INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.2.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 sovere

-0.75

STEM

-0.74

 tradem

-0.73

Fra

-0.72

Cube

-0.70

 Nanto

-0.69

ricanes

-0.66

 Encyclopedia

-0.66

uca

-0.65

omsky

-0.65

POSITIVE LOGITS

omething

0.67

 Yards

0.66

 processed

0.66

>>>>

0.62

 flask

0.61

 Donation

0.61

 warrants

0.59

heet

0.59

heny

0.59

eries

0.59

Activations Density 0.000%

No Known Activations

This feature has no known activations.