INDEX

Explanations

numerical data and indicators related to measurements or statistics

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

------+------+

-0.17

unm

-0.16

acci

-0.15

umni

-0.15

 flash

-0.15

poz

-0.14

 miss

-0.14

æ¥½

-0.14

 Gear

-0.14

brook

-0.14

POSITIVE LOGITS

chl

0.16

zf

0.16

chine

0.16

agra

0.15

éĢīæĭ

0.14

chia

0.14

hive

0.14

zan

0.14

 Larson

0.14

alic

0.13

Activations Density 0.001%

numerical data and indicators related to measurements or statistics

No Comments

No Known Activations