INDEX

Explanations

terms related to engagement and participation

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

ers

-0.17

ahan

-0.16

iggins

-0.15

plevel

-0.14

Ĭ

-0.14

æ°ı

-0.13

éĲĺ

-0.13

ãĥĵãĥ¼

-0.13

pher

-0.13

ering

-0.13

POSITIVE LOGITS

yonel

0.17

uated

0.17

/pass

0.17

/react

0.16

_inactive

0.16

0.15

uar

0.15

endar

0.15

-active

0.14

uator

0.14

Activations Density 0.043%

terms related to engagement and participation

No Comments

No Known Activations