INDEX

Explanations

legal terms or activities in a text

oai_token-act-pair · gpt-3.5-turbo

references to legal concepts or legal matters

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

xual

-0.83

mble

-0.71

warts

-0.70

ococ

-0.69

pread

-0.68

alin

-0.67

cliffe

-0.67

eful

-0.67

Ble

-0.67

hoff

-0.66

POSITIVE LOGITS

 enforcement

0.85

 precedent

0.84

 highs

0.82

 legal

0.80

IZ

0.77

 counsel

0.76

 adviser

0.75

izes

0.73

pedia

0.73

 recourse

0.72

Activations Density 0.030%

legal terms or activities in a text

references to legal concepts or legal matters

No Comments

No Known Activations