INDEX

Explanations

action words related to intentionally not paying attention to something

oai_token-act-pair · gpt-3.5-turbo

instances of the word "ignore" and its variations

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ccording

-0.78

ray

-0.75

iov

-0.73

si

-0.71

sav

-0.71

rim

-0.70

isher

-0.70

agra

-0.69

gans

-0.69

ernels

-0.68

POSITIVE LOGITS

ibly

0.80

illy

0.74

 warnings

0.66

 cues

0.66

ibility

0.65

fulness

0.63

 altogether

0.60

 stares

0.60

 hears

0.59

 aside

0.57

Activations Density 0.041%

action words related to intentionally not paying attention to something

instances of the word "ignore" and its variations

No Comments

No Known Activations