INDEX

Explanations

references to fingers and hands

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 18-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L18R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.18.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

addin

-0.19

shal

-0.17

ards

-0.15

ellido

-0.15

Ø¬Ùĩ

-0.15

-0.14

WARD

-0.14

å¸¯

-0.14

ÐµÑĢÑĸÐ²

-0.14

quez

-0.14

POSITIVE LOGITS

/thumb

0.25

 finger

0.19

prints

0.19

 Finger

0.18

less

0.17

tip

0.16

hold

0.16

lobal

0.16

 pointing

0.15

horn

0.15

Activations Density 0.019%

references to fingers and hands

No Comments

No Known Activations