INDEX

Explanations

instances of numerical or quantifiable information

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

za

-0.16

ewidth

-0.15

 designed

-0.15

-0.14

app

-0.14

dog

-0.14

isco

-0.14

ith

-0.13

 Canada

-0.13

POSITIVE LOGITS

ledon

0.17

ÙĬÙĨØ©

0.16

åĪ¥

0.16

0.15

 tÄĽ

0.14

å¡

0.14

ÏħÎ¼

0.14

 backpage

0.14

ansom

0.14

readystatechange

0.14

Activations Density 0.053%

instances of numerical or quantifiable information

No Comments

No Known Activations