INDEX

Explanations

terms related to gender, including gender-neutral or gender-based concepts

oai_token-act-pair · gpt-3.5-turbo

phrases related to gender and related policies

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

shire

-0.74

hiba

-0.72

×ķ

-0.70

hower

-0.65

KC

-0.64

Bus

-0.64

deck

-0.64

 Cheong

-0.64

×©

-0.64

NK

-0.64

POSITIVE LOGITS

itives

0.84

 ethnicity

0.76

 inheritance

0.75

ethnic

0.72

 influences

0.71

 Hispanic

0.71

 minorities

0.70

 ancestry

0.70

 pronouns

0.69

 representation

0.69

Activations Density 0.270%

terms related to gender, including gender-neutral or gender-based concepts

phrases related to gender and related policies

No Comments

No Known Activations