INDEX

Explanations

adverbs that modify the activity level of an action or event

oai_token-act-pair · gpt-3.5-turbo

adjectives that describe the extent or degree of something

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

ilater

-0.90

iday

-0.72

bucks

-0.65

 bowling

-0.64

Reviewer

-0.64

okÃ©

-0.63

 payday

-0.62

eport

-0.62

 shutter

-0.61

 hatch

-0.61

POSITIVE LOGITS

 Sabha

0.79

 Helpful

0.71

forth

0.71

hea

0.70

[+

0.70

oeuv

0.68

 istg

0.66

forward

0.65

inguishable

0.65

zer

0.65

Activations Density 0.043%

adverbs that modify the activity level of an action or event

adjectives that describe the extent or degree of something

No Comments

No Known Activations