INDEX

Explanations

references to audience or viewers

oai_token-act-pair · gpt-3.5-turbo

references to the audience or viewers

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

mine

-0.81

orie

-0.70

agna

-0.66

 reorgan

-0.65

Ram

-0.60

redo

-0.59

 Davidson

-0.59

death

-0.58

red

-0.58

 Hurricanes

-0.58

POSITIVE LOGITS

 viewers

1.38

 viewership

1.11

 viewer

1.03

wat

0.99

ÃįÃį

0.96

 audiences

0.91

 tuned

0.88

 audience

0.85

 listeners

0.82

erning

0.79

Activations Density 0.006%

references to audience or viewers

references to the audience or viewers

No Comments

No Known Activations