INDEX

Explanations

phrases indicating a journey or process, particularly those that emphasize the path taken

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

ulet

-0.18

ystone

-0.16

wheel

-0.16

rav

-0.16

 wheel

-0.16

isz

-0.15

-wheel

-0.14

rapper

-0.14

ule

-0.14

ë¹Ī

-0.14

POSITIVE LOGITS

way

0.39

Way

0.26

-way

0.26

way

0.26

_way

0.24

WAY

0.24

.way

0.22

 lines

0.21

Way

0.21

Lines

0.19

Activations Density 0.008%

phrases indicating a journey or process, particularly those that emphasize the path taken

No Comments

No Known Activations