INDEX

Explanations

phrases indicating contrast or exception

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

not

-0.21

 nicht

-0.18

NOT

-0.18

 ALSO

-0.17

 also

-0.17

à¹Ħà¸¡

-0.17

ä¸įåĪ°

-0.16

 khÃ´ng

-0.16

no

-0.16

ello

-0.16

POSITIVE LOGITS

 certainly

0.35

nor

0.34

Nor

0.31

 sÃŃ

0.31

nor

0.28

Nor

0.27

 Certainly

0.26

Nevertheless

0.26

 definitely

0.24

 sondern

0.24

Activations Density 0.267%

phrases indicating contrast or exception

No Comments

No Known Activations