Training in progress, step 20000, checkpoint

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:879ead83bd2c1fe46b7e68211ea0caa2723b2414cf5d08c8c7b5b29823750c25
+oid sha256:bff1d47cdb13e83e8fe6a9d20b6c173dfae48b029aecf5c92bf458772133f5b2
 size 36730224

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8a6fddc34b7b7d3235b23dac04ebf76b113434f948440bce4f7bae9c4f95c22a
+oid sha256:c26f8a7499e23aa16477b380bf134293878cb8936b98f8cad1d8a606fc561e1a
 size 73588346

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:aa7c19bbb017976e63ac5e9fec3eea0eda0cf5868e0275e6b00fde078e2c850d
+oid sha256:b205d8106465470d82a9f668e113b1a4c937f3fe768b385e78f85eb171e49cc2
 size 14244

last-checkpoint/scaler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:abdb0a20afab4dc0c80b1c5f76a237a0247272f8d00055102fe692e4d58c1f05
+oid sha256:344e181f9677a4d18097c7085524ad711f8dfb151d42735faf9658fba6102bed
 size 988

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:488aaa88c1afe9c9c3279271cc45e297a0d20ece7902b965486e19f7d9811602
+oid sha256:6fc471d6cffa6d4b3db986d1e54c37de8f3b7f607849783a6e6b32bce97519eb
 size 1064

last-checkpoint/trainer_state.json

@@ -2,9 +2,9 @@
   "best_global_step": 18750,
   "best_metric": 1.4680087566375732,
   "best_model_checkpoint": "./ar-diffusion-checkpoints/checkpoint-15500",
-  "epoch": 1.4998846242596724,
+  "epoch": 1.5383432043688947,
   "eval_steps": 250,
-  "global_step": 19500,
+  "global_step": 20000,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -3362,6 +3362,92 @@
       "eval_samples_per_second": 55.491,
       "eval_steps_per_second": 13.873,
       "step": 19500
+    },
+    {
+      "epoch": 1.5037304822705946,
+      "grad_norm": 0.8779070377349854,
+      "learning_rate": 3.995000183816771e-05,
+      "loss": 1.3728,
+      "step": 19550
+    },
+    {
+      "epoch": 1.5075763402815168,
+      "grad_norm": 1.6068123579025269,
+      "learning_rate": 3.9692658358148595e-05,
+      "loss": 1.5204,
+      "step": 19600
+    },
+    {
+      "epoch": 1.511422198292439,
+      "grad_norm": 1.7712832689285278,
+      "learning_rate": 3.9435314878129475e-05,
+      "loss": 1.514,
+      "step": 19650
+    },
+    {
+      "epoch": 1.5152680563033614,
+      "grad_norm": 1.2519572973251343,
+      "learning_rate": 3.917797139811036e-05,
+      "loss": 1.3953,
+      "step": 19700
+    },
+    {
+      "epoch": 1.5191139143142836,
+      "grad_norm": 1.5644786357879639,
+      "learning_rate": 3.892062791809125e-05,
+      "loss": 1.4772,
+      "step": 19750
+    },
+    {
+      "epoch": 1.5191139143142836,
+      "eval_loss": 1.4710900783538818,
+      "eval_runtime": 18.4205,
+      "eval_samples_per_second": 54.287,
+      "eval_steps_per_second": 13.572,
+      "step": 19750
+    },
+    {
+      "epoch": 1.5229597723252057,
+      "grad_norm": 1.6755670309066772,
+      "learning_rate": 3.866328443807212e-05,
+      "loss": 1.4148,
+      "step": 19800
+    },
+    {
+      "epoch": 1.5268056303361282,
+      "grad_norm": 1.7168843746185303,
+      "learning_rate": 3.840594095805301e-05,
+      "loss": 1.4211,
+      "step": 19850
+    },
+    {
+      "epoch": 1.5306514883470501,
+      "grad_norm": 1.5205817222595215,
+      "learning_rate": 3.8148597478033894e-05,
+      "loss": 1.4663,
+      "step": 19900
+    },
+    {
+      "epoch": 1.5344973463579725,
+      "grad_norm": 1.608231544494629,
+      "learning_rate": 3.789125399801477e-05,
+      "loss": 1.3634,
+      "step": 19950
+    },
+    {
+      "epoch": 1.5383432043688947,
+      "grad_norm": 1.5260729789733887,
+      "learning_rate": 3.763391051799566e-05,
+      "loss": 1.4114,
+      "step": 20000
+    },
+    {
+      "epoch": 1.5383432043688947,
+      "eval_loss": 1.4733539819717407,
+      "eval_runtime": 18.105,
+      "eval_samples_per_second": 55.233,
+      "eval_steps_per_second": 13.808,
+      "step": 20000
     }
   ],
   "logging_steps": 50,