INDEX

Explanations

phrases related to a specific location - "Grove"

oai_token-act-pair · gpt-3.5-turbo

references to different types of 'Grove.'

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

als

-0.80

ement

-0.73

ÑĮ

-0.73

amen

-0.72

oling

-0.71

ently

-0.70

manship

-0.68

arians

-0.68

sol

-0.68

myra

-0.67

POSITIVE LOGITS

 Grove

0.95

lore

0.81

tesque

0.80

veyard

0.80

antry

0.71

 Gardens

0.67

iety

0.65

combe

0.64

 Vand

0.63

keepers

0.63

Activations Density 0.021%

phrases related to a specific location - "Grove"

references to different types of 'Grove.'

No Comments

No Known Activations