INDEX

Explanations

dates or number ranges formatted with '-' and 'to'

oai_token-act-pair · gpt-3.5-turbo

references to age and age-related categories

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

ãĤ©

-0.70

eless

-0.67

ucci

-0.66

etsk

-0.64

©¶æ

-0.62

earable

-0.62

£ı

-0.62

 âĵĺ

-0.61

nels

-0.61

 McCull

-0.60

POSITIVE LOGITS

0.78

century

0.73

PsyNetMessage

0.70

0.69

0.66

minute

0.66

undreds

0.65

OLD

0.65

Activations Density 0.132%

dates or number ranges formatted with '-' and 'to'

references to age and age-related categories

No Comments

No Known Activations