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

 hers

-0.69

 lifes

-0.64

 reserv

-0.64

lance

-0.63

ktop

-0.62

 awareness

-0.61

skilled

-0.61

 removal

-0.61

 jaws

-0.59

thodox

-0.59

POSITIVE LOGITS

ween

0.73

Date

0.68

orio

0.64

angelo

0.63

Ļ

0.63

hua

0.63

Ĺ

0.63

nan

0.63

¤

0.62

Adams

0.62

Activations Density 0.000%

No Known Activations

This feature has no known activations.