INDEX

Explanations

proper nouns, particularly related to vilification and character assassination

oai_token-act-pair · gpt-3.5-turbo

words related to vilification or negative labeling of individuals

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

soDeliveryDate

-0.76

 cropped

-0.74

ĸļ

-0.71

unct

-0.70

riad

-0.67

uyomi

-0.66

 Brill

-0.65

hr

-0.64

ItemTracker

-0.63

advertisement

-0.63

POSITIVE LOGITS

Vil

1.39

eness

0.94

vil

0.86

uv

0.80

icious

0.80

estone

0.78

estones

0.77

apo

0.77

vil

0.77

atan

0.76

Activations Density 0.014%

proper nouns, particularly related to vilification and character assassination

words related to vilification or negative labeling of individuals

No Comments

No Known Activations