codemichaeld commited on
Commit
640b3c2
·
verified ·
1 Parent(s): 3e304e1

Upload README.md with huggingface_hub

Browse files
Files changed (1) hide show
  1. README.md +4 -4
README.md CHANGED
@@ -11,9 +11,9 @@ tags:
11
  - **Source**: `https://huggingface.co/LifuWang/DistillT5`
12
  - **File**: `model.safetensors`
13
  - **FP8 Format**: `E5M2`
14
- - **Architecture**: text_encoder
15
  - **Precision Recovery Type**: LoRA
16
- - **Precision Recovery File**: `model-lora-r64-text_encoder.safetensors` if available
17
  - **FP8 File**: `model-fp8-e5m2.safetensors`
18
 
19
  ## Usage (Inference)
@@ -26,8 +26,8 @@ fp8_state = load_file("model-fp8-e5m2.safetensors")
26
 
27
  # Load precision recovery file if available
28
  recovery_state = {}
29
- if "model-lora-r64-text_encoder.safetensors":
30
- recovery_state = load_file("model-lora-r64-text_encoder.safetensors")
31
 
32
  # Reconstruct high-precision weights
33
  reconstructed = {}
 
11
  - **Source**: `https://huggingface.co/LifuWang/DistillT5`
12
  - **File**: `model.safetensors`
13
  - **FP8 Format**: `E5M2`
14
+ - **Architecture**: all
15
  - **Precision Recovery Type**: LoRA
16
+ - **Precision Recovery File**: `model-lora-r64-all.safetensors` if available
17
  - **FP8 File**: `model-fp8-e5m2.safetensors`
18
 
19
  ## Usage (Inference)
 
26
 
27
  # Load precision recovery file if available
28
  recovery_state = {}
29
+ if "model-lora-r64-all.safetensors":
30
+ recovery_state = load_file("model-lora-r64-all.safetensors")
31
 
32
  # Reconstruct high-precision weights
33
  reconstructed = {}