Eviation commited on
Commit
9cb7abe
·
verified ·
1 Parent(s): 83a97c4

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +1 -0
README.md CHANGED
@@ -54,6 +54,7 @@ Using [llama.cpp quantize cae9fb4](https://github.com/ggerganov/llama.cpp/commit
54
  - Bravo IQ1_S worse than Alpha?
55
  - [Latent loss](https://huggingface.co/Eviation/flux-imatrix/blob/main/images/latent_loss.png)
56
  - [Per layer quantization cost](https://huggingface.co/Eviation/flux-imatrix/blob/main/images/casting_cost.png) from [chrisgoringe/casting_cost](https://github.com/chrisgoringe/mixed-gguf-converter/blob/main/costs/casting_cost.yaml)
 
57
 
58
  ### Per weight type quantization loss in increasing order:
59
  - double_blocks.%d.txt_attn.proj.weight
 
54
  - Bravo IQ1_S worse than Alpha?
55
  - [Latent loss](https://huggingface.co/Eviation/flux-imatrix/blob/main/images/latent_loss.png)
56
  - [Per layer quantization cost](https://huggingface.co/Eviation/flux-imatrix/blob/main/images/casting_cost.png) from [chrisgoringe/casting_cost](https://github.com/chrisgoringe/mixed-gguf-converter/blob/main/costs/casting_cost.yaml)
57
+ - [Ablation latent loss per weight type](https://huggingface.co/Eviation/flux-imatrix/blob/main/images/latent_loss_ablation.png)
58
 
59
  ### Per weight type quantization loss in increasing order:
60
  - double_blocks.%d.txt_attn.proj.weight