INDEX

Explanations

numerical values and sequences

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

or

-0.17

Bor

-0.16

-0.15

el

-0.15

ine

-0.15

 Holidays

-0.15

bor

-0.15

enton

-0.15

le

-0.14

POSITIVE LOGITS

Ð²Ð¸ÑĩÐ°Ð¹

0.20

kov

0.20

PathComponent

0.16

 plÃ¡

0.16

.owl

0.16

ptune

0.16

ãİ¡

0.15

ãĥ©ãĤ¯

0.15

lien

0.15

åī¯

0.15

Activations Density 0.029%

numerical values and sequences

No Comments

No Known Activations