INDEX

Explanations

warnings or instances of caution

oai_token-act-pair · gpt-3.5-turbo

words related to the concept of caution or warnings

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

haul

-0.70

elf

-0.69

elt

-0.68

á

-0.67

hof

-0.66

iop

-0.65

intern

-0.63

Ķ

-0.63

estone

-0.63

ergy

-0.63

POSITIVE LOGITS

 caution

1.48

autions

0.89

 precaution

0.85

 cautioned

0.83

ution

0.78

 cautious

0.76

 condem

0.72

avascript

0.72

ashtra

0.71

 defic

0.70

Activations Density 0.008%

warnings or instances of caution

words related to the concept of caution or warnings

No Comments

No Known Activations