INDEX

Explanations

mentions of being banned

oai_token-act-pair · gpt-3.5-turbo

instances of the word "banned" across various contexts

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

eah

-0.80

issance

-0.73

 IMAGES

-0.73

 Generations

-0.69

Auth

-0.68

sie

-0.66

rious

-0.66

ickey

-0.64

ilant

-0.64

prise

-0.64

POSITIVE LOGITS

hee

0.91

ishment

0.80

 banning

0.77

netflix

0.76

 banned

0.76

 bans

0.73

 substances

0.73

 smoking

0.72

hammer

0.72

wana

0.71

Activations Density 0.019%

mentions of being banned

instances of the word "banned" across various contexts

No Comments

No Known Activations