INDEX

Explanations

words related to violent criminal activities, specifically murder

oai_token-act-pair · gpt-3.5-turbo

words associated with murder or violent acts

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

 Tibetan

-0.65

OHN

-0.65

 Pixie

-0.65

AU

-0.64

 Shinra

-0.64

 Fundamental

-0.64

 Shining

-0.62

orers

-0.59

 Hercules

-0.58

 Palo

-0.58

POSITIVE LOGITS

mur

1.38

ãĥ£

0.94

boats

0.88

Mur

0.86

geon

0.86

boat

0.86

izont

0.84

cloth

0.83

amaz

0.82

chief

0.82

Activations Density 0.005%

words related to violent criminal activities, specifically murder

words associated with murder or violent acts

No Comments

No Known Activations