INDEX

Explanations

strong affirmations and superlatives describing people or experiences

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

isko

-0.15

 ÙħÙĤØ¯

-0.14

acobian

-0.14

ruk

-0.14

 ä»»

-0.14

asl

-0.14

isky

-0.14

logan

-0.13

.mvp

-0.13

weg

-0.13

POSITIVE LOGITS

gem

0.36

Gem

0.30

 treat

0.30

Gem

0.30

gem

0.27

 winner

0.27

 pleasure

0.26

 gems

0.25

 blast

0.24

ces

0.24

Activations Density 0.163%

strong affirmations and superlatives describing people or experiences

No Comments

No Known Activations