INDEX

Explanations

phrases related to legal issues and disputes

oai_token-act-pair · gpt-3.5-turbo

phrases indicating vulnerability or risk to particular groups or entities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

 Recre

-0.70

 renovations

-0.64

Aer

-0.63

 Seasons

-0.61

 renov

-0.60

 Navigation

-0.59

Initial

-0.58

perty

-0.56

 swapped

-0.56

 Compan

-0.56

POSITIVE LOGITS

 hypocritical

0.80

 disingen

0.75

AFP

0.73

 indict

0.73

 hypocrisy

0.73

 peril

0.71

 alarm

0.71

 shudder

0.70

CHA

0.70

 damning

0.70

Activations Density 0.683%

phrases related to legal issues and disputes

phrases indicating vulnerability or risk to particular groups or entities

No Comments

No Known Activations