INDEX

Explanations

technical descriptions and metadata related to documents or coding

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.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

igh

-0.17

Sez

-0.15

 Feld

-0.15

 Glover

-0.15

á»¹

-0.15

hai

-0.15

agu

-0.15

aggi

-0.14

 Activation

-0.14

ystick

-0.14

POSITIVE LOGITS

wr

0.16

IBUT

0.16

 Rapids

0.15

WR

0.15

aven

0.15

 Ð²Ð¸Ð½

0.15

atta

0.15

 basket

0.15

 Orch

0.14

åĻ

0.14

Activations Density 0.016%

technical descriptions and metadata related to documents or coding

No Comments

No Known Activations