INDEX

Explanations

questions related to health and wellness topics

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 31-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L31R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.31.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

 somehow

-0.17

odate

-0.14

orna

-0.13

liches

-0.13

vido

-0.13

 mÄ±

-0.13

 belki

-0.13

lest

-0.13

 occasional

-0.13

POSITIVE LOGITS

 exactly

0.66

 Exactly

0.52

 precisely

0.43

Exactly

0.41

 anyway

0.39

 anyways

0.37

 exact

0.36

 really

0.35

 genau

0.33

 supposed

0.33

Activations Density 0.128%

questions related to health and wellness topics

No Comments

No Known Activations