INDEX

Explanations

expressions related to going above and beyond in service or effort

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

æĪĲ

-0.16

icha

-0.16

adle

-0.15

aised

-0.14

Ð¾Ð²Ð°ÑĤÑĮÑģÑı

-0.14

ãĥ©ãĥĥãĤ¯

-0.14

onu

-0.14

gÃ¼n

-0.14

enna

-0.14

awl

-0.13

POSITIVE LOGITS

 extra

0.44

 EXTRA

0.35

 above

0.35

-extra

0.35

extra

0.34

 Extra

0.34

 Above

0.34

above

0.32

Extra

0.29

 ABOVE

0.29

Activations Density 0.019%

expressions related to going above and beyond in service or effort

No Comments

No Known Activations