INDEX

Explanations

phrases related to clothing or physical appearance

oai_token-act-pair · gpt-3.5-turbo

forms of the word "bare" and its derivations, indicating a focus on exposure or vulnerability

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

OD

-0.72

Mit

-0.69

oret

-0.67

engers

-0.67

Gray

-0.65

YS

-0.61

orem

-0.61

Bom

-0.60

ues

-0.59

OY

-0.59

POSITIVE LOGITS

paren

0.90

yout

0.82

tsky

0.80

hot

0.75

fi

0.73

atches

0.72

ctic

0.71

thro

0.70

entin

0.70

mares

0.70

Activations Density 0.011%

phrases related to clothing or physical appearance

forms of the word "bare" and its derivations, indicating a focus on exposure or vulnerability

No Comments

No Known Activations