INDEX

Explanations

terms related to individuals or entities that play a role in various processes, such as donors, patients, users, and players

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 12-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L12R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.12.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

áº»

-0.15

Cr

-0.15

ITO

-0.14

apiro

-0.14

maal

-0.14

olley

-0.14

daf

-0.14

unto

-0.14

azel

-0.14

/***/

-0.13

POSITIVE LOGITS

(s

0.23

 YÃ¼z

0.16

(es

0.15

 herself

0.14

 SÃ¶z

0.14

.inc

0.14

åıĬåħ¶

0.14

 Explosion

0.13

sea

0.13

andas

0.13

Activations Density 0.135%

terms related to individuals or entities that play a role in various processes, such as donors, patients, users, and players

No Comments

No Known Activations