INDEX

Explanations

subgroups or categories within a larger group or population

oai_token-act-pair · gpt-3.5-turbo

references to various groups in a study or analysis context

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

shire

-0.86

 concess

-0.71

 appre

-0.68

LV

-0.67

CLASSIFIED

-0.66

DAY

-0.66

issance

-0.65

clinton

-0.64

yy

-0.63

 Dame

-0.63

POSITIVE LOGITS

roups

1.02

arettes

0.97

emonium

0.89

naires

0.89

rats

0.89

ings

0.87

istical

0.85

arette

0.84

think

0.83

ages

0.81

Activations Density 0.037%

subgroups or categories within a larger group or population

references to various groups in a study or analysis context

No Comments

No Known Activations