INDEX

Explanations

references to licenses and legal information related to software usage

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

foon

-0.17

ephy

-0.16

¼

-0.16

į¼

-0.15

assa

-0.14

pte

-0.14

anders

-0.14

xbe

-0.14

kur

-0.14

habit

-0.14

POSITIVE LOGITS

oga

0.14

zdy

0.14

oba

0.14

 lane

0.14

.foundation

0.14

ãĤ¹ãĥŀ

0.14

ENU

0.14

VO

0.13

STALL

0.13

ContentLoaded

0.13

Activations Density 0.005%

references to licenses and legal information related to software usage

No Comments

No Known Activations