INDEX

Explanations

references to political or legal mandates

oai_token-act-pair · gpt-3.5-turbo

references to legal mandates or mandatory policies

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

istics

-1.03

folk

-0.91

stocks

-0.88

hop

-0.87

ophon

-0.84

ienne

-0.82

Hop

-0.78

can

-0.77

cases

-0.77

ihad

-0.76

POSITIVE LOGITS

 compliance

0.79

 enforced

0.76

 creep

0.76

 decree

0.75

 mandate

0.75

 conformity

0.74

 mand

0.73

 obedience

0.72

 imposed

0.72

 breaker

0.71

Activations Density 0.077%

references to political or legal mandates

references to legal mandates or mandatory policies

No Comments

No Known Activations