INDEX

Explanations

various forms of the word "formula" and its associated context

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

-0.18

 Ø¨ÙĪØ§Ø¨Ø©

-0.16

mark

-0.16

age

-0.15

Ã¶h

-0.15

ucu

-0.15

rot

-0.14

reo

-0.14

uft

-0.14

fall

-0.14

POSITIVE LOGITS

0.20

 vitae

0.18

ic

0.17

ÑģÑĮ

0.17

naire

0.16

ural

0.16

Ã§Ã£o

0.15

oin

0.15

errupted

0.15

ardÄ±m

0.15

Activations Density 0.046%

various forms of the word "formula" and its associated context

No Comments

No Known Activations