INDEX

Explanations

terms related to qualifications and conditions for actions or events

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

Con

-0.28

Con

-0.17

-Con

-0.17

 conjug

-0.16

Å¡it

-0.15

Thu

-0.15

chu

-0.15

 Cone

-0.15

ufs

-0.14

ube

-0.14

POSITIVE LOGITS

icon

0.20

on

0.19

eon

0.19

cons

0.18

icons

0.18

-k

0.18

 Ú©Ø§ÙĨ

0.18

oon

0.18

-cons

0.17

å»ºè¨Ń

0.17

Activations Density 0.062%

terms related to qualifications and conditions for actions or events

No Comments

No Known Activations