INDEX

Explanations

the concept of being willing or ready to do something

oai_token-act-pair · gpt-3.5-turbo

expressions of readiness or willingness to take action

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

 Anthem

-0.89

isol

-0.72

adish

-0.69

ARS

-0.69

riot

-0.68

rave

-0.68

alien

-0.66

arette

-0.66

gran

-0.65

our

-0.65

POSITIVE LOGITS

theless

0.84

 willing

0.82

 enough

0.79

 gladly

0.75

 unwilling

0.75

 unres

0.73

terday

0.70

ledged

0.70

 incent

0.69

 lend

0.69

Activations Density 0.020%

the concept of being willing or ready to do something

expressions of readiness or willingness to take action

No Comments

No Known Activations