INDEX

Explanations

mentions of radio talk shows or programs

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "talk" in various contexts

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

aples

-0.72

served

-0.69

uilt

-0.68

boa

-0.66

 Stores

-0.62

Constructed

-0.61

AppData

-0.60

 crest

-0.60

arching

-0.60

 landfall

-0.60

POSITIVE LOGITS

ĳå£«

0.78

cloth

0.78

athon

0.77

 Talk

0.77

ies

0.76

bag

0.75

 talk

0.74

axe

0.73

ers

0.73

hran

0.72

Activations Density 0.025%

mentions of radio talk shows or programs

occurrences of the word "talk" in various contexts

No Comments

No Known Activations