INDEX

Explanations

phrases emphasizing significance or noteworthy information

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.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

-0.15

æ´ĭ

-0.15

.unlink

-0.14

ä¸ĭåİ»

-0.14

ades

-0.14

 hints

-0.14

.PerformLayout

-0.13

tÃŃ

-0.13

ìľ¨

-0.13

gue

-0.13

POSITIVE LOGITS

 note

0.40

 stress

0.34

 noted

0.34

 remember

0.33

note

0.32

 stressed

0.31

 Note

0.31

Note

0.30

remember

0.30

 point

0.29

Activations Density 0.129%

phrases emphasizing significance or noteworthy information

No Comments

No Known Activations