Training in progress, step 27000, checkpoint

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f619cbe8165ed1e2332f6436e4fdf6c896db5237d638a5179f03f772e54deae1
+oid sha256:0c71cb6762a3f2359c9b24c07c0f2c18222a0855ef4122d25dd24f1be773077c
 size 36730224

last-checkpoint/ar_diffusion_info.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0e41f1a18d6f9531a4bd645cd462d9090e6f54162f6fddf017afc585f3099347
+oid sha256:c7562e2e11d39589f1d1677624c9cbfb42e6035b41d3ee891b0f53a0cc6bfe79
 size 1544

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f75a42181e10c429bdf4cde150fa8bd273b194edd1af255a2c46af2a16e4d777
+oid sha256:dc1dfd01f40336797b539a0d0ca763264a6fc70cdf83f8431ae379e9a2d8dff9
 size 73588346

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5b56501523df118c1a33e60d970ee258e92691efddadd68cb368e352ca4fb0c1
+oid sha256:a7cb9cecde71619f5455094db9dbc10b01ced14b7fff166fcfd1f46df6e480f1
 size 14244

last-checkpoint/scaler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3bfa502239c4234c76e8429085577f9d92a7a11c10a2a19927df506f4396be9a
+oid sha256:e2a94b30a1039d003ca39bb6fcc94430e37b0c6c69456a996151d312d106dd61
 size 988

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:cde03cf528fa40100e793e6bc03d3e643c03bab96500bb8c5705c4e1018be13d
+oid sha256:1412b21a67b3e14f57e3f3d27ac855c98641b3c473e09eb37b8289635d633513
 size 1064

last-checkpoint/trainer_state.json

@@ -2,9 +2,9 @@
   "best_global_step": 24500,
   "best_metric": 1.445096731185913,
   "best_model_checkpoint": "./ar-diffusion-checkpoints-fixed/checkpoint-24500",
-  "epoch": 2.0383047457887855,
+  "epoch": 2.076763325898008,
   "eval_steps": 250,
-  "global_step": 26500,
+  "global_step": 27000,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -4566,6 +4566,92 @@
       "eval_samples_per_second": 55.83,
       "eval_steps_per_second": 13.958,
       "step": 26500
+    },
+    {
+      "epoch": 2.042150603799708,
+      "grad_norm": 1.0810669660568237,
+      "learning_rate": 6.475859023972158e-05,
+      "loss": 1.2936,
+      "step": 26550
+    },
+    {
+      "epoch": 2.04599646181063,
+      "grad_norm": 1.23382568359375,
+      "learning_rate": 6.44988702179051e-05,
+      "loss": 1.3922,
+      "step": 26600
+    },
+    {
+      "epoch": 2.0498423198215523,
+      "grad_norm": 1.8218950033187866,
+      "learning_rate": 6.423915019608862e-05,
+      "loss": 1.4041,
+      "step": 26650
+    },
+    {
+      "epoch": 2.0536881778324743,
+      "grad_norm": 0.6482899785041809,
+      "learning_rate": 6.397943017427213e-05,
+      "loss": 1.3609,
+      "step": 26700
+    },
+    {
+      "epoch": 2.0575340358433967,
+      "grad_norm": 1.9538156986236572,
+      "learning_rate": 6.371971015245566e-05,
+      "loss": 1.4558,
+      "step": 26750
+    },
+    {
+      "epoch": 2.0575340358433967,
+      "eval_loss": 1.454428791999817,
+      "eval_runtime": 18.2441,
+      "eval_samples_per_second": 54.812,
+      "eval_steps_per_second": 13.703,
+      "step": 26750
+    },
+    {
+      "epoch": 2.061379893854319,
+      "grad_norm": 1.087234377861023,
+      "learning_rate": 6.345999013063917e-05,
+      "loss": 1.3967,
+      "step": 26800
+    },
+    {
+      "epoch": 2.065225751865241,
+      "grad_norm": 1.638533592224121,
+      "learning_rate": 6.32002701088227e-05,
+      "loss": 1.4558,
+      "step": 26850
+    },
+    {
+      "epoch": 2.0690716098761635,
+      "grad_norm": 1.4552900791168213,
+      "learning_rate": 6.294055008700621e-05,
+      "loss": 1.4553,
+      "step": 26900
+    },
+    {
+      "epoch": 2.0729174678870854,
+      "grad_norm": 1.6330054998397827,
+      "learning_rate": 6.268083006518974e-05,
+      "loss": 1.4977,
+      "step": 26950
+    },
+    {
+      "epoch": 2.076763325898008,
+      "grad_norm": 1.497938632965088,
+      "learning_rate": 6.242111004337325e-05,
+      "loss": 1.4449,
+      "step": 27000
+    },
+    {
+      "epoch": 2.076763325898008,
+      "eval_loss": 1.4610888957977295,
+      "eval_runtime": 17.9657,
+      "eval_samples_per_second": 55.662,
+      "eval_steps_per_second": 13.915,
+      "step": 27000
     }
   ],
   "logging_steps": 50,