INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

Benz

-0.86

Mech

-0.76

Balt

-0.74

Bloom

-0.73

thro

-0.73

Merit

-0.72

Bas

-0.71

GY

-0.70

Dialogue

-0.69

Ont

-0.68

POSITIVE LOGITS

hammad

0.76

ldon

0.71

 ransom

0.70

igger

0.69

owa

0.66

olation

0.65

urious

0.65

yne

0.64

reau

0.64

horn

0.63

Activations Density 0.000%

No Known Activations

This feature has no known activations.