Training in progress, step 8500, checkpoint

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:eb24e75cd0d9c0241c7a377dfdee869bf75df5f8323d7de323c20a4ab9611b72
+oid sha256:267a0aeb8560d88ad88517cd9fd8662ce7d595564aa3e0f53a4ad59438882fb9
 size 36730224

last-checkpoint/ar_diffusion_info.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d03f32b278295513a310a1b0d9bff7cb144799f247621052afb62359d974a28f
+oid sha256:727b6b299bbefe56d0c7a86c1d32ed70935812db806c55a2d94fdc69eac969c5
 size 1544

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d903a7de961e36fa691b8642db3b6862344a9f993ee30fb71db94133105da9e1
+oid sha256:9b3d8571245b57c1124757a0245973277c23134608ddef1e6cb62706c349989a
 size 73588346

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0d6bc30a519d7cae432bede0d2778a805102f7cbeae9a244275827f72067499b
-size 14180
+oid sha256:a993eb83d217882b8d4270d606bf3996e74fd8745c32e817e4ed4a0c36ae01a8
+size 14244

last-checkpoint/scaler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:490750601169a2038998363f62664f342c96cc7ab2cf792ba7560d9ea46de685
+oid sha256:273b848f717f8541d5af997621dc6b78ddd5d9517d19f49a443ead9755008eb4
 size 988

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:eac25a0c04cc4fd8619bcf4c75b6e95131c92ecfe30295ff7072dab6e9fc2f51
+oid sha256:69b1fdee44d912af93d7f5915eae6dc9e82d40f7e3d50091f8df864d7a3e6f30
 size 1064

last-checkpoint/trainer_state.json

@@ -1,10 +1,10 @@
 {
-  "best_global_step": 7250,
-  "best_metric": 4.42822265625,
-  "best_model_checkpoint": "./ar-diffusion-checkpoints-fixed/checkpoint-7000",
-  "epoch": 0.6153372817475579,
+  "best_global_step": 8500,
+  "best_metric": 4.409055709838867,
+  "best_model_checkpoint": "./ar-diffusion-checkpoints-fixed/checkpoint-8500",
+  "epoch": 0.6537958618567803,
   "eval_steps": 250,
-  "global_step": 8000,
+  "global_step": 8500,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -1384,6 +1384,92 @@
       "eval_samples_per_second": 53.833,
       "eval_steps_per_second": 13.458,
       "step": 8000
+    },
+    {
+      "epoch": 0.6191831397584802,
+      "grad_norm": 4.5992255210876465,
+      "learning_rate": 0.00016083422071007454,
+      "loss": 4.3146,
+      "step": 8050
+    },
+    {
+      "epoch": 0.6230289977694023,
+      "grad_norm": 5.412031650543213,
+      "learning_rate": 0.00016057450068825805,
+      "loss": 4.5841,
+      "step": 8100
+    },
+    {
+      "epoch": 0.6268748557803245,
+      "grad_norm": 4.4531779289245605,
+      "learning_rate": 0.0001603147806664416,
+      "loss": 4.5084,
+      "step": 8150
+    },
+    {
+      "epoch": 0.6307207137912468,
+      "grad_norm": 3.913174867630005,
+      "learning_rate": 0.0001600550606446251,
+      "loss": 4.4183,
+      "step": 8200
+    },
+    {
+      "epoch": 0.634566571802169,
+      "grad_norm": 3.3952407836914062,
+      "learning_rate": 0.00015979534062280862,
+      "loss": 4.3496,
+      "step": 8250
+    },
+    {
+      "epoch": 0.634566571802169,
+      "eval_loss": 4.46174430847168,
+      "eval_runtime": 18.5381,
+      "eval_samples_per_second": 53.943,
+      "eval_steps_per_second": 13.486,
+      "step": 8250
+    },
+    {
+      "epoch": 0.6384124298130913,
+      "grad_norm": 4.156221389770508,
+      "learning_rate": 0.00015953562060099213,
+      "loss": 4.361,
+      "step": 8300
+    },
+    {
+      "epoch": 0.6422582878240135,
+      "grad_norm": 3.2021920680999756,
+      "learning_rate": 0.00015927590057917564,
+      "loss": 4.473,
+      "step": 8350
+    },
+    {
+      "epoch": 0.6461041458349358,
+      "grad_norm": 5.048036575317383,
+      "learning_rate": 0.00015901618055735918,
+      "loss": 4.4142,
+      "step": 8400
+    },
+    {
+      "epoch": 0.649950003845858,
+      "grad_norm": 3.91768217086792,
+      "learning_rate": 0.0001587564605355427,
+      "loss": 4.4672,
+      "step": 8450
+    },
+    {
+      "epoch": 0.6537958618567803,
+      "grad_norm": 9.229452133178711,
+      "learning_rate": 0.0001584967405137262,
+      "loss": 4.7837,
+      "step": 8500
+    },
+    {
+      "epoch": 0.6537958618567803,
+      "eval_loss": 4.409055709838867,
+      "eval_runtime": 18.5943,
+      "eval_samples_per_second": 53.78,
+      "eval_steps_per_second": 13.445,
+      "step": 8500
     }
   ],
   "logging_steps": 50,