INDEX

Explanations

technical terms and references related to software and programming

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.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

Rew

-0.17

Ð²ÑģÑı

-0.16

ÑĤÑĥ

-0.16

ret

-0.16

RET

-0.15

edback

-0.15

zet

-0.14

rete

-0.14

uze

-0.14

POSITIVE LOGITS

 later

0.29

later

0.27

 Later

0.27

Later

0.25

 spÃ¤ter

0.20

 pozdÄĽ

0.17

ober

0.17

.copyOf

0.17

åĲİ

0.16

 afterwards

0.16

Activations Density 0.020%

technical terms and references related to software and programming

No Comments

No Known Activations