INDEX

Explanations

references to sources, indicating the importance of citations or documented information

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

erva

-0.16

uzzi

-0.15

acon

-0.15

otel

-0.15

pok

-0.14

ampus

-0.14

Weg

-0.14

 Schwartz

-0.14

 institution

-0.14

alysis

-0.14

POSITIVE LOGITS

zell

0.16

iyon

0.14

æı

0.14

-Ð²

0.14

asd

0.14

mÃ¡n

0.14

Ð¾Ð½ÑĮ

0.14

hetic

0.14

utz

0.13

à¤¿à¤£

0.13

Activations Density 0.006%

references to sources, indicating the importance of citations or documented information

No Comments

No Known Activations