INDEX

Explanations

phrases related to different types of professions or roles

oai_token-act-pair · gpt-3.5-turbo

phrases that discuss the concept of "being" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

itures

-0.74

rosso

-0.68

estyles

-0.67

ighting

-0.66

oday

-0.66

 reperto

-0.66

tle

-0.65

ongyang

-0.64

 interchange

-0.62

 contrasts

-0.61

POSITIVE LOGITS

 unemployed

1.03

 born

0.93

 addicted

0.89

 employed

0.85

 circumcised

0.85

 uninsured

0.85

 bitten

0.85

 alive

0.84

 involved

0.84

 atheist

0.83

Activations Density 0.275%

phrases related to different types of professions or roles

phrases that discuss the concept of "being" in various contexts

No Comments

No Known Activations