INDEX

Explanations

references to media reviews and promotional materials

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.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

ÙħØ§ÙĦ

-0.16

enco

-0.14

ÑĪÐºÐ°

-0.14

robots

-0.14

rollo

-0.14

jde

-0.13

mens

-0.13

ç«

-0.13

ÏģÏī

-0.13

Ð²Ð¸ÑĤ

-0.13

POSITIVE LOGITS

zier

0.20

MMdd

0.16

LEGRO

0.16

å»Ĭ

0.15

oker

0.14

yonel

0.14

ARGIN

0.14

urname

0.14

TRGL

0.14

fully

0.13

Activations Density 0.302%

references to media reviews and promotional materials

No Comments

No Known Activations