INDEX

Explanations

expressions of confidence and change

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 19-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L19R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.19.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

 zwar

-0.27

 sice

-0.27

èĻ½çĦ¶

-0.20

 indeed

-0.20

 therefore

-0.19

btw

-0.19

setItem

-0.18

 though

-0.17

 donc

-0.17

çĦ¶èĢĮ

-0.17

POSITIVE LOGITS

 also

0.26

 still

0.23

also

0.22

 nonetheless

0.22

 nevertheless

0.21

è¿ĺæĺ¯

0.21

 ALSO

0.21

åį»

0.20

 Ø£ÙĬØ¶Ø§

0.20

åį´

0.19

Activations Density 0.681%

expressions of confidence and change

No Comments

No Known Activations