INDEX

Explanations

legal terms and concepts

oai_token-act-pair · gpt-3.5-turbo

terms related to laws or legal issues

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

accompan

-0.79

Ct

-0.68

Cong

-0.66

Sov

-0.63

Zot

-0.62

 cannabin

-0.61

Offic

-0.60

 Nare

-0.59

PCI

-0.59

Trend

-0.59

POSITIVE LOGITS

iencies

0.82

phrine

0.81

sembly

0.80

selves

0.75

wered

0.72

icultural

0.70

ishly

0.70

imity

0.69

hed

0.69

ilus

0.68

Activations Density 0.211%

legal terms and concepts

terms related to laws or legal issues

No Comments

No Known Activations