INDEX

Explanations

specific words related to diverse areas like sports, news, and genetics

oai_token-act-pair · gpt-3.5-turbo

specific nouns and action-oriented verbs related to various topics

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

 respectively

-0.71

Ïī

-0.67

 ..."

-0.65

 âĢ¦"

-0.62

 thereof

-0.62

 thereto

-0.61

 otherwise

-0.61

 trans

-0.60

.","

-0.58

..."

-0.57

POSITIVE LOGITS

resa

1.10

ogether

0.91

mosp

0.84

vertising

0.83

icularly

0.83

tymology

0.81

odore

0.81

roximately

0.80

autions

0.79

xiety

0.78

Activations Density 0.532%

specific words related to diverse areas like sports, news, and genetics

specific nouns and action-oriented verbs related to various topics

No Comments

No Known Activations