INDEX

Explanations

explicit mentions of rules or restrictions being imposed on specific behaviors, actions, or groups

oai_token-act-pair · gpt-3.5-turbo

phrases related to permissions and prohibitions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

 Soldier

-0.70

xon

-0.67

lves

-0.65

lust

-0.62

posure

-0.62

borough

-0.61

athan

-0.61

 Generation

-0.61

center

-0.60

sis

-0.59

POSITIVE LOGITS

Reviewer

1.09

uthor

0.87

 exemptions

0.79

ommod

0.76

ľ

0.74

 allowed

0.72

 disclaim

0.70

ptin

0.70

 permitted

0.70

ravel

0.69

Activations Density 0.038%

explicit mentions of rules or restrictions being imposed on specific behaviors, actions, or groups

phrases related to permissions and prohibitions

No Comments

No Known Activations