INDEX

Explanations

mentions of warfare or fighting

oai_token-act-pair · gpt-3.5-turbo

mentions of the term "War."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

soever

-0.85

ollow

-0.81

ĵĺ

-0.79

Newsletter

-0.79

enegger

-0.78

*/(

-0.77

ô

-0.76

umph

-0.76

lear

-0.75

¦

-0.75

POSITIVE LOGITS

 Typh

0.74

 Malf

0.73

 Mushroom

0.69

ATA

0.69

zu

0.67

RAW

0.67

 latitude

0.66

inet

0.65

cloth

0.65

 Minotaur

0.63

Activations Density 0.000%

mentions of warfare or fighting

mentions of the term "War."

No Comments

No Known Activations