INDEX

Explanations

instances of the symbol 'ðŁ' and character sequences that indicate narrative elements or actions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 31-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L31R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.31.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

JT

-0.15

 Japanese

-0.15

Thai

-0.15

ãĢģæĹ¥æľ¬

-0.15

Japanese

-0.14

owler

-0.14

JNI

-0.14

mrb

-0.14

teb

-0.14

Interpreter

-0.14

POSITIVE LOGITS

 magic

0.24

 witches

0.23

 Magic

0.22

 magical

0.22

 Magical

0.22

 Fairy

0.21

éŃĶæ³ķ

0.20

Magic

0.19

magic

0.19

éŃĶ

0.19

Activations Density 0.010%

instances of the symbol 'ðŁ' and character sequences that indicate narrative elements or actions

No Comments

No Known Activations