INDEX

Explanations

mentions of specific medical conditions or technical terms, particularly related to psychology and health

oai_token-act-pair · gpt-3.5-turbo

terms related to psychiatric conditions and their symptoms

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

 FANTASY

-0.66

gaard

-0.64

hold

-0.64

 abiding

-0.63

ORGE

-0.63

VERT

-0.63

 ACTION

-0.62

separ

-0.61

 Aval

-0.59

 Parenthood

-0.59

POSITIVE LOGITS

yp

1.24

olicy

1.00

olit

0.96

aired

0.94

posium

0.91

ython

0.89

osite

0.88

ocalypse

0.88

utic

0.86

oen

0.86

Activations Density 0.010%

mentions of specific medical conditions or technical terms, particularly related to psychology and health

terms related to psychiatric conditions and their symptoms

No Comments

No Known Activations