INDEX

Explanations

code structure and formatting, specifically related to functions and test cases in programming

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

æľĭ

-0.16

 Ukra

-0.15

shortcode

-0.15

ymi

-0.15

Ã³r

-0.15

asca

-0.14

retty

-0.14

adge

-0.14

.BackgroundColor

-0.14

é¹

-0.14

POSITIVE LOGITS

 done

0.19

done

0.18

arness

0.16

(done

0.16

 Done

0.16

igu

0.15

ëł¤

0.15

 artificially

0.15

 scenario

0.15

(()

0.15

Activations Density 0.010%

code structure and formatting, specifically related to functions and test cases in programming

No Comments

No Known Activations