INDEX

Explanations

numerical data and references to articles or datasets

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 31-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L31R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.31.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

an

-0.31

-0.30

the

-0.25

-0.24

un

-0.24

in

-0.23

st

-0.23

re

-0.23

POSITIVE LOGITS

0.44

0.38

0.31

,↵

0.26

↵

0.24

0.23

↵↵

0.21

Activations Density 0.474%

numerical data and references to articles or datasets

No Comments

No Known Activations