INDEX

Explanations

phrases related to conflicts or disputes between different parties

oai_token-act-pair · gpt-3.5-turbo

instances of the word "and" in contexts of conflict or relationships

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

spir

-0.72

eker

-0.70

INESS

-0.69

raq

-0.68

needed

-0.68

bard

-0.67

scroll

-0.67

oche

-0.66

 Citation

-0.66

wow

-0.65

POSITIVE LOGITS

 Philadelphia

0.68

 destiny

0.66

 Gujarat

0.63

 Vietnam

0.62

 Camden

0.62

 Chinatown

0.61

 acquire

0.61

 Norfolk

0.61

 Sons

0.61

its

0.60

Activations Density 0.107%

phrases related to conflicts or disputes between different parties

instances of the word "and" in contexts of conflict or relationships

No Comments

No Known Activations