INDEX

Explanations

phrases that indicate hardship or challenging circumstances

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

oya

-0.16

agher

-0.16

ame

-0.16

autos

-0.15

']!='

-0.14

ackage

-0.14

umn

-0.14

steen

-0.14

rott

-0.14

olume

-0.13

POSITIVE LOGITS

 trying

0.40

 difficult

0.38

trying

0.36

 Trying

0.35

 times

0.35

 tough

0.35

 challenging

0.34

 uncertain

0.32

Trying

0.31

 dark

0.28

Activations Density 0.187%

phrases that indicate hardship or challenging circumstances

No Comments

No Known Activations