Training in progress, step 7000, checkpoint

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2e4fa545226f8a1bc971fe1a1680b858ab3cba84884085de4d7bb95fce3027ee
+oid sha256:52b997298d45a4fc6cd35ecfe5ccf8a749558ac28f75afb2d0909fe1d125004b
 size 36730224

last-checkpoint/ar_diffusion_info.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b20c2693c849c11648664638444158f688e791753d1fd5b0a3839f3bf439279b
+oid sha256:f95ce72a1c3a7c4d5aab7200d9dec95f462be645783ddec5e51a5155488e47cc
 size 1736

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9b38307c83b8af70cb28809299297b128e7799c2bcec4e16e0bf925a16820657
+oid sha256:215f3ea94412e8b86475daca81b7346b3039dd518294dd8628b764b4fe24e130
 size 73588346

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e3dc41260bd3c4c88915ea4a89c036f2cf6425407a322ee64551935a201e505c
-size 14244
+oid sha256:517f7f125f44f7edc89c0d28a191b1cd6fb9d0b94d17a4ddcd53f5fcaa8aefcd
+size 14180

last-checkpoint/scaler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:76773cd6ef78ee26a2f9b89a5cdc92391f18eaca0e15a3c1b947462d6c8a47c1
+oid sha256:7df83cd74eb098f609a6963afc6cf853a656c30456c390df44a7c43c2165d798
 size 988

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:63a3451545c9d3c80af7346ccecc86c947b582a85ff2768518fe0c4acdf7dcd7
+oid sha256:93f284fcd12d6e29e5cbdf1e4a0021a53156461eb5dba25291a9ab08dfb3c1a8
 size 1064

last-checkpoint/trainer_state.json

@@ -1,10 +1,10 @@
 {
-  "best_global_step": 6000,
-  "best_metric": 1.4524279832839966,
-  "best_model_checkpoint": "./ar-diffusion-checkpoints-progressive-attention/checkpoint-6000",
-  "epoch": 0.4999615414198908,
+  "best_global_step": 7000,
+  "best_metric": 1.4396251440048218,
+  "best_model_checkpoint": "./ar-diffusion-checkpoints-progressive-attention/checkpoint-7000",
+  "epoch": 0.5384201215291131,
   "eval_steps": 250,
-  "global_step": 6500,
+  "global_step": 7000,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -1126,6 +1126,92 @@
       "eval_samples_per_second": 59.703,
       "eval_steps_per_second": 14.926,
       "step": 6500
+    },
+    {
+      "epoch": 0.503807399430813,
+      "grad_norm": 1.6324172019958496,
+      "learning_rate": 0.00016859984936238735,
+      "loss": 1.5181,
+      "step": 6550
+    },
+    {
+      "epoch": 0.5076532574417353,
+      "grad_norm": 2.2709999084472656,
+      "learning_rate": 0.00016834012934057086,
+      "loss": 1.4517,
+      "step": 6600
+    },
+    {
+      "epoch": 0.5114991154526575,
+      "grad_norm": 1.0540615320205688,
+      "learning_rate": 0.0001680804093187544,
+      "loss": 1.4737,
+      "step": 6650
+    },
+    {
+      "epoch": 0.5153449734635798,
+      "grad_norm": 1.497239589691162,
+      "learning_rate": 0.00016782068929693792,
+      "loss": 1.4038,
+      "step": 6700
+    },
+    {
+      "epoch": 0.519190831474502,
+      "grad_norm": 1.5478484630584717,
+      "learning_rate": 0.00016756096927512143,
+      "loss": 1.5019,
+      "step": 6750
+    },
+    {
+      "epoch": 0.519190831474502,
+      "eval_loss": 1.446391224861145,
+      "eval_runtime": 16.8309,
+      "eval_samples_per_second": 59.415,
+      "eval_steps_per_second": 14.854,
+      "step": 6750
+    },
+    {
+      "epoch": 0.5230366894854241,
+      "grad_norm": 1.1217519044876099,
+      "learning_rate": 0.00016730124925330494,
+      "loss": 1.3677,
+      "step": 6800
+    },
+    {
+      "epoch": 0.5268825474963464,
+      "grad_norm": 1.8287678956985474,
+      "learning_rate": 0.00016704152923148846,
+      "loss": 1.4758,
+      "step": 6850
+    },
+    {
+      "epoch": 0.5307284055072686,
+      "grad_norm": 2.8644931316375732,
+      "learning_rate": 0.000166781809209672,
+      "loss": 1.4328,
+      "step": 6900
+    },
+    {
+      "epoch": 0.5345742635181909,
+      "grad_norm": 0.8893502354621887,
+      "learning_rate": 0.0001665220891878555,
+      "loss": 1.4648,
+      "step": 6950
+    },
+    {
+      "epoch": 0.5384201215291131,
+      "grad_norm": 1.1741907596588135,
+      "learning_rate": 0.00016626236916603902,
+      "loss": 1.4338,
+      "step": 7000
+    },
+    {
+      "epoch": 0.5384201215291131,
+      "eval_loss": 1.4396251440048218,
+      "eval_runtime": 16.8849,
+      "eval_samples_per_second": 59.225,
+      "eval_steps_per_second": 14.806,
+      "step": 7000
     }
   ],
   "logging_steps": 50,