INDEX

Explanations

instances of the word "publish" and its variants, indicating a focus on publication and publishing activities

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

 Publishers

-0.16

Publisher

-0.16

Published

-0.16

Publication

-0.16

 publications

-0.16

 Publications

-0.16

 purchases

-0.15

Ð¼Ð¾Ð²ÑĸÑĢ

-0.15

 Publication

-0.15

ãĤ©

-0.15

POSITIVE LOGITS

(crate

0.21

ICATION

0.20

houses

0.18

/export

0.18

 online

0.17

/post

0.17

çī©

0.16

ment

0.16

 khoa

0.16

jabi

0.16

Activations Density 0.046%

instances of the word "publish" and its variants, indicating a focus on publication and publishing activities

No Comments

No Known Activations