INDEX

Explanations

phrases related to physical actions or movements

oai_token-act-pair · gpt-3.5-turbo

phrases related to sports and physical activity

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

 Membership

-0.88

 Anonymous

-0.75

Anonymous

-0.72

 membership

-0.69

Pand

-0.68

 Adult

-0.67

amily

-0.66

 Patreon

-0.66

 multicultural

-0.66

 Residents

-0.66

POSITIVE LOGITS

 manoeuv

1.31

 braking

1.19

 crou

1.18

 maneuver

1.17

 reflex

1.11

 momentum

1.10

 dodging

1.09

 drib

1.08

 angles

1.08

 bounces

1.06

Activations Density 0.769%

phrases related to physical actions or movements

phrases related to sports and physical activity

No Comments

No Known Activations