INDEX

Explanations

references to governmental or bureaucratic proceedings

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 18-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L18R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.18.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

mts

-0.15

-0.14

azon

-0.14

parency

-0.14

orp

-0.14

ÑıÑģÑĮ

-0.14

-0.13

 mult

-0.13

 Cole

-0.13

Ã¥n

-0.13

POSITIVE LOGITS

ilk

0.17

memberOf

0.16

Î¹ÎºÎ¿

0.16

æĿ¥æºĲ

0.15

ANI

0.15

ophil

0.15

ollo

0.14

ertino

0.14

aga

0.14

kus

0.14

Activations Density 0.067%

references to governmental or bureaucratic proceedings

No Comments

No Known Activations