INDEX

Explanations

instances of the word "quick" and its variations, indicating a focus on speed or efficiency

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.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

eof

-0.15

ledi

-0.15

ationally

-0.14

GenerationStrategy

-0.14

 Franti

-0.14

edly

-0.14

illard

-0.14

 çľŁ

-0.14

alcon

-0.13

mtree

-0.13

POSITIVE LOGITS

silver

0.42

sand

0.39

ie

0.36

ening

0.34

ened

0.34

lime

0.29

-fix

0.27

ies

0.25

ens

0.25

fire

0.23

Activations Density 0.024%

instances of the word "quick" and its variations, indicating a focus on speed or efficiency

No Comments

No Known Activations