INDEX

Explanations

language connecting different concepts and ideas

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 4-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L4R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.4.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

MASK

-0.15

pel

-0.15

legates

-0.15

WebResponse

-0.14

 profil

-0.14

ono

-0.14

 Fast

-0.14

pg

-0.14

ä¹ī

-0.14

POSITIVE LOGITS

quip

0.15

aley

0.15

_strerror

0.14

leigh

0.14

ÑģÑĮ

0.13

allee

0.13

/REC

0.13

 Vladim

0.13

cri

0.13

_rect

0.13

Activations Density 0.049%

language connecting different concepts and ideas

No Comments

No Known Activations