INDEX

Explanations

geographical locations or references within the document

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 21-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L21R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.21.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

adio

-0.19

agher

-0.17

ostel

-0.16

agna

-0.15

ackers

-0.15

odium

-0.15

nev

-0.15

cura

-0.14

Ø§ÙĨÙĬØ©

-0.14

 southern

-0.14

POSITIVE LOGITS

à¥

0.16

ylland

0.15

zone

0.15

/Dk

0.15

vtk

0.15

 Shelf

0.14

imir

0.14

 Ø§ÙĦØªØ¬

0.14

imore

0.14

Î±ÏģÎ¬

0.14

Activations Density 0.100%

geographical locations or references within the document

No Comments

No Known Activations