INDEX

Explanations

phrases that relate to simplifying tasks and making processes easier

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

anners

-0.19

mad

-0.17

ìĹ´

-0.15

mad

-0.15

IXEL

-0.14

adox

-0.14

forums

-0.13

chemes

-0.13

Ã±o

-0.13

Ì£

-0.13

POSITIVE LOGITS

.ali

0.16

 ease

0.16

ulur

0.16

 convenience

0.15

erule

0.15

aminer

0.15

iasi

0.14

yat

0.14

/qu

0.14

Activations Density 0.118%

phrases that relate to simplifying tasks and making processes easier

No Comments

No Known Activations