INDEX

Explanations

quantifiable statistics and numerical values related to performance

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.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

 Mats

-0.17

stra

-0.14

pery

-0.14

åĿĤ

-0.13

ugu

-0.13

aira

-0.13

VD

-0.13

uct

-0.13

POSITIVE LOGITS

ç¹

0.15

dee

0.15

ìĬ¹

0.15

erence

0.15

ateria

0.15

sein

0.15

asion

0.14

oren

0.14

ì¼ĢìĿ´

0.14

ä»¥ä¸Ĭ

0.13

Activations Density 0.306%

quantifiable statistics and numerical values related to performance

No Comments

No Known Activations