INDEX

Explanations

references to specific individuals or entities, including authors and titles

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.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

jni

-0.15

Ø²ÙĬ

-0.14

ank

-0.14

Zi

-0.14

 Bing

-0.14

 Encyclopedia

-0.14

Å¾i

-0.13

.metric

-0.13

org

-0.13

bul

-0.13

POSITIVE LOGITS

monds

0.16

tero

0.16

ptime

0.15

 Gallagher

0.15

ç¶

0.15

:UITableView

0.14

 VÃ¡c

0.14

ÅĻen

0.14

æĸ¹éĿ¢

0.14

pine

0.14

Activations Density 0.009%

references to specific individuals or entities, including authors and titles

No Comments

No Known Activations