INDEX

Explanations

references to high quality or premium products

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

ancode

-0.15

umerator

-0.15

à¸Ľ

-0.15

ssize

-0.15

eff

-0.15

trial

-0.14

ifo

-0.14

\CMS

-0.14

orative

-0.13

al

-0.13

POSITIVE LOGITS

/high

0.20

(er

0.19

ley

0.17

à¤¤à¤®

0.17

sword

0.16

lands

0.16

ened

0.15

)prepare

0.15

-rise

0.15

ening

0.14

Activations Density 0.082%

references to high quality or premium products

No Comments

No Known Activations