INDEX

Explanations

words related to sickness at varying levels of severity

oai_token-act-pair · gpt-3.5-turbo

the word "sick" 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 @ 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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

RFC

-0.69

adr

-0.67

aders

-0.67

ETHOD

-0.66

ader

-0.66

AUT

-0.66

PsyNetMessage

-0.65

ailable

-0.60

successful

-0.58

 guided

-0.58

POSITIVE LOGITS

nesses

1.00

ening

0.98

bay

0.95

ened

0.94

 Sick

0.87

 sick

0.82

ness

0.81

isbury

0.80

etts

0.79

arse

0.74

Activations Density 0.005%

words related to sickness at varying levels of severity

the word "sick" in various contexts

No Comments

No Known Activations