INDEX

Explanations

references to governance and authority in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

stdafx

-0.16

avo

-0.15

hone

-0.15

LAY

-0.14

ç·´

-0.14

=''

-0.14

arrant

-0.14

koa

-0.14

assis

-0.13

auer

-0.13

POSITIVE LOGITS

 covered

0.18

ucer

0.16

agus

0.16

 Covered

0.16

0.15

æĿĲ

0.15

buz

0.14

Cum

0.14

cum

0.14

ernote

0.14

Activations Density 0.009%

references to governance and authority in various contexts

No Comments

No Known Activations

references to governance and authority in various contexts

No Comments

No Known Activations