INDEX

Explanations

instances of legal or regulatory terms and documents

oai_token-act-pair · gpt-3.5-turbo

terms and phrases related to legislation and legal matters

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

Ops

-0.69

 Solitaire

-0.64

oken

-0.63

VIDEO

-0.63

skill

-0.62

Kis

-0.61

 Dresden

-0.61

akeru

-0.61

ularity

-0.61

ebus

-0.60

POSITIVE LOGITS

 forb

0.72

 mandated

0.69

Enlarge

0.69

 guiActiveUn

0.69

 muster

0.66

 levied

0.66

 "$:/

0.66

 cens

0.65

 regulators

0.65

 outlaw

0.63

Activations Density 0.760%

instances of legal or regulatory terms and documents

terms and phrases related to legislation and legal matters

No Comments

No Known Activations