INDEX

Explanations

phrases indicating distance or extent

oai_token-act-pair · gpt-3.5-turbo

phrases related to proximity and speed

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

emaker

-0.78

ãĤ´ãĥ³

-0.75

ħĭ

-0.75

kaya

-0.74

bda

-0.73

ãģĮ

-0.72

eteen

-0.72

iour

-0.71

jew

-0.70

»

-0.70

POSITIVE LOGITS

 they

1.09

we

0.90

 these

0.85

 THEY

0.83

 things

0.81

 Europeans

0.78

it

0.77

 humans

0.77

he

0.76

 each

0.74

Activations Density 0.109%

phrases indicating distance or extent

phrases related to proximity and speed

No Comments

No Known Activations