INDEX

Explanations

references to product categories or specifications in technology

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 12-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L12R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.12.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

è¼Ŀ

-0.15

hawk

-0.14

.libs

-0.14

 pars

-0.14

aar

-0.14

ufen

-0.13

fol

-0.13

zon

-0.13

rette

-0.13

ToPoint

-0.13

POSITIVE LOGITS

raquo

0.19

.scalablytyped

0.16

cta

0.16

æ¸¸

0.15

 Truy

0.14

bst

0.14

etooth

0.14

xis

0.14

ecut

0.14

Ø¹ÙĨÙĪØ§ÙĨ

0.14

Activations Density 0.037%

references to product categories or specifications in technology

No Comments

No Known Activations