Training in progress, step 9500, checkpoint

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c025d03493052da11f2e276acb3d21a6d929e8421e1c7592ebeacd87d114cd85
+oid sha256:46059868a56444f774675df268eb914452898f29b7df82787796af4c7fa5d48d
 size 36730224

last-checkpoint/ar_diffusion_info.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ed85d0da178bc8a897d04aed9fccd3af1551001178952fa7b7d9b07378424006
+oid sha256:d22244b9bcee513c0db5fa0dd4382afaa1310731018b62ed1ec48732dc32403d
 size 1736

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:79658bc05ff8e823814b286e695d1366121826201ccee57ea590843dcddf9763
+oid sha256:24d290d02846044f2e528ff22c1dc13c928158f90f0b3902a3633693e875f22d
 size 73588346

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c9fe45a2e23b5067bfb664c7c87d67578826fdd3c8538fe28491021efb33c2e2
-size 14180
+oid sha256:330590f2dd931f25cb518a6afcc3373fe0b53fe3178be67482b5a6355a86a689
+size 14244

last-checkpoint/scaler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0ed1ff55cffa4c7997d35d9e9aec781ef5a02c8194294d8fa15d89ecfc4185f6
+oid sha256:90087180835653f1b7a3f029c8126085005e238a2fd17bd1f7457696339c6b13
 size 988

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4ea10c4ee31d77ef3033b6c398543d815e76e7ca17a3e28c3601be5a6eebbb15
+oid sha256:121ffc46f9f634ab597be8168f47dd7955b764f9421f382e55c4e5758e2a1624
 size 1064

last-checkpoint/trainer_state.json

@@ -1,10 +1,10 @@
 {
-  "best_global_step": 9000,
-  "best_metric": 1.902275800704956,
-  "best_model_checkpoint": "./ar-diffusion-checkpoints-progressive-attention/checkpoint-9000",
-  "epoch": 0.6922544419660026,
+  "best_global_step": 9500,
+  "best_metric": 1.8725571632385254,
+  "best_model_checkpoint": "./ar-diffusion-checkpoints-progressive-attention/checkpoint-9500",
+  "epoch": 0.7307130220752249,
   "eval_steps": 250,
-  "global_step": 9000,
+  "global_step": 9500,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -1556,6 +1556,92 @@
       "eval_samples_per_second": 58.972,
       "eval_steps_per_second": 14.743,
       "step": 9000
+    },
+    {
+      "epoch": 0.6961002999769248,
+      "grad_norm": 3.6892011165618896,
+      "learning_rate": 0.00015562423707243591,
+      "loss": 1.9551,
+      "step": 9050
+    },
+    {
+      "epoch": 0.6999461579878471,
+      "grad_norm": 1.524671196937561,
+      "learning_rate": 0.00015536451705061943,
+      "loss": 1.9109,
+      "step": 9100
+    },
+    {
+      "epoch": 0.7037920159987693,
+      "grad_norm": 1.5293575525283813,
+      "learning_rate": 0.00015510479702880297,
+      "loss": 1.8891,
+      "step": 9150
+    },
+    {
+      "epoch": 0.7076378740096916,
+      "grad_norm": 2.321476697921753,
+      "learning_rate": 0.00015484507700698648,
+      "loss": 1.9021,
+      "step": 9200
+    },
+    {
+      "epoch": 0.7114837320206138,
+      "grad_norm": 2.710942029953003,
+      "learning_rate": 0.00015458535698517,
+      "loss": 1.9414,
+      "step": 9250
+    },
+    {
+      "epoch": 0.7114837320206138,
+      "eval_loss": 1.9107917547225952,
+      "eval_runtime": 16.997,
+      "eval_samples_per_second": 58.834,
+      "eval_steps_per_second": 14.708,
+      "step": 9250
+    },
+    {
+      "epoch": 0.7153295900315361,
+      "grad_norm": 1.9385954141616821,
+      "learning_rate": 0.0001543256369633535,
+      "loss": 1.9401,
+      "step": 9300
+    },
+    {
+      "epoch": 0.7191754480424583,
+      "grad_norm": 2.589629650115967,
+      "learning_rate": 0.00015406591694153702,
+      "loss": 1.9054,
+      "step": 9350
+    },
+    {
+      "epoch": 0.7230213060533806,
+      "grad_norm": 1.6431207656860352,
+      "learning_rate": 0.00015380619691972056,
+      "loss": 1.9324,
+      "step": 9400
+    },
+    {
+      "epoch": 0.7268671640643027,
+      "grad_norm": 4.840892314910889,
+      "learning_rate": 0.00015354647689790407,
+      "loss": 1.993,
+      "step": 9450
+    },
+    {
+      "epoch": 0.7307130220752249,
+      "grad_norm": 0.9328492879867554,
+      "learning_rate": 0.00015328675687608758,
+      "loss": 1.8637,
+      "step": 9500
+    },
+    {
+      "epoch": 0.7307130220752249,
+      "eval_loss": 1.8725571632385254,
+      "eval_runtime": 16.9918,
+      "eval_samples_per_second": 58.852,
+      "eval_steps_per_second": 14.713,
+      "step": 9500
     }
   ],
   "logging_steps": 50,