INDEX

Explanations

terms related to broadcasting and media networks

oai_token-act-pair · gpt-3.5-turbo

the term "Broad" and its related variations in the context of broadcasting and media

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

dj

-0.74

pez

-0.69

 undo

-0.68

mig

-0.66

 minus

-0.66

 saving

-0.63

 stains

-0.63

pg

-0.63

gie

-0.62

Jah

-0.62

POSITIVE LOGITS

Broad

3.75

 Broad

3.06

broad

2.72

OAD

1.25

 Broadcast

1.17

 broad

1.09

ighton

1.01

 Fuller

1.01

Brook

1.00

Wid

0.98

Activations Density 0.014%

terms related to broadcasting and media networks

the term "Broad" and its related variations in the context of broadcasting and media

No Comments

No Known Activations