INDEX

Explanations

proper nouns and significant identifiers

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 21-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L21R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.21.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

ãĤ¿ãĥ«

-0.15

 Cain

-0.14

Â´t

-0.14

 ï¿½

-0.14

ape

-0.14

ä¹İ

-0.14

(es

-0.14

spb

-0.12

 nabÃŃd

-0.12

.parallel

-0.12

POSITIVE LOGITS

's

0.56

’s

0.48

çļĦ

0.40

ìĿĺ

0.35

ãģ®

0.31

çļĦæĥħ

0.28

Â´s

0.28

'S

0.27

ãģ®å¤§

0.27

çļĦå¤§

0.26

Activations Density 0.001%

proper nouns and significant identifiers

No Comments

No Known Activations