INDEX

Explanations

numerical values and percentages associated with time stamps or measurements

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 4-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L4R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.4.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

uzzi

-0.17

fen

-0.16

amer

-0.16

erosis

-0.15

á»ĩu

-0.15

Ð½ÐµÑģ

-0.14

eya

-0.14

opsy

-0.14

affer

-0.14

ophe

-0.14

POSITIVE LOGITS

 Vinci

0.15

urance

0.14

 Chapman

0.14

fat

0.14

achs

0.14

sville

0.14

ib

0.13

Ah

0.13

uran

0.13

 Gross

0.13

Activations Density 0.078%

numerical values and percentages associated with time stamps or measurements

No Comments

No Known Activations