INDEX

Explanations

references to partiality or division of concepts and influences

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 25-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L25R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.25.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

 solely

-0.19

 mostly

-0.18

 primarily

-0.17

 basically

-0.17

 either

-0.17

 mainly

-0.17

sole

-0.16

mostly

-0.16

 somehow

-0.15

 predominantly

-0.15

POSITIVE LOGITS

 responsible

0.20

due

0.17

due

0.16

 thanks

0.15

 Ð±Ð»Ð°Ð³Ð¾Ð´Ð°ÑĢÑı

0.15

Due

0.15

thanks

0.15

ael

0.15

 responsable

0.15

çĶ±äºİ

0.14

Activations Density 0.036%

references to partiality or division of concepts and influences

No Comments

No Known Activations