INDEX

Explanations

descriptors of styles or types, particularly in relation to character attributes or appearances

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.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

 Ð¿Ð¾ÐºÐ¾Ð»

-0.17

ela

-0.15

 Ð¾Ð±ÑĢÐ°Ð·Ð¾Ð²Ð°Ð½Ð¸Ñı

-0.15

ila

-0.15

 Ð¿ÑĢÐ¸Ð½ÑĨÐ¸Ð¿

-0.15

Ñĭ

-0.14

 Ð¾ÑĤÐ½Ð¾ÑĪÐµÐ½Ð¸Ñı

-0.14

 somehow

-0.14

ÑģÐºÐ¾Ðµ

-0.14

POSITIVE LOGITS

 Playlist

0.16

aits

0.16

Î»Î¹Î±

0.15

 Minute

0.15

loy

0.15

 Mehr

0.15

ÑĪÐµÐ¼

0.14

 specular

0.14

 superficial

0.14

reserve

0.14

Activations Density 0.017%

descriptors of styles or types, particularly in relation to character attributes or appearances

No Comments

No Known Activations