INDEX

Explanations

academic language and structures related to research findings

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

elow

-0.15

Graph

-0.14

pij

-0.14

Ð´Ð¸Ð½Ð°

-0.14

/**<

-0.13

áº£m

-0.13

ient

-0.13

thood

-0.13

 interrupts

-0.13

POSITIVE LOGITS

 recently

0.25

 authors

0.22

recent

0.20

Recently

0.19

authors

0.19

æľĢè¿ĳ

0.19

eld

0.19

 Recently

0.18

 works

0.18

 later

0.17

Activations Density 0.068%

academic language and structures related to research findings

No Comments

No Known Activations