INDEX

Explanations

phrases where the speaker is stating intentions or plans

oai_token-act-pair · gpt-3.5-turbo

statements of intention or commitment

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

 Farms

-0.65

 Kitt

-0.65

Args

-0.62

artney

-0.62

ERE

-0.61

Gap

-0.61

 Byrne

-0.58

 Rossi

-0.58

bies

-0.56

lif

-0.56

POSITIVE LOGITS

 myself

1.15

 paraph

0.75

 recommend

0.74

xtap

0.73

 glad

0.72

 personally

0.72

 fortunate

0.72

 summarize

0.72

 sugg

0.71

anmar

0.70

Activations Density 0.422%

phrases where the speaker is stating intentions or plans

statements of intention or commitment

No Comments

No Known Activations