INDEX

Explanations

phrases that indicate exceptions or conditions

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

 bÃ¥de

-0.19

iliar

-0.16

 BOTH

-0.15

’autres

-0.15

 nejen

-0.15

 both

-0.15

ãģªãģı

-0.15

lasses

-0.15

uctose

-0.14

POSITIVE LOGITS

 maybe

0.30

 occasional

0.29

 perhaps

0.28

maybe

0.26

few

0.24

 minor

0.24

 occasionally

0.23

 possibly

0.23

perhaps

0.23

 Maybe

0.22

Activations Density 0.110%

phrases that indicate exceptions or conditions

No Comments

No Known Activations