INDEX

Explanations

phrases related to physical harm, especially stabbing

oai_token-act-pair · gpt-3.5-turbo

references to stabbing incidents or related injuries

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

 Cosmos

-0.75

oS

-0.72

mberg

-0.70

 Geographic

-0.69

XM

-0.68

 Afric

-0.64

 Organization

-0.64

AMA

-0.63

 Leban

-0.63

VO

-0.63

POSITIVE LOGITS

 wounds

1.05

lished

0.98

 slit

0.89

 throats

0.84

nery

0.82

 rampage

0.82

 stabbing

0.81

 stab

0.81

nesday

0.80

 wrists

0.80

Activations Density 0.029%

phrases related to physical harm, especially stabbing

references to stabbing incidents or related injuries

No Comments

No Known Activations