INDEX

Explanations

timestamps and publication information

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 22-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L22R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.22.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

osl

-0.18

_mpi

-0.15

ve

-0.15

cube

-0.14

icast

-0.14

vÃ©

-0.14

 Tear

-0.14

enheim

-0.14

rn

-0.14

corn

-0.14

POSITIVE LOGITS

á»ĳi

0.17

rais

0.14

atform

0.14

alsy

0.14

 darm

0.14

arium

0.14

ellite

0.14

 Burada

0.14

âķĳ

0.14

Cro

0.13

Activations Density 0.003%

timestamps and publication information

No Comments

No Known Activations