INDEX

Explanations

various instances of symbols and operators commonly used in programming or metadata annotations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 4-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L4R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.4.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

-0.17

 Dust

-0.15

 Suggestions

-0.15

ral

-0.14

arians

-0.14

 dust

-0.14

zel

-0.14

ekim

-0.14

io

-0.14

iere

-0.14

POSITIVE LOGITS

 Scient

0.15

"indices

0.15

-alist

0.15

cret

0.15

cente

0.14

Î»Ïİ

0.14

ãĥ¡ãĥ³ãĥĪ

0.14

ereum

0.14

 ìĥĿ

0.14

abin

0.13

Activations Density 0.005%

various instances of symbols and operators commonly used in programming or metadata annotations

No Comments

No Known Activations