INDEX

Explanations

terms related to myths, stories, or famous individuals

oai_token-act-pair · gpt-3.5-turbo

references to legends or legendary figures

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

abus

-0.80

 Parenthood

-0.75

orneys

-0.74

artment

-0.71

lyak

-0.70

pex

-0.70

Interstitial

-0.67

zers

-0.67

ells

-0.66

tan

-0.66

POSITIVE LOGITS

arily

1.13

lore

0.85

 tales

0.85

 legends

0.85

Legend

0.84

arium

0.82

 legend

0.82

arious

0.76

 Legend

0.74

 lore

0.73

Activations Density 0.016%

terms related to myths, stories, or famous individuals

references to legends or legendary figures

No Comments

No Known Activations