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

Kut

-0.73

 Ster

-0.73

Vs

-0.72

 Rove

-0.72

Rye

-0.71

abad

-0.70

hod

-0.70

Coy

-0.69

uve

-0.68

 Evolution

-0.66

POSITIVE LOGITS

erate

1.11

istries

0.81

cknow

0.80

LLOW

0.77

ccording

0.72

nels

0.69

etitive

0.69

ears

0.66

pletion

0.64

pace

0.64

Activations Density 0.000%

No Known Activations

This feature has no known activations.