INDEX

Explanations

phrases related to different types of rooms and their activities or features within a setting such as a hotel or household

oai_token-act-pair · gpt-3.5-turbo

references to rooms, particularly hotel or lodging rooms

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

utsch

-0.77

AFP

-0.64

clipse

-0.63

icide

-0.63

hog

-0.62

za

-0.62

strong

-0.61

athan

-0.59

AFP

-0.59

antle

-0.59

POSITIVE LOGITS

 rooms

3.74

 Rooms

2.83

rooms

2.73

 bedrooms

2.30

 room

2.07

 Room

1.97

room

1.91

 halls

1.88

Room

1.86

 chambers

1.83

Activations Density 0.021%

phrases related to different types of rooms and their activities or features within a setting such as a hotel or household

references to rooms, particularly hotel or lodging rooms

No Comments

No Known Activations