INDEX

Explanations

terms related to safety and hazards in consumer products

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 18-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L18R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.18.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

ihan

-0.16

ibble

-0.16

à¥įà¤µà¤°

-0.16

fst

-0.15

Ã¼nd

-0.15

æ±¡

-0.15

ãĤ«ãĥ¼

-0.15

Fraction

-0.14

ÑĤÑĮ

-0.14

 annonces

-0.14

POSITIVE LOGITS

 safety

0.46

 Safety

0.41

Safety

0.38

afety

0.35

 fire

0.34

 unsafe

0.33

 Fire

0.31

 Unsafe

0.30

 safer

0.28

 safe

0.27

Activations Density 0.150%

terms related to safety and hazards in consumer products

No Comments

No Known Activations