INDEX

Explanations

instances of the word "in" as well as formatting elements indicating sections or chapters in a text

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

fg

-0.16

hurst

-0.15

uble

-0.15

ymoon

-0.15

Ð¼ÐµÑĩ

-0.14

ung

-0.14

(éĩĳ

-0.14

#ga

-0.14

duck

-0.14

itudes

-0.14

POSITIVE LOGITS

Î¼Î¹

0.17

0.16

ohn

0.16

enc

0.15

reator

0.15

emez

0.15

0.14

erval

0.14

Ã£

0.14

argest

0.14

Activations Density 0.021%

instances of the word "in" as well as formatting elements indicating sections or chapters in a text

No Comments

No Known Activations