INDEX

Explanations

references to awards or recognitions in sports or competitive 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

iesz

-0.17

isha

-0.17

ESIS

-0.16

oldt

-0.15

itivity

-0.15

jom

-0.14

venes

-0.14

nicos

-0.14

uns

-0.14

iyas

-0.14

POSITIVE LOGITS

ato

0.35

ando

0.33

ava

0.32

ata

0.31

ati

0.26

are

0.25

ano

0.25

azione

0.25

Ã²

0.23

atura

0.23

Activations Density 0.018%

references to awards or recognitions in sports or competitive contexts

No Comments

No Known Activations