INDEX

Explanations

names of authors and contributors affiliated with various works

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 12-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L12R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.12.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

Ã©p

-0.17

mpar

-0.16

 Highlander

-0.15

stery

-0.15

 ÑģÐ±

-0.15

icked

-0.14

abit

-0.14

ONO

-0.13

apon

-0.13

iasi

-0.13

POSITIVE LOGITS

0.15

 pitched

0.14

autop

0.14

argout

0.14

 consect

0.14

 rout

0.14

 åĨĨ

0.14

emb

0.14

iner

0.13

 Ð¿ÐµÑĢÐµÐ¿

0.13

Activations Density 0.092%

names of authors and contributors affiliated with various works

No Comments

No Known Activations