INDEX

Explanations

phrases related to publication and data integrity

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

[array

-0.14

 Leone

-0.14

rej

-0.14

hay

-0.14

ucci

-0.14

orth

-0.13

 Sens

-0.13

 Depression

-0.13

POSITIVE LOGITS

ÐµÐ»Ð¸Ðº

0.17

Éµ

0.16

allel

0.15

iyim

0.15

eced

0.15

ivals

0.14

Ymd

0.14

Î»Î¹Îº

0.14

_grad

0.14

elize

0.14

Activations Density 0.092%

phrases related to publication and data integrity

No Comments

No Known Activations