INDEX

Explanations

complete abstract or title sections of academic papers or reports

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 2-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L2R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.2.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

elerik

-0.14

mitt

-0.14

arya

-0.14

è¨ĢãģĨ

-0.14

agi

-0.14

itzer

-0.14

ĩ

-0.14

á»įn

-0.14

ciler

-0.14

åºľ

-0.13

POSITIVE LOGITS

 Shank

0.15

issen

0.15

GDK

0.14

 Commons

0.14

tober

0.14

 Triangle

0.14

 project

0.13

 Kare

0.13

holds

0.13

 Lambert

0.13

Activations Density 0.151%

complete abstract or title sections of academic papers or reports

No Comments

No Known Activations