INDEX

Explanations

references to disabilities

oai_token-act-pair · gpt-3.5-turbo

mentions of disabilities and related terms

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

 ellipt

-0.75

anza

-0.74

furt

-0.70

orically

-0.70

lar

-0.69

gres

-0.69

kov

-0.68

hunt

-0.67

ãĥĥãĥĪ

-0.67

ogle

-0.66

POSITIVE LOGITS

 disability

1.01

 disabilities

1.01

 Disability

0.99

 Disabled

0.84

 impair

0.81

 Discrimination

0.79

afe

0.74

rights

0.72

 cripp

0.72

abled

0.71

Activations Density 0.016%

references to disabilities

mentions of disabilities and related terms

No Comments

No Known Activations