INDEX

Explanations

references to disciplinary actions and procedures within educational contexts

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

yw

-0.15

åĢĴ

-0.15

 legit

-0.14

untu

-0.14

blick

-0.14

íĳ¸

-0.14

ìĤ°

-0.14

weg

-0.13

orado

-0.13

½æķ°

-0.13

POSITIVE LOGITS

akan

0.15

ingleton

0.15

mana

0.15

ValueCollection

0.14

amina

0.14

 Colleg

0.14

ONUS

0.14

 Brandon

0.14

onus

0.14

insula

0.14

Activations Density 0.097%

references to disciplinary actions and procedures within educational contexts

No Comments

No Known Activations