INDEX

Explanations

phrases indicating legal or moral entitlement

oai_token-act-pair · gpt-3.5-turbo

phrases related to individual rights and liberties

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

 Madness

-0.60

iries

-0.59

 diligent

-0.57

 Spiel

-0.57

 unsuspecting

-0.56

 batches

-0.55

 Journals

-0.54

metics

-0.54

 Hort

-0.54

duction

-0.54

POSITIVE LOGITS

 whatsoever

0.85

0.77

 vested

0.74

ointed

0.74

 veto

0.69

kees

0.69

 entit

0.69

âĺ

0.68

amus

0.68

ï¸

0.67

Activations Density 0.159%

phrases indicating legal or moral entitlement

phrases related to individual rights and liberties

No Comments

No Known Activations