INDEX

Explanations

commands or suggestions to remain attentive or engaged

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 23-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L23R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.23.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

ickle

-0.17

VER

-0.14

arrow

-0.14

 Beet

-0.14

.reserve

-0.14

ķĮ

-0.14

atti

-0.14

.heroku

-0.13

POSITIVE LOGITS

 tuned

0.34

Tun

0.29

 tuning

0.26

 tune

0.24

 tunes

0.23

tun

0.23

 Tune

0.21

 tuner

0.17

 Connected

0.16

esson

0.16

Activations Density 0.022%

commands or suggestions to remain attentive or engaged

No Comments

No Known Activations

commands or suggestions to remain attentive or engaged

No Comments

No Known Activations