INDEX

Explanations

mentions of gifts and related items

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.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

hoot

-0.18

hci

-0.17

eled

-0.16

haft

-0.16

hu

-0.15

eriod

-0.15

ngth

-0.15

dre

-0.15

erer

-0.15

ductive

-0.15

POSITIVE LOGITS

wrap

0.31

wrapped

0.29

 wrap

0.26

 certificate

0.26

-wrap

0.25

 ideas

0.25

ware

0.25

 certificates

0.25

 basket

0.24

 baskets

0.24

Activations Density 0.015%

mentions of gifts and related items

No Comments

No Known Activations