INDEX

Explanations

mentions of a specific television network

oai_token-act-pair · gpt-3.5-turbo

references to television networks

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

sql

-0.75

tml

-0.72

matter

-0.70

potion

-0.69

istry

-0.69

umatic

-0.68

arcity

-0.67

ravel

-0.67

vironment

-0.66

 Flask

-0.66

POSITIVE LOGITS

 television

0.97

 Broadcasting

0.94

TV

0.92

tv

0.92

 Networks

0.91

 telev

0.91

 sitcom

0.89

 broadcaster

0.87

 broadcasts

0.86

 airing

0.86

Activations Density 0.034%

mentions of a specific television network

references to television networks

No Comments

No Known Activations