INDEX

Explanations

examples of the word "brand" used in context

oai_token-act-pair · gpt-3.5-turbo

references to the concept of 'brand new'

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

ulhu

-0.88

veyard

-0.84

pmwiki

-0.79

 Able

-0.76

SEE

-0.73

 Poverty

-0.73

*/(

-0.72

cale

-0.71

abama

-0.69

vae

-0.69

POSITIVE LOGITS

ishing

0.94

enburg

0.89

brand

0.86

 loyalty

0.85

ished

0.82

eli

0.81

ages

0.78

aging

0.78

ishes

0.75

 brand

0.74

Activations Density 0.013%

examples of the word "brand" used in context

references to the concept of 'brand new'

No Comments

No Known Activations