INDEX

Explanations

the word "Alpha" with varying degrees of emphasis (activation levels)

oai_token-act-pair · gpt-3.5-turbo

mentions of different types of alpha particles or compounds

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

enegger

-0.97

Ö¼

-0.96

puter

-0.77

alities

-0.75

nee

-0.75

SIGN

-0.74

tons

-0.74

ROR

-0.72

die

-0.72

alling

-0.72

POSITIVE LOGITS

 Centauri

1.22

 male

0.75

velop

0.75

 males

0.71

 Male

0.71

Draft

0.70

 Males

0.68

 essence

0.68

 channel

0.66

imity

0.66

Activations Density 0.011%

the word "Alpha" with varying degrees of emphasis (activation levels)

mentions of different types of alpha particles or compounds

No Comments

No Known Activations