INDEX

Explanations

keywords related to bias in various contexts

oai_token-act-pair · gpt-3.5-turbo

references to various forms of bias

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

Interstitial

-0.85

export

-0.81

ership

-0.80

clamation

-0.76

sa

-0.75

ODUCT

-0.73

VID

-0.72

eu

-0.72

agra

-0.72

si

-0.71

POSITIVE LOGITS

 bias

1.23

 biases

1.17

 prejud

0.95

 prejudice

0.92

 detector

0.86

 inference

0.85

 prejudices

0.84

 biased

0.82

 skew

0.80

 towards

0.78

Activations Density 0.009%

keywords related to bias in various contexts

references to various forms of bias

No Comments

No Known Activations