Trained for 0 epochs and 6000 steps.

Trained with datasets ['text-embeds-sd3-nofilter', 'photo-concept-bucket', 'dalle3']
Learning rate 1e-06, batch size 6, and 2 gradient accumulation steps.
Used DDPM noise scheduler for training with epsilon prediction type and rescaled_betas_zero_snr=False
Using 'trailing' timestep spacing.
Base model: stabilityai/stable-diffusion-3-medium-diffusers
VAE: None

README.md

@@ -1164,7 +1164,7 @@ You may reuse the base model text encoder for inference.
 ## Training settings
 
 - Training epochs: 0
-- Training steps: 5500
+- Training steps: 6000
 - Learning rate: 1e-06
 - Effective batch size: 96
   - Micro-batch size: 6

optimizer.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a4ade6957115be9048cb6099f15af8d109f8769e8a053ddf81a6f16477cd3f6b
+oid sha256:06e7b3bd3c2e41169a6a32ca435e94825f5121ec67eb3e792f1d1786810b9bf6
 size 12170595712

random_states_0.pkl

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d566e64750d1ba8193a83c0eb0c4508e61c56ac58e70954cb07455e2f55fe56e
-size 16100
+oid sha256:f49480729ff3ab718b4a2a8d9256aaa1ce5882bd8f387a6d421b3cd262e24f71
+size 16036

scheduler.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:dd5751e81810f4414c5432dd8d91eb6503635529d6526f0d5acee0007aa8a967
+oid sha256:7f30c9fbfa479a9edb8fa992b83695b0ce1122eace54970a7b35dbcc7f5d4a75
 size 1128

training_state.json

@@ -1 +1 @@
-{"global_step": 5500, "epoch_step": 5500, "epoch": 1, "exhausted_backends": [], "repeats": {}}
+{"global_step": 6000, "epoch_step": 6000, "epoch": 1, "exhausted_backends": [], "repeats": {}}

transformer/diffusion_pytorch_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9910bbab644f610dd6a0320af2026ce2da2b9b3dbdb2acfc1fa264bf2ed39ec5
+oid sha256:c3332da85b0460b065149f8212f7b56da455a1a9981f087f31ed53a48ca97dfb
 size 4169982088