INDEX

Explanations

words related to eliminating or restricting something

oai_token-act-pair · gpt-3.5-turbo

pronouns and related personal terms

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

 Sorceress

-0.71

 Painter

-0.63

AVG

-0.58

 Nanto

-0.57

 accuser

-0.57

 Instruct

-0.56

 Jackets

-0.56

 Pastebin

-0.56

 mediation

-0.56

 Manager

-0.56

POSITIVE LOGITS

andre

0.94

ailability

0.80

anding

0.76

abeth

0.75

vertisement

0.73

OND

0.73

aution

0.71

aire

0.70

ATIONS

0.68

onde

0.68

Activations Density 0.117%

words related to eliminating or restricting something

pronouns and related personal terms

No Comments

No Known Activations