INDEX

Explanations

measurements of area and density

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

egers

-0.16

Dys

-0.15

sat

-0.15

Î´ÎµÎ¹

-0.15

allah

-0.14

unders

-0.14

gings

-0.14

ynÄ±

-0.14

sink

-0.14

AGER

-0.14

POSITIVE LOGITS

Â²

0.28

 ÙħØ±Ø¨Ø¹

0.22

0.21

 vuÃ´ng

0.19

 Äįt

0.18

 squared

0.18

xmm

0.17

ÏįÎ½

0.17

 squares

0.16

0.15

Activations Density 0.010%

measurements of area and density

No Comments

No Known Activations