INDEX

Explanations

proper nouns, likely names or titles of individuals

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

uib

-0.17

rette

-0.17

qli

-0.16

utow

-0.15

Ð¾Ð»Ð¶

-0.14

Å¼

-0.14

ICLE

-0.14

éŃ

-0.14

etes

-0.14

roti

-0.14

POSITIVE LOGITS

acons

0.15

(er

0.15

jp

0.15

anas

0.15

zar

0.13

åĦ¿

0.13

jt

0.13

 aspir

0.13

Activations Density 0.030%

proper nouns, likely names or titles of individuals

No Comments

No Known Activations