INDEX

Explanations

references to searching or seeking, particularly for lost items or locations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

acus

-0.15

abal

-0.14

 Learned

-0.14

 taraf

-0.14

ico

-0.14

ÙıÙĪ

-0.14

å®Ļ

-0.14

 bondage

-0.13

 powerless

-0.13

POSITIVE LOGITS

 lost

0.36

 hidden

0.35

 buried

0.34

 answers

0.32

lost

0.31

hidden

0.31

 elusive

0.29

 missing

0.29

 Lost

0.29

 treasure

0.28

Activations Density 0.215%

references to searching or seeking, particularly for lost items or locations

No Comments

No Known Activations