INDEX

Explanations

phrases indicating a final summary or conclusion

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

unk

-0.16

adu

-0.15

.LogWarning

-0.15

IEL

-0.15

-0.14

iel

-0.14

avana

-0.14

 Ð¸Ñģ

-0.14

pesan

-0.14

 Parr

-0.14

POSITIVE LOGITS

 Bottom

0.28

bottom

0.27

 bottom

0.26

 BOTTOM

0.26

.Bottom

0.25

(bottom

0.24

Bottom

0.24

-bottom

0.22

BOTTOM

0.21

/top

0.20

Activations Density 0.017%

phrases indicating a final summary or conclusion

No Comments

No Known Activations