INDEX

Explanations

references to legal terms, specifically related to probation

oai_token-act-pair · gpt-3.5-turbo

terms related to probation and its implications

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

xxxxxxxx

-0.70

 Medic

-0.67

lli

-0.66

Medic

-0.65

Jet

-0.65

advertisement

-0.64

vol

-0.63

mind

-0.62

Temperature

-0.61

Dust

-0.60

POSITIVE LOGITS

 probation

1.42

 confir

1.08

 parole

1.05

avorite

0.91

etheless

0.89

 revocation

0.87

 transgress

0.82

 describ

0.79

 agre

0.78

ally

0.76

Activations Density 0.010%

references to legal terms, specifically related to probation

terms related to probation and its implications

No Comments

No Known Activations