INDEX

Explanations

verbs indicating actions or behaviors performed by individuals

oai_token-act-pair · gpt-3.5-turbo

frequent verbs and phrases indicating actions taken by individuals

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

 Coach

-0.67

 Mages

-0.64

 Almighty

-0.64

ascript

-0.62

Ü

-0.61

`,

-0.58

 thous

-0.57

ogun

-0.56

 Sounds

-0.56

 Guys

-0.56

POSITIVE LOGITS

iw

0.75

conom

0.75

rake

0.70

ashtra

0.67

MpServer

0.66

zon

0.66

lect

0.65

 outright

0.65

Reviewed

0.65

hurst

0.64

Activations Density 0.314%

verbs indicating actions or behaviors performed by individuals

frequent verbs and phrases indicating actions taken by individuals

No Comments

No Known Activations