INDEX

Explanations

references to American institutions or organizations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 11-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L11R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.11.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

uru

-0.18

alom

-0.15

 Herrera

-0.15

 à¤®à¤²

-0.15

 RectTransform

-0.15

pector

-0.15

ride

-0.14

ĵĺ

-0.14

urus

-0.14

icha

-0.14

POSITIVE LOGITS

onet

0.16

yny

0.16

 streak

0.16

ikan

0.16

uguay

0.15

strup

0.15

ivan

0.15

stants

0.15

SES

0.14

åĽ

0.14

Activations Density 0.000%

references to American institutions or organizations

No Comments

No Known Activations