INDEX

Explanations

punctuation and formatting marks indicating the end of sentences

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

onga

-0.17

 chance

-0.15

ptr

-0.15

-0.14

»

-0.14

 wing

-0.13

aid

-0.13

 perimeter

-0.13

POSITIVE LOGITS

åĲ

0.16

 coli

0.16

 Ð´Ð¾Ð³

0.15

orce

0.15

æ¯ķ

0.15

 autopsy

0.15

zano

0.15

org

0.15

ÐµÐºÑĥ

0.14

eref

0.14

Activations Density 0.108%

punctuation and formatting marks indicating the end of sentences

No Comments

No Known Activations