INDEX

Explanations

references to additional benefits or perks

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

 Further

-0.19

put

-0.18

 further

-0.18

Further

-0.16

baar

-0.16

isko

-0.15

soever

-0.15

stone

-0.15

RECT

-0.15

istle

-0.15

POSITIVE LOGITS

ordinary

0.39

ordin

0.29

-large

0.28

ord

0.28

ORD

0.27

-extra

0.26

-cur

0.26

ordinal

0.24

-special

0.24

 virgin

0.23

Activations Density 0.019%

references to additional benefits or perks

No Comments

No Known Activations