INDEX

Explanations

instances of the word "labor" or related words

oai_token-act-pair · gpt-3.5-turbo

references to labor and related terms

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

ebook

-0.74

washer

-0.67

*/(

-0.67

BOOK

-0.65

protection

-0.63

²¾

-0.62

Course

-0.62

edom

-0.60

 Calories

-0.59

milo

-0.59

POSITIVE LOGITS

atory

1.05

abor

0.99

atories

0.83

ators

0.83

icult

0.82

ator

0.82

ately

0.81

atorium

0.79

atively

0.77

icus

0.74

Activations Density 0.014%

instances of the word "labor" or related words

references to labor and related terms

No Comments

No Known Activations