INDEX

Explanations

phrases containing the word "usual" followed by other descriptors

oai_token-act-pair · gpt-3.5-turbo

references to common or typical patterns and behaviors

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

acus

-0.81

zon

-0.79

vic

-0.75

rea

-0.72

mented

-0.70

wic

-0.70

<?

-0.69

hani

-0.66

justice

-0.63

 Darius

-0.63

POSITIVE LOGITS

 disclaimer

0.86

 caveat

0.83

 caveats

0.77

 tendency

0.72

 assumption

0.72

 picture

0.71

 amount

0.71

 assortment

0.70

 thing

0.70

"+

0.69

Activations Density 0.106%

phrases containing the word "usual" followed by other descriptors

references to common or typical patterns and behaviors

No Comments

No Known Activations