INDEX

Explanations

phrases related to guidance or instructions for creating or preparing items

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 18-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L18R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.18.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

ond

-0.15

otts

-0.14

bject

-0.14

umping

-0.14

.showMessage

-0.14

anyl

-0.14

omentum

-0.14

icense

-0.14

reso

-0.14

_BP

-0.13

POSITIVE LOGITS

vo

0.45

Vo

0.39

Vo

0.35

vo

0.35

 prest

0.30

VO

0.28

 viol

0.27

VO

0.25

.vo

0.24

 Done

0.23

Activations Density 0.189%

phrases related to guidance or instructions for creating or preparing items

No Comments

No Known Activations