Training in progress, step 35000, checkpoint

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0a6ce4e2a543d8f0f121c7e0ba42ed65d29ab7f5a943ee9bc4c73eecc6e0c784
+oid sha256:fcbe886ef81e28ac4bf6ddf4529a6f997971fc7b47d5cc6b8463fde09b58e295
 size 36730224

last-checkpoint/ar_diffusion_info.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:67460d2a4cacdb12abbf640ed12faba90192eaa19550a17b7d7a4813aff0b245
+oid sha256:4e405c76f4c76f6b143004e04e6f5ef178d5eb79df9ddf579a7e536122d01ff4
 size 1736

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:098b3f939c72fb0e3883581fc3697edb7b3da16af98c5ea0dc091500b65f610b
+oid sha256:16b9dfda36c1191f9290f5c582e1fc3c2ff79c08ab7ee90848abb8367068ccb4
 size 73588346

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:627b267edccd634ef2533204cc126cbeee8cf658319bb821bc512f27abfec0db
+oid sha256:aa38e1254b2184d0a59e46181afc6b800a0f129dcd35d1277176baed5ca3bc70
 size 14244

last-checkpoint/scaler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ddfbea6602f43e7038cb128e2aa769d4c52b655d7b10760c5b89273a9ecca4ec
+oid sha256:b02b69899240ec8f846e73a10979f64f0f7812ffecfcd635d5625acc349acd03
 size 988

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:47cc0a6ee4ac28094e00435ddeaf659d411952db9ee5f95a42cb3b796293c8cc
+oid sha256:3c76a5b616a8159a5073e044f9b5bf193ac0fcff7d79d7b67f44d5065ac4e6fa
 size 1064

last-checkpoint/trainer_state.json

@@ -1,10 +1,10 @@
 {
-  "best_global_step": 34500,
-  "best_metric": -30.46938705444336,
-  "best_model_checkpoint": "./ar-diffusion-checkpoints-progressive-attention/checkpoint-34500",
-  "epoch": 2.6536420275363435,
+  "best_global_step": 35000,
+  "best_metric": -30.46967315673828,
+  "best_model_checkpoint": "./ar-diffusion-checkpoints-progressive-attention/checkpoint-35000",
+  "epoch": 2.692100607645566,
   "eval_steps": 250,
-  "global_step": 34500,
+  "global_step": 35000,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -5942,6 +5942,92 @@
       "eval_samples_per_second": 59.369,
       "eval_steps_per_second": 14.842,
       "step": 34500
+    },
+    {
+      "epoch": 2.6574878855472654,
+      "grad_norm": 0.7920024991035461,
+      "learning_rate": 2.3213775549957147e-05,
+      "loss": -30.4646,
+      "step": 34550
+    },
+    {
+      "epoch": 2.661333743558188,
+      "grad_norm": 1.470051646232605,
+      "learning_rate": 2.2954055528140667e-05,
+      "loss": -30.4651,
+      "step": 34600
+    },
+    {
+      "epoch": 2.6651796015691103,
+      "grad_norm": 0.7992119193077087,
+      "learning_rate": 2.2694335506324183e-05,
+      "loss": -30.4645,
+      "step": 34650
+    },
+    {
+      "epoch": 2.6690254595800322,
+      "grad_norm": 0.9150895476341248,
+      "learning_rate": 2.2434615484507702e-05,
+      "loss": -30.4651,
+      "step": 34700
+    },
+    {
+      "epoch": 2.6728713175909546,
+      "grad_norm": 1.7401885986328125,
+      "learning_rate": 2.217489546269122e-05,
+      "loss": -30.4645,
+      "step": 34750
+    },
+    {
+      "epoch": 2.6728713175909546,
+      "eval_loss": -30.468637466430664,
+      "eval_runtime": 16.8394,
+      "eval_samples_per_second": 59.384,
+      "eval_steps_per_second": 14.846,
+      "step": 34750
+    },
+    {
+      "epoch": 2.676717175601877,
+      "grad_norm": 1.014699935913086,
+      "learning_rate": 2.1915175440874738e-05,
+      "loss": -30.465,
+      "step": 34800
+    },
+    {
+      "epoch": 2.680563033612799,
+      "grad_norm": 0.9657973051071167,
+      "learning_rate": 2.1655455419058258e-05,
+      "loss": -30.465,
+      "step": 34850
+    },
+    {
+      "epoch": 2.684408891623721,
+      "grad_norm": 1.1033378839492798,
+      "learning_rate": 2.1395735397241774e-05,
+      "loss": -30.465,
+      "step": 34900
+    },
+    {
+      "epoch": 2.6882547496346434,
+      "grad_norm": 3.0065767765045166,
+      "learning_rate": 2.113601537542529e-05,
+      "loss": -30.4651,
+      "step": 34950
+    },
+    {
+      "epoch": 2.692100607645566,
+      "grad_norm": 1.226468563079834,
+      "learning_rate": 2.087629535360881e-05,
+      "loss": -30.4653,
+      "step": 35000
+    },
+    {
+      "epoch": 2.692100607645566,
+      "eval_loss": -30.46967315673828,
+      "eval_runtime": 16.9427,
+      "eval_samples_per_second": 59.022,
+      "eval_steps_per_second": 14.756,
+      "step": 35000
     }
   ],
   "logging_steps": 50,