INDEX

Explanations

mentions of legal terms related to consequences and obligations, specifically the term "probation"

oai_token-act-pair · gpt-3.5-turbo

references to probation and related legal terms

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

xxxxxxxx

-0.69

bern

-0.67

vag

-0.66

Medic

-0.66

ynchron

-0.65

mind

-0.64

advertisement

-0.63

 Occupations

-0.62

lycer

-0.62

 Medic

-0.61

POSITIVE LOGITS

 probation

1.27

 confir

0.97

 parole

0.91

 revocation

0.83

etheless

0.77

 describ

0.76

ally

0.74

ary

0.73

avorite

0.72

urous

0.71

Activations Density 0.007%

mentions of legal terms related to consequences and obligations, specifically the term "probation"

references to probation and related legal terms

No Comments

No Known Activations