INDEX

Explanations

references to numerical data and metrics related to 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 @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.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

ange

-0.16

hape

-0.15

aight

-0.15

uÃ§

-0.14

åł±

-0.14

ONO

-0.14

ireccion

-0.14

steder

-0.14

abouts

-0.14

 Seeder

-0.14

POSITIVE LOGITS

arn

0.17

amas

0.16

olla

0.16

Om

0.15

WM

0.15

 Studio

0.15

à¹Ģà¸Ħà¸£

0.15

oll

0.14

 Rick

0.14

rax

0.14

Activations Density 0.040%

references to numerical data and metrics related to various contexts

No Comments

No Known Activations