INDEX

Explanations

phrases related to acknowledging or giving recognition to someone or something

oai_token-act-pair · gpt-3.5-turbo

references to giving credit or recognition

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Lans

-0.80

VER

-0.67

 mosquit

-0.66

 Alter

-0.64

 Weird

-0.63

wid

-0.58

terday

-0.58

FO

-0.58

 apprehens

-0.57

viol

-0.57

POSITIVE LOGITS

card

0.83

worthiness

0.79

itism

0.76

rious

0.76

 card

0.75

cards

0.75

saf

0.72

rating

0.72

enza

0.72

abilities

0.72

Activations Density 0.019%

phrases related to acknowledging or giving recognition to someone or something

references to giving credit or recognition

No Comments

No Known Activations