INDEX

Explanations

phrases where someone is explaining or arguing a point

oai_token-act-pair · gpt-3.5-turbo

verbs used to attribute statements or arguments to individuals

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

inary

-0.76

spir

-0.71

acts

-0.71

empt

-0.66

eneg

-0.66

rats

-0.66

plug

-0.65

icates

-0.64

rel

-0.64

 impe

-0.63

POSITIVE LOGITS

ï¸ı

0.85

 Polly

0.83

 David

0.82

 Jonathan

0.82

 Joel

0.80

 Laura

0.79

 Joyce

0.78

 Pamela

0.77

 Lynn

0.76

 Eliot

0.76

Activations Density 0.046%

phrases where someone is explaining or arguing a point

verbs used to attribute statements or arguments to individuals

No Comments

No Known Activations