INDEX

Explanations

references to well-known individuals or significant figures in various contexts

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

unker

-0.15

çľĭçľĭ

-0.15

ka

-0.15

ToFront

-0.15

apolis

-0.15

åĳ¢

-0.14

esp

-0.14

 uÄį

-0.14

rame

-0.14

 Frequency

-0.14

POSITIVE LOGITS

 well

0.23

 WELL

0.21

well

0.19

 better

0.18

 drill

0.18

 mieux

0.17

Well

0.17

 dobÅĻe

0.16

andan

0.16

UILTIN

0.16

Activations Density 0.155%

references to well-known individuals or significant figures in various contexts

No Comments

No Known Activations