INDEX

Explanations

legal and bureaucratic terms or phrases within legislative texts

oai_token-act-pair · gpt-3.5-turbo

references to specific legal provisions or regulations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

Magikarp

-0.73

ampa

-0.70

ritz

-0.64

CV

-0.63

 Doodle

-0.61

arton

-0.59

 gasp

-0.59

 actionGroup

-0.59

Arc

-0.58

kson

-0.58

POSITIVE LOGITS

ality

1.11

ally

1.10

witz

1.03

alities

0.95

ional

0.91

arians

0.88

arian

0.87

ions

0.86

aler

0.77

mental

0.76

Activations Density 0.028%

legal and bureaucratic terms or phrases within legislative texts

references to specific legal provisions or regulations

No Comments

No Known Activations