INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

abet

-0.78

rats

-0.74

fired

-0.72

rations

-0.71

pes

-0.69

carry

-0.69

rans

-0.69

vernment

-0.68

unal

-0.66

Reviewer

-0.64

POSITIVE LOGITS

ufact

0.70

 Noir

0.67

 Archdemon

0.66

ctor

0.66

eson

0.64

)=(

0.62

 Catalyst

0.61

isbury

0.61

 molecule

0.61

 Pioneer

0.61

Activations Density 0.000%

No Known Activations

This feature has no known activations.