INDEX

Explanations

references to fatal incidents or medical conditions that have the potential to be deadly

oai_token-act-pair · gpt-3.5-turbo

terms related to fatal outcomes or serious consequences

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

mble

-0.78

Kit

-0.77

lease

-0.76

EY

-0.75

ï¸

-0.75

HER

-0.75

lies

-0.74

rence

-0.73

lyak

-0.73

uden

-0.72

POSITIVE LOGITS

istic

1.02

ities

1.00

istically

0.89

 indign

0.87

 vigilance

0.78

istics

0.78

 Fatal

0.77

isy

0.77

ist

0.77

ism

0.76

Activations Density 0.055%

references to fatal incidents or medical conditions that have the potential to be deadly

terms related to fatal outcomes or serious consequences

No Comments

No Known Activations