INDEX

Explanations

terms related to the concept of leaving or exiting

oai_token-act-pair · gpt-3.5-turbo

references to the concept of "exit," particularly in political or significant contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

acebook

-0.82

elf

-0.79

illac

-0.78

eta

-0.75

yright

-0.73

hed

-0.73

hemy

-0.71

yer

-0.71

eday

-0.68

enegger

-0.68

POSITIVE LOGITS

 pupil

0.75

 spoiler

0.74

 door

0.71

 exit

0.69

 doors

0.67

 Door

0.65

Exit

0.65

 velocity

0.64

 combustion

0.62

 valve

0.62

Activations Density 0.023%

terms related to the concept of leaving or exiting

references to the concept of "exit," particularly in political or significant contexts

No Comments

No Known Activations