INDEX

Explanations

mentions of people involved in significant incidents or cases

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.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

 luder

-0.16

mime

-0.15

 Morse

-0.14

 Mickey

-0.14

epam

-0.14

.Interop

-0.14

meth

-0.14

omor

-0.14

mount

-0.14

mons

-0.13

POSITIVE LOGITS

Mal

1.63

mal

1.52

Mal

1.51

mal

1.37

MAL

1.34

MAL

1.22

 Malcolm

1.06

 ÐľÐ°Ð»

0.96

 Malta

0.90

 ÙħØ§ÙĦ

0.88

Activations Density 0.176%

mentions of people involved in significant incidents or cases

No Comments

No Known Activations