INDEX

Explanations

company names and references to their products or actions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 21-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L21R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.21.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

emsp

-0.18

geber

-0.17

ÏģÎ±Î²

-0.16

elman

-0.16

ä¹ĭä¸Ģ

-0.14

esiz

-0.14

Italic

-0.14

eyJ

-0.14

@student

-0.14

Providers

-0.13

POSITIVE LOGITS

 itself

0.26

’s

0.24

's

0.24

/the

0.20

 employees

0.19

 officials

0.19

CEO

0.18

 corporate

0.18

 executives

0.17

 spokes

0.17

Activations Density 0.091%

company names and references to their products or actions

No Comments

No Known Activations