INDEX

Explanations

phrases indicating conditional situations or recommendations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 16-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L16R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.16.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

 otherwise

-0.17

 hack

-0.16

 then

-0.16

å®ŀåľ¨

-0.14

 Warm

-0.14

ront

-0.13

NX

-0.13

ZA

-0.13

 train

-0.13

edback

-0.13

POSITIVE LOGITS

worthy

0.21

ãĥ«ãĥĪ

0.16

acted

0.15

olini

0.15

ä¹ĭä¸Ģ

0.15

 INTERRUPTION

0.15

->[

0.15

ãĥ¡ãĥ³ãĥĪ

0.14

shaft

0.14

 Marker

0.14

Activations Density 0.199%

phrases indicating conditional situations or recommendations

No Comments

No Known Activations