INDEX

Explanations

names of people or entities

oai_token-act-pair · gpt-3.5-turbo

references to individuals or groups involved in political or significant organizational activities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

 tremend

-0.62

 appe

-0.58

 gobl

-0.57

 retribution

-0.56

CHR

-0.56

 outwe

-0.56

schild

-0.55

 spont

-0.54

heimer

-0.54

 toget

-0.53

POSITIVE LOGITS

auld

0.78

igham

0.76

*/(

0.69

taboola

0.68

itaire

0.67

ierre

0.64

Sud

0.64

 Soup

0.63

ibus

0.59

agall

0.58

Activations Density 1.049%

names of people or entities

references to individuals or groups involved in political or significant organizational activities

No Comments

No Known Activations