INDEX

Explanations

terms related to individual contributions and personal experiences

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

thus

-0.15

æ¨¹

-0.14

atsapp

-0.14

.eng

-0.14

 thus

-0.14

logan

-0.14

istant

-0.14

ÃŃÅ¾

-0.14

apeake

-0.14

ryn

-0.14

POSITIVE LOGITS

alone

0.31

 personal

0.30

 independently

0.29

 alone

0.28

 solo

0.28

 contribution

0.28

 independ

0.28

own

0.27

 independent

0.26

 Independ

0.25

Activations Density 0.012%

terms related to individual contributions and personal experiences

No Comments

No Known Activations