INDEX

Explanations

references to windows

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 28-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L28R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.28.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

vÃ¤

-0.17

geb

-0.15

PAD

-0.15

ideographic

-0.15

egov

-0.15

loff

-0.15

ugi

-0.15

lej

-0.15

rada

-0.15

ilon

-0.14

POSITIVE LOGITS

pane

0.40

ill

0.28

ills

0.27

pan

0.27

-pane

0.24

æĪ·

0.23

ed

0.23

less

0.22

 sill

0.22

shade

0.21

Activations Density 0.027%

references to windows

No Comments

No Known Activations