INDEX

Explanations

legal or judicial terms and concepts

oai_token-act-pair · gpt-3.5-turbo

phrases related to legal consequences and evaluations of evidence

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

 Mandarin

-0.55

mber

-0.55

itialized

-0.52

Ago

-0.50

cember

-0.49

Introduced

-0.49

 mornings

-0.49

 Gutenberg

-0.48

Thu

-0.48

nesday

-0.48

POSITIVE LOGITS

ADA

0.55

ilty

0.51

OUGH

0.51

iciency

0.49

ilt

0.48

etc

0.48

efully

0.48

 defenses

0.47

chemy

0.47

 passage

0.47

Activations Density 0.543%

legal or judicial terms and concepts

phrases related to legal consequences and evaluations of evidence

No Comments

No Known Activations