INDEX

Explanations

words related to legal issues or obstacles

oai_token-act-pair · gpt-3.5-turbo

terms related to obstruction and interference in justice or legal processes

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Grind

-0.75

ndra

-0.74

MER

-0.71

morph

-0.70

 Volt

-0.68

 Anarchy

-0.68

www

-0.67

 Soph

-0.67

ann

-0.66

derived

-0.65

POSITIVE LOGITS

 obstruction

1.44

 obstruct

1.21

structed

0.93

struction

0.86

 impede

0.83

agall

0.83

insula

0.81

 undermin

0.78

 hemor

0.77

acles

0.76

Activations Density 0.009%

words related to legal issues or obstacles

terms related to obstruction and interference in justice or legal processes

No Comments

No Known Activations