Training in progress, step 12500, checkpoint

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:da26990df234d94448e406f402bbaab14595948f6b2bb6aaf6233e11502b39c8
+oid sha256:6a623f2e366d56a653f11e3f01819889dc1645d4d1136c5b510bf0c1c92dab66
 size 36730224

last-checkpoint/ar_diffusion_info.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4fd8f2ea0f8f3f2b6317bf012fe6c554f0ab9b95b3c3bc864dd5cae3b6eef992
+oid sha256:6015f4fec963cdcdbd9534ce342304afb8094ad5f761558a674ddf3068ab0335
 size 1736

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:70a89db0f278efb20b8c741665f32f39c512b7877cd595dd7f15cfb9e97916f3
+oid sha256:8f314be3a1ba2ce8f5f725cf645fa00681145b061c07cf8ad59e1e13356924cb
 size 73588346

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1696a7b978a5af1f9de4c1c4ce7f56376b2325426ded966dd6d734962f026c10
-size 14244
+oid sha256:082b149e63f0f47e9e60d37040a9d015e2fa6a3d80388f6fa7d859aa3870fd21
+size 14308

last-checkpoint/scaler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7f076f7a63b00dbdf4d7a2eb7609297124339e4b64abf631b5119fa735b47e6a
+oid sha256:0678e6da71fe39cdc83977acb14900d45184a734b31b0b8a20ead5262367273c
 size 988

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b7a1d25be56e05cb4cbafc0981ade0df33b24e8c1de440cdbae85f1fea41cdfa
+oid sha256:a7680f728f95fcd934cd32ef342f02b8f2b507472a27e9e8de734cc5468e1428
 size 1064

last-checkpoint/trainer_state.json

@@ -1,10 +1,10 @@
 {
-  "best_global_step": 12000,
-  "best_metric": -30.337785720825195,
-  "best_model_checkpoint": "./ar-diffusion-checkpoints-progressive-attention/checkpoint-12000",
-  "epoch": 0.9230059226213368,
+  "best_global_step": 12500,
+  "best_metric": -30.455406188964844,
+  "best_model_checkpoint": "./ar-diffusion-checkpoints-progressive-attention/checkpoint-12500",
+  "epoch": 0.9614645027305592,
   "eval_steps": 250,
-  "global_step": 12000,
+  "global_step": 12500,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -2072,6 +2072,92 @@
       "eval_samples_per_second": 61.247,
       "eval_steps_per_second": 15.312,
       "step": 12000
+    },
+    {
+      "epoch": 0.926851780632259,
+      "grad_norm": 1.3591136932373047,
+      "learning_rate": 0.00014004623016388335,
+      "loss": -30.3233,
+      "step": 12050
+    },
+    {
+      "epoch": 0.9306976386431813,
+      "grad_norm": 1.8113991022109985,
+      "learning_rate": 0.00013978651014206686,
+      "loss": -30.3411,
+      "step": 12100
+    },
+    {
+      "epoch": 0.9345434966541035,
+      "grad_norm": 2.9664206504821777,
+      "learning_rate": 0.00013952679012025037,
+      "loss": -30.3763,
+      "step": 12150
+    },
+    {
+      "epoch": 0.9383893546650258,
+      "grad_norm": 4.572649955749512,
+      "learning_rate": 0.0001392670700984339,
+      "loss": -30.3894,
+      "step": 12200
+    },
+    {
+      "epoch": 0.942235212675948,
+      "grad_norm": 0.7282872200012207,
+      "learning_rate": 0.0001390073500766174,
+      "loss": -30.3943,
+      "step": 12250
+    },
+    {
+      "epoch": 0.942235212675948,
+      "eval_loss": -30.41870880126953,
+      "eval_runtime": 16.3228,
+      "eval_samples_per_second": 61.264,
+      "eval_steps_per_second": 15.316,
+      "step": 12250
+    },
+    {
+      "epoch": 0.9460810706868702,
+      "grad_norm": 0.5192355513572693,
+      "learning_rate": 0.00013874763005480094,
+      "loss": -30.4192,
+      "step": 12300
+    },
+    {
+      "epoch": 0.9499269286977925,
+      "grad_norm": 0.4913904070854187,
+      "learning_rate": 0.00013848791003298445,
+      "loss": -30.4326,
+      "step": 12350
+    },
+    {
+      "epoch": 0.9537727867087147,
+      "grad_norm": 0.36102983355522156,
+      "learning_rate": 0.00013822819001116797,
+      "loss": -30.4425,
+      "step": 12400
+    },
+    {
+      "epoch": 0.957618644719637,
+      "grad_norm": 0.1661052107810974,
+      "learning_rate": 0.00013796846998935148,
+      "loss": -30.4485,
+      "step": 12450
+    },
+    {
+      "epoch": 0.9614645027305592,
+      "grad_norm": 0.43040284514427185,
+      "learning_rate": 0.000137708749967535,
+      "loss": -30.4529,
+      "step": 12500
+    },
+    {
+      "epoch": 0.9614645027305592,
+      "eval_loss": -30.455406188964844,
+      "eval_runtime": 16.2806,
+      "eval_samples_per_second": 61.423,
+      "eval_steps_per_second": 15.356,
+      "step": 12500
     }
   ],
   "logging_steps": 50,