INDEX

Explanations

instances of names or titles related to people or organizations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 2-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L2R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.2.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

ISCO

-0.15

agy

-0.15

odiac

-0.14

uida

-0.14

iche

-0.14

Ĥæķ°

-0.14

-Javadoc

-0.14

ooter

-0.13

uja

-0.13

 squad

-0.13

POSITIVE LOGITS

asia

0.15

illos

0.14

illo

0.14

rif

0.14

 mine

0.14

mie

0.14

ilia

0.14

mort

0.14

atings

0.14

blade

0.14

Activations Density 0.336%

instances of names or titles related to people or organizations

No Comments

No Known Activations