Training in progress, step 1000, checkpoint

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:fae9e588eaaf24844115c1eddab6702485f0e276176d9905391edb7633c50bea
+oid sha256:5f02ec6b82a831f2902a910cf8fafe953154448b3b45acb102d57d8391c23b96
 size 36730224

last-checkpoint/ar_diffusion_info.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:46cbbd66b0fad4cd21043dde397564b4a5a43cb08fd3e63a8a9d298088a80ea8
+oid sha256:cbe39982c4bebe4e0b3f0f76afe1c485909ad223aac0a865ef49d364896c3cff
 size 1736

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:fecf71fcdecddc7b1d484764aa9989b5c7a376940de2345dbcd4727173f1a8cc
+oid sha256:c326c13c75092245145edf95faedf4bbc2e6d95bbc78269d58ccc6219fd83b99
 size 73588346

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6091cec722ce83576335fe69e6a4694ee54be33f028be0bbcfa0367b80c50012
+oid sha256:31bb3f2abd407f62d555fa53e113286139c6abb36d00dab486b238b592c7d71a
 size 14244

last-checkpoint/scaler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ad6acbbacc4296909519356cab55a9b5586a68ca85ba3af712eed9f037b1e773
+oid sha256:2229a15bdf42c8dbae70a65f18e33cde5c16363745edce9ae10c3d4c94d5ed8e
 size 988

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:df8159a5d54d37b240325feb44ba0ce415d32bac74fe814243062911256e4717
+oid sha256:9d88837be00a1ec9be1702d7436c136fd03c05aff969a737326a957a88901e17
 size 1064

last-checkpoint/trainer_state.json

@@ -1,10 +1,10 @@
 {
-  "best_global_step": 500,
-  "best_metric": 1.7362440824508667,
-  "best_model_checkpoint": "./ar-diffusion-checkpoints-progressive-attention/checkpoint-500",
-  "epoch": 0.038458580109222366,
+  "best_global_step": 1000,
+  "best_metric": 1.6695575714111328,
+  "best_model_checkpoint": "./ar-diffusion-checkpoints-progressive-attention/checkpoint-1000",
+  "epoch": 0.07691716021844473,
   "eval_steps": 250,
-  "global_step": 500,
+  "global_step": 1000,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -94,6 +94,92 @@
       "eval_samples_per_second": 59.399,
       "eval_steps_per_second": 14.85,
       "step": 500
+    },
+    {
+      "epoch": 0.0423044381201446,
+      "grad_norm": 1.872428297996521,
+      "learning_rate": 0.00019976105757992885,
+      "loss": 1.7652,
+      "step": 550
+    },
+    {
+      "epoch": 0.04615029613106684,
+      "grad_norm": 1.5747038125991821,
+      "learning_rate": 0.00019950133755811236,
+      "loss": 1.7437,
+      "step": 600
+    },
+    {
+      "epoch": 0.04999615414198908,
+      "grad_norm": 2.0243659019470215,
+      "learning_rate": 0.00019924161753629587,
+      "loss": 1.6628,
+      "step": 650
+    },
+    {
+      "epoch": 0.05384201215291132,
+      "grad_norm": 1.1568059921264648,
+      "learning_rate": 0.0001989818975144794,
+      "loss": 1.6787,
+      "step": 700
+    },
+    {
+      "epoch": 0.05768787016383355,
+      "grad_norm": 1.8709088563919067,
+      "learning_rate": 0.0001987221774926629,
+      "loss": 1.7599,
+      "step": 750
+    },
+    {
+      "epoch": 0.05768787016383355,
+      "eval_loss": 1.6890833377838135,
+      "eval_runtime": 16.858,
+      "eval_samples_per_second": 59.319,
+      "eval_steps_per_second": 14.83,
+      "step": 750
+    },
+    {
+      "epoch": 0.06153372817475579,
+      "grad_norm": 1.4875328540802002,
+      "learning_rate": 0.00019846245747084644,
+      "loss": 1.7187,
+      "step": 800
+    },
+    {
+      "epoch": 0.06537958618567802,
+      "grad_norm": 1.9822659492492676,
+      "learning_rate": 0.00019820273744902995,
+      "loss": 1.7005,
+      "step": 850
+    },
+    {
+      "epoch": 0.06922544419660026,
+      "grad_norm": 1.3554214239120483,
+      "learning_rate": 0.0001979430174272135,
+      "loss": 1.7001,
+      "step": 900
+    },
+    {
+      "epoch": 0.0730713022075225,
+      "grad_norm": 0.9904000163078308,
+      "learning_rate": 0.000197683297405397,
+      "loss": 1.6042,
+      "step": 950
+    },
+    {
+      "epoch": 0.07691716021844473,
+      "grad_norm": 1.4929814338684082,
+      "learning_rate": 0.00019742877178401683,
+      "loss": 1.7251,
+      "step": 1000
+    },
+    {
+      "epoch": 0.07691716021844473,
+      "eval_loss": 1.6695575714111328,
+      "eval_runtime": 16.8334,
+      "eval_samples_per_second": 59.406,
+      "eval_steps_per_second": 14.851,
+      "step": 1000
     }
   ],
   "logging_steps": 50,