INDEX

Explanations

specific names and proper nouns

oai_token-act-pair · gpt-3.5-turbo

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Âł Âł

-0.70

İĭ

-0.69

 Trees

-0.68

CLASSIFIED

-0.65

\'

-0.64

 Yellowstone

-0.64

ĸļ

-0.64

 Âł Âł Âł Âł Âł Âł Âł Âł

-0.62

 GOODMAN

-0.62

 Electricity

-0.62

POSITIVE LOGITS

ap

0.85

lip

0.84

hop

0.80

vati

0.80

ob

0.79

mor

0.79

apt

0.78

ar

0.78

sin

0.77

Activations Density 0.275%

specific names and proper nouns

No Comments

No Known Activations