INDEX

Explanations

statistical analysis and methodology in research

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 23-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L23R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.23.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

tÃŃ

-0.17

radient

-0.16

 Chain

-0.15

inki

-0.15

eniz

-0.15

XHR

-0.15

ensing

-0.14

amment

-0.14

 chain

-0.14

asil

-0.14

POSITIVE LOGITS

 Mann

0.32

chi

0.31

Wil

0.28

Chi

0.26

ANO

0.26

 Student

0.26

AN

0.26

Chi

0.26

Wil

0.25

chi

0.24

Activations Density 0.041%

statistical analysis and methodology in research

No Comments

No Known Activations