INDEX

Explanations

references to television networks and programming deals

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

cih

-0.18

NSK

-0.16

nosti

-0.15

ibbon

-0.15

 CascadeType

-0.15

ÑĥÑĪ

-0.14

oro

-0.14

ghi

-0.14

arih

-0.14

 backups

-0.14

POSITIVE LOGITS

 cable

0.32

 Cable

0.31

 programming

0.31

Vi

0.29

 Programming

0.29

 stations

0.28

 station

0.27

Programming

0.26

 entertainment

0.26

 Stations

0.26

Activations Density 0.154%

references to television networks and programming deals

No Comments

No Known Activations