INDEX

Explanations

references to sports performance metrics and player achievements

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

arkin

-0.16

aspers

-0.16

wner

-0.15

wich

-0.15

esti

-0.15

arse

-0.15

enville

-0.14

horn

-0.14

-vars

-0.14

POSITIVE LOGITS

íĬ

0.15

imoto

0.15

enko

0.15

rescia

0.14

 Rolls

0.14

reon

0.14

 performance

0.14

bib

0.14

 nonatomic

0.14

ERN

0.13

Activations Density 0.288%

references to sports performance metrics and player achievements

No Comments

No Known Activations