INDEX

Explanations

references to official titles and positions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

 SolidColorBrush

-0.14

 DÃ¼n

-0.13

ascar

-0.12

eyle

-0.12

ancode

-0.12

ADIO

-0.12

biÃªn

-0.12

âĢ¦

-0.12

intColor

-0.12

ãĥ³ãĥĩ

-0.12

POSITIVE LOGITS

ulton

0.14

Ã©v

0.13

 Morrow

0.13

/commons

0.13

rych

0.13

.TryParse

0.13

woff

0.12

ÃħÅ¸

0.12

oire

0.12

 Ã¢

0.12

Activations Density 0.156%

references to official titles and positions

No Comments

No Known Activations