INDEX

Explanations

phrases indicating warnings or cautionary advice

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 28-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L28R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.28.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

ittle

-0.17

oot

-0.15

Answers

-0.15

layan

-0.14

 Hairst

-0.14

á»Ĩ

-0.14

endez

-0.14

 ANSW

-0.13

 NhÃ¢n

-0.13

 Expenses

-0.13

POSITIVE LOGITS

 warning

1.01

 warnings

0.93

 Warning

0.85

 warn

0.83

warning

0.81

Warning

0.78

 warned

0.75

-warning

0.73

warnings

0.72

 Warn

0.72

Activations Density 0.277%

phrases indicating warnings or cautionary advice

No Comments

No Known Activations