INDEX

Explanations

mentions of news personalities and their affiliations

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

 Intermediate

-0.17

 Sanat

-0.16

 Beckham

-0.15

à¸²à¸ģà¸¥

-0.15

okus

-0.14

olia

-0.14

à¸²à¸Ķ

-0.14

advertisement

-0.14

.appspot

-0.14

ereco

-0.13

POSITIVE LOGITS

Fox

0.33

 anchor

0.30

Fox

0.28

 anchors

0.27

anchor

0.26

 network

0.25

anchors

0.24

 cable

0.24

Anchor

0.24

chy

0.23

Activations Density 0.146%

mentions of news personalities and their affiliations

No Comments

No Known Activations