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

pson

-0.76

icative

-0.71

iox

-0.68

Pax

-0.66

swick

-0.64

rical

-0.63

iq

-0.63

 Okin

-0.63

itarian

-0.63

illance

-0.62

POSITIVE LOGITS

abilities

0.86

artifacts

0.86

untarily

0.76

 Parables

0.71

uca

0.69

resso

0.65

better

0.63

illard

0.62

ascal

0.62

 preferably

0.61

Activations Density 0.000%

No Known Activations

This feature has no known activations.