INDEX

Explanations

phrases indicating complaints or criticisms

oai_token-act-pair · gpt-3.5-turbo

phrases expressing complaints

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

ulhu

-0.77

IQ

-0.75

NET

-0.74

fman

-0.74

utical

-0.70

omething

-0.67

oak

-0.67

itect

-0.66

nostic

-0.65

soDeliveryDate

-0.65

POSITIVE LOGITS

 incess

0.85

 inaction

0.80

 sore

0.78

 injust

0.72

 grievances

0.72

 injustice

0.71

 perceived

0.71

 mobs

0.70

 poor

0.69

 bitterly

0.69

Activations Density 0.228%

phrases indicating complaints or criticisms

phrases expressing complaints

No Comments

No Known Activations