Training in progress, step 29000, checkpoint

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f6600bba0d4ff6b96aebe142a8f1d29d3689bd6d93b2ebd674227ef088bff7b8
+oid sha256:3f7e588a197a8570159058327c70025893ca766d0fbd2a25bb0403e673011256
 size 36730224

last-checkpoint/ar_diffusion_info.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:bf9bcfbac04fd11bcd15980c9df2ea854394b3bb5d97d731829d75371fbaa955
+oid sha256:84eb396661bc75d364b2d44d83a4c4960855665d430d64c5097382769b27773b
 size 1736

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:72afdb8f3d5d0bd15446a4f8cd0fbad5718bb8f7796245e8f6d72c43db0cdada
+oid sha256:95894271fa4fbddfeacbe1ad5d29056d6607d4cbede968d833f230635c10694c
 size 73588346

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:25bf93d553915ff8c50a7931af03f2d7b533b6db51da15a573325147ba05e268
-size 14244
+oid sha256:bdc223e0883d03981b825b4fc5e59f6538251017a05a433c157f563714a2f9ba
+size 14180

last-checkpoint/scaler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5bfb7a12fdede60278d2b32e52f50469e6e77b619720bcb0e4bd039e0a83e832
+oid sha256:24e0d66d723576c324a7fe34e5dbff8e5703e0dde1366b79228a8fb754fe8fd9
 size 988

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e4e46f9024365cd773095a35799f9dd86ead6c135c47c788f2a21bedfef6883c
+oid sha256:e9c653c5f990cfe923931e3bba4aa0b8bc89aeef6f3280f590db66eda67a330f
 size 1064

last-checkpoint/trainer_state.json

@@ -1,10 +1,10 @@
 {
-  "best_global_step": 28250,
-  "best_metric": 0.7313055396080017,
-  "best_model_checkpoint": "./ar-diffusion-checkpoints-fixed/checkpoint-27500",
-  "epoch": 2.192139066225675,
+  "best_global_step": 29000,
+  "best_metric": 0.7284311652183533,
+  "best_model_checkpoint": "./ar-diffusion-checkpoints-fixed/checkpoint-29000",
+  "epoch": 2.230597646334897,
   "eval_steps": 250,
-  "global_step": 28500,
+  "global_step": 29000,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -4910,6 +4910,92 @@
       "eval_samples_per_second": 55.478,
       "eval_steps_per_second": 13.87,
       "step": 28500
+    },
+    {
+      "epoch": 2.195984924236597,
+      "grad_norm": 0.8648149371147156,
+      "learning_rate": 5.435940056618965e-05,
+      "loss": 0.6874,
+      "step": 28550
+    },
+    {
+      "epoch": 2.1998307822475196,
+      "grad_norm": 0.6547895073890686,
+      "learning_rate": 5.4099680544373166e-05,
+      "loss": 0.7115,
+      "step": 28600
+    },
+    {
+      "epoch": 2.2036766402584416,
+      "grad_norm": 0.7705133557319641,
+      "learning_rate": 5.3839960522556685e-05,
+      "loss": 0.7498,
+      "step": 28650
+    },
+    {
+      "epoch": 2.207522498269364,
+      "grad_norm": 1.2679988145828247,
+      "learning_rate": 5.358024050074021e-05,
+      "loss": 0.6956,
+      "step": 28700
+    },
+    {
+      "epoch": 2.211368356280286,
+      "grad_norm": 0.4992118775844574,
+      "learning_rate": 5.332052047892372e-05,
+      "loss": 0.7124,
+      "step": 28750
+    },
+    {
+      "epoch": 2.211368356280286,
+      "eval_loss": 0.7316113710403442,
+      "eval_runtime": 17.8507,
+      "eval_samples_per_second": 56.02,
+      "eval_steps_per_second": 14.005,
+      "step": 28750
+    },
+    {
+      "epoch": 2.2152142142912084,
+      "grad_norm": 0.8134187459945679,
+      "learning_rate": 5.306080045710724e-05,
+      "loss": 0.733,
+      "step": 28800
+    },
+    {
+      "epoch": 2.2190600723021308,
+      "grad_norm": 0.5038111805915833,
+      "learning_rate": 5.2801080435290764e-05,
+      "loss": 0.7239,
+      "step": 28850
+    },
+    {
+      "epoch": 2.2229059303130527,
+      "grad_norm": 1.0252164602279663,
+      "learning_rate": 5.254136041347427e-05,
+      "loss": 0.7102,
+      "step": 28900
+    },
+    {
+      "epoch": 2.226751788323975,
+      "grad_norm": 0.823451817035675,
+      "learning_rate": 5.2281640391657796e-05,
+      "loss": 0.709,
+      "step": 28950
+    },
+    {
+      "epoch": 2.230597646334897,
+      "grad_norm": 0.7642868161201477,
+      "learning_rate": 5.2021920369841316e-05,
+      "loss": 0.7326,
+      "step": 29000
+    },
+    {
+      "epoch": 2.230597646334897,
+      "eval_loss": 0.7284311652183533,
+      "eval_runtime": 17.8949,
+      "eval_samples_per_second": 55.882,
+      "eval_steps_per_second": 13.97,
+      "step": 29000
     }
   ],
   "logging_steps": 50,