Training in progress, step 8000, checkpoint

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1db6fd6ee1d14415c983da271a1a5fced7a19d546dcff5e9c52892498d1357ea
+oid sha256:f8cf14ec041a3170fc93bd13962e58c3433da275c26e3ddcd248f44fe9261656
 size 36730224

last-checkpoint/ar_diffusion_info.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:90e5824e7895d4d885fcae3d6d27b58bf56f7ed9334ac10f6cfbd2e2106a176c
+oid sha256:c5afd0a70049fd01ffc5b0b2af6dbbfcaccc76963457c541c72cb41caeec8cf4
 size 1736

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:636d50d708bc2858a479246fdb19d62e4927d92e0d54e0a5e51e6cc5b17b5a5c
+oid sha256:1383466fc8ec131ed5aea99415fe444f7a641621908ee8efda1f95cd39e73d3d
 size 73588346

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:340279b529434b0154ba013e9a01bef12d3018e0cbfd933ce4ed580c0bdf70f8
+oid sha256:565e9a1f20b4e27434c54601d243feb5ec67c27b2d167d2e9e70202e4cc1bb10
 size 14180

last-checkpoint/scaler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d034bc1613cdbec4742e0b9343e54d205e633fa15e81421b5d524ce6b4165764
+oid sha256:f4875a6ebf2ebe9f7eec596109836a4abaa1f225d707484c4b3129f5845c484e
 size 988

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2541b57467e961f49663f187822c9553808b853819403f17ca1464ff8f14b868
+oid sha256:763a8e91c3a700b449d0501649e8a0112790ca7b7dbb67a7e5e5dcfd39db55d8
 size 1064

last-checkpoint/trainer_state.json

@@ -1,10 +1,10 @@
 {
-  "best_global_step": 7500,
-  "best_metric": 1.424470067024231,
-  "best_model_checkpoint": "./ar-diffusion-checkpoints-progressive-attention/checkpoint-7500",
-  "epoch": 0.5768787016383355,
+  "best_global_step": 8000,
+  "best_metric": 1.410463809967041,
+  "best_model_checkpoint": "./ar-diffusion-checkpoints-progressive-attention/checkpoint-8000",
+  "epoch": 0.6153372817475579,
   "eval_steps": 250,
-  "global_step": 7500,
+  "global_step": 8000,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -1298,6 +1298,92 @@
       "eval_samples_per_second": 59.328,
       "eval_steps_per_second": 14.832,
       "step": 7500
+    },
+    {
+      "epoch": 0.5807245596492577,
+      "grad_norm": 1.6724326610565186,
+      "learning_rate": 0.00016341064332649406,
+      "loss": 1.4254,
+      "step": 7550
+    },
+    {
+      "epoch": 0.58457041766018,
+      "grad_norm": 0.871095597743988,
+      "learning_rate": 0.00016315092330467757,
+      "loss": 1.3679,
+      "step": 7600
+    },
+    {
+      "epoch": 0.5884162756711022,
+      "grad_norm": 1.2308543920516968,
+      "learning_rate": 0.00016289120328286108,
+      "loss": 1.4699,
+      "step": 7650
+    },
+    {
+      "epoch": 0.5922621336820244,
+      "grad_norm": 1.0057512521743774,
+      "learning_rate": 0.0001626314832610446,
+      "loss": 1.3838,
+      "step": 7700
+    },
+    {
+      "epoch": 0.5961079916929467,
+      "grad_norm": 2.5127899646759033,
+      "learning_rate": 0.0001623717632392281,
+      "loss": 1.4696,
+      "step": 7750
+    },
+    {
+      "epoch": 0.5961079916929467,
+      "eval_loss": 1.4130176305770874,
+      "eval_runtime": 16.9059,
+      "eval_samples_per_second": 59.151,
+      "eval_steps_per_second": 14.788,
+      "step": 7750
+    },
+    {
+      "epoch": 0.5999538497038689,
+      "grad_norm": 0.9064082503318787,
+      "learning_rate": 0.00016211204321741165,
+      "loss": 1.4517,
+      "step": 7800
+    },
+    {
+      "epoch": 0.6037997077147912,
+      "grad_norm": 0.8912540674209595,
+      "learning_rate": 0.00016185232319559516,
+      "loss": 1.4393,
+      "step": 7850
+    },
+    {
+      "epoch": 0.6076455657257134,
+      "grad_norm": 1.231679916381836,
+      "learning_rate": 0.00016159260317377867,
+      "loss": 1.4221,
+      "step": 7900
+    },
+    {
+      "epoch": 0.6114914237366357,
+      "grad_norm": 0.9336662292480469,
+      "learning_rate": 0.0001613328831519622,
+      "loss": 1.4739,
+      "step": 7950
+    },
+    {
+      "epoch": 0.6153372817475579,
+      "grad_norm": 1.313412070274353,
+      "learning_rate": 0.0001610731631301457,
+      "loss": 1.4792,
+      "step": 8000
+    },
+    {
+      "epoch": 0.6153372817475579,
+      "eval_loss": 1.410463809967041,
+      "eval_runtime": 16.8522,
+      "eval_samples_per_second": 59.34,
+      "eval_steps_per_second": 14.835,
+      "step": 8000
     }
   ],
   "logging_steps": 50,