INDEX

Explanations

the presence of proper nouns and specific entities related to locations or organizations

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

_revision

-0.15

}}{{

-0.14

 Landing

-0.14

WebHost

-0.14

 Luna

-0.14

èģ

-0.14

ober

-0.14

èħ

-0.13

etty

-0.13

á»¡

-0.13

POSITIVE LOGITS

 Chatt

0.17

 Bengal

0.16

chine

0.15

 cust

0.15

 Samp

0.15

/commons

0.15

 Gram

0.15

 India

0.14

idel

0.14

inois

0.14

Activations Density 0.028%

the presence of proper nouns and specific entities related to locations or organizations

No Comments

No Known Activations