INDEX

Explanations

geographical references and country-related terms

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 13-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L13R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.13.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

mares

-0.15

↵

-0.15

achs

-0.15

ãĥŃãĥ³

-0.14

URES

-0.14

een

-0.14

ħ§

-0.13

 infrastructure

-0.13

exc

-0.13

edo

-0.13

POSITIVE LOGITS

xBA

0.15

vla

0.15

æ¡

0.15

alli

0.15

Ã¢r

0.15

lub

0.15

esin

0.14

VML

0.14

 ActiveForm

0.14

CJK

0.14

Activations Density 0.002%

geographical references and country-related terms

No Comments

No Known Activations