INDEX

Explanations

mentions of an audience in various contexts

oai_token-act-pair · gpt-3.5-turbo

mentions of an audience

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

omes

-0.73

ced

-0.71

erald

-0.68

ces

-0.68

empt

-0.68

pleting

-0.66

omic

-0.65

jong

-0.64

omon

-0.62

cies

-0.62

POSITIVE LOGITS

 audience

1.24

 audiences

1.04

ÃįÃį

0.81

eers

0.79

 viewership

0.77

 attendees

0.75

 interviewer

0.73

wat

0.73

goers

0.73

 listener

0.72

Activations Density 0.012%

mentions of an audience in various contexts

mentions of an audience

No Comments

No Known Activations