INDEX

Explanations

mentions of historical or political contexts involving trauma and conflict

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

 Hawai

-0.22

mex

-0.21

 Hawaiian

-0.20

 Hawaii

-0.19

_mex

-0.18

 japan

-0.17

 Maui

-0.17

 Jamaica

-0.17

Mex

-0.17

Oregon

-0.17

POSITIVE LOGITS

 Kosovo

0.57

 Serbian

0.51

 Serbia

0.49

 Alban

0.49

Kos

0.46

Ser

0.45

 Albania

0.39

Ser

0.36

 Yugosl

0.36

 Balk

0.34

Activations Density 0.017%

mentions of historical or political contexts involving trauma and conflict

No Comments

No Known Activations