INDEX

Explanations

conjunctions and connecting words in text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 20-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L20R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.20.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

OTA

-0.16

apper

-0.15

amment

-0.14

anship

-0.14

Ø§ÙģÙĩ

-0.14

inkel

-0.14

Ã¡vÄĽ

-0.14

ucken

-0.13

Legendary

-0.13

arlar

-0.13

POSITIVE LOGITS

 holder

0.23

 member

0.22

 part

0.21

an

0.21

 provider

0.20

 recipient

0.20

 winner

0.20

0.19

 someone

0.18

 responsible

0.18

Activations Density 0.202%

conjunctions and connecting words in text

No Comments

No Known Activations