INDEX

Explanations

details related to recipes and cooking methods

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

 Ranch

-0.17

 brush

-0.17

fruit

-0.16

 Brush

-0.16

 Mast

-0.15

 Bart

-0.14

Pie

-0.14

 clipboard

-0.14

 Orchard

-0.14

 éĤ

-0.14

POSITIVE LOGITS

 rice

0.42

 Rice

0.41

rice

0.35

 grains

0.25

RVA

0.24

 ÑĢÐ¸Ñģ

0.23

bas

0.21

ç±³

0.21

 Arbor

0.20

 cÆ¡m

0.19

Activations Density 0.029%

details related to recipes and cooking methods

No Comments

No Known Activations

details related to recipes and cooking methods

No Comments

No Known Activations