INDEX

Explanations

references to specific locations or countries

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

iaux

-0.18

 Cork

-0.17

 McCartney

-0.17

 Missouri

-0.17

 Maine

-0.17

 Belfast

-0.17

 Vermont

-0.16

 Kansas

-0.16

 Texas

-0.16

 Mississippi

-0.16

POSITIVE LOGITS

 Indonesian

0.59

 Indonesia

0.58

 Jakarta

0.57

 Indones

0.56

 indonesia

0.42

 Bali

0.40

 Java

0.35

Jak

0.35

 Papua

0.35

Rp

0.34

Activations Density 0.101%

references to specific locations or countries

No Comments

No Known Activations