Training in progress, step 28000, checkpoint

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:395d6f2f3db4017a2737909d6b4ea9e9031c52589c6f6d040f85bb9bc9a61574
+oid sha256:62439bae41ea1998b90f6037dd9fa884cf659248f0c85d5c8a744390c2886fa9
 size 36730224

last-checkpoint/ar_diffusion_info.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:33bfe006e215fece89dfe8777344c7d6ce9d541fd4ce48109267ec6314d1a517
+oid sha256:a0db08bb0e64e6fdc76f068a4dce33c014fc080b283b9a64c9a60ab6193a7f49
 size 1736

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b9df54078255f6bdb0657e3789fa1dc2e87d9dd9aedcc964174fb1c8597c5616
+oid sha256:370e87188f637f83ed8349bf7a8b60a845c5cbc7b0913ca617016b425f823961
 size 73588346

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:547f82cfc12d87310740e670a337bffc2b27a4a004bc784a756302a28851318f
+oid sha256:0f0206c699fd966569a4deff507474ca9114467664863e2a7c8efd89e088e4c4
 size 14244

last-checkpoint/scaler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:aafb160e86ca3dae2151e7d5410c1f0f3d6a45d02cb075a2fd74acbbb6438223
+oid sha256:a80611879827bf00dd9a4f888e494a8a2af0eea4eada726997ec5489de20e821
 size 988

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:00e4057ddb93d4201ababcbc4c1556bd476cd599690ef79c5880a1b8a05ecfad
+oid sha256:59377233a54e779461bcaa1b904cc8d8243d8ae8f5afd52f30216a544a19a619
 size 1064

last-checkpoint/trainer_state.json

@@ -1,10 +1,10 @@
 {
-  "best_global_step": 27500,
-  "best_metric": 0.7388279438018799,
+  "best_global_step": 27750,
+  "best_metric": 0.736056387424469,
   "best_model_checkpoint": "./ar-diffusion-checkpoints-fixed/checkpoint-27500",
-  "epoch": 2.11522190600723,
+  "epoch": 2.1536804861164525,
   "eval_steps": 250,
-  "global_step": 27500,
+  "global_step": 28000,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -4738,6 +4738,92 @@
       "eval_samples_per_second": 55.86,
       "eval_steps_per_second": 13.965,
       "step": 27500
+    },
+    {
+      "epoch": 2.1190677640181526,
+      "grad_norm": 0.8702675104141235,
+      "learning_rate": 5.955380100251928e-05,
+      "loss": 0.7034,
+      "step": 27550
+    },
+    {
+      "epoch": 2.1229136220290745,
+      "grad_norm": 0.9476292729377747,
+      "learning_rate": 5.929408098070281e-05,
+      "loss": 0.7075,
+      "step": 27600
+    },
+    {
+      "epoch": 2.126759480039997,
+      "grad_norm": 0.7763323783874512,
+      "learning_rate": 5.903436095888633e-05,
+      "loss": 0.7401,
+      "step": 27650
+    },
+    {
+      "epoch": 2.1306053380509193,
+      "grad_norm": 0.9647789001464844,
+      "learning_rate": 5.8774640937069847e-05,
+      "loss": 0.7308,
+      "step": 27700
+    },
+    {
+      "epoch": 2.1344511960618413,
+      "grad_norm": 0.7125420570373535,
+      "learning_rate": 5.851492091525336e-05,
+      "loss": 0.7337,
+      "step": 27750
+    },
+    {
+      "epoch": 2.1344511960618413,
+      "eval_loss": 0.736056387424469,
+      "eval_runtime": 18.0862,
+      "eval_samples_per_second": 55.291,
+      "eval_steps_per_second": 13.823,
+      "step": 27750
+    },
+    {
+      "epoch": 2.1382970540727637,
+      "grad_norm": 1.1053630113601685,
+      "learning_rate": 5.825520089343688e-05,
+      "loss": 0.7434,
+      "step": 27800
+    },
+    {
+      "epoch": 2.1421429120836857,
+      "grad_norm": 0.751206636428833,
+      "learning_rate": 5.79954808716204e-05,
+      "loss": 0.7208,
+      "step": 27850
+    },
+    {
+      "epoch": 2.145988770094608,
+      "grad_norm": 1.0049408674240112,
+      "learning_rate": 5.773576084980391e-05,
+      "loss": 0.7265,
+      "step": 27900
+    },
+    {
+      "epoch": 2.1498346281055305,
+      "grad_norm": 0.9738804697990417,
+      "learning_rate": 5.747604082798743e-05,
+      "loss": 0.7515,
+      "step": 27950
+    },
+    {
+      "epoch": 2.1536804861164525,
+      "grad_norm": 0.7807592153549194,
+      "learning_rate": 5.721632080617095e-05,
+      "loss": 0.7309,
+      "step": 28000
+    },
+    {
+      "epoch": 2.1536804861164525,
+      "eval_loss": 0.736193835735321,
+      "eval_runtime": 17.9436,
+      "eval_samples_per_second": 55.73,
+      "eval_steps_per_second": 13.933,
+      "step": 28000
     }
   ],
   "logging_steps": 50,