 38750
+    },
+    {
+      "epoch": 2.9805399584647336,
+      "eval_loss": -30.47084617614746,
+      "eval_runtime": 16.7842,
+      "eval_samples_per_second": 59.58,
+      "eval_steps_per_second": 14.895,
+      "step": 38750
+    },
+    {
+      "epoch": 2.9843858164756556,
+      "grad_norm": 0.4299401640892029,
+      "learning_rate": 1.1375736955561905e-06,
+      "loss": -30.467,
+      "step": 38800
+    },
+    {
+      "epoch": 2.988231674486578,
+      "grad_norm": 0.8476674556732178,
+      "learning_rate": 8.778536737397087e-07,
+      "loss": -30.467,
+      "step": 38850
+    },
+    {
+      "epoch": 2.9920775324975004,
+      "grad_norm": 1.1194632053375244,
+      "learning_rate": 6.181336519232268e-07,
+      "loss": -30.4671,
+      "step": 38900
+    },
+    {
+      "epoch": 2.9959233905084224,
+      "grad_norm": 0.7896573543548584,
+      "learning_rate": 3.584136301067449e-07,
+      "loss": -30.4671,
+      "step": 38950
+    },
+    {
+      "epoch": 2.999769248519345,
+      "grad_norm": 0.32893452048301697,
+      "learning_rate": 9.869360829026311e-08,
+      "loss": -30.4671,
+      "step": 39000
+    },
+    {
+      "epoch": 2.999769248519345,
+      "eval_loss": -30.471084594726562,
+      "eval_runtime": 16.7188,
+      "eval_samples_per_second": 59.813,
+      "eval_steps_per_second": 14.953,
+      "step": 39000
     }
   ],
   "logging_steps": 50,