INDEX

Explanations

references to geographic locations or administrative divisions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.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

entes

-0.16

ibaba

-0.16

ensi

-0.16

oyal

-0.15

aes

-0.15

endi

-0.14

è¢ĸ

-0.14

 PureComponent

-0.14

isin

-0.14

POSITIVE LOGITS

ucer

0.16

Ð¾ÑĢÐ¾Ð²

0.15

 belong

0.15

 Forge

0.15

 belonging

0.14

comed

0.14

Forge

0.14

 belongs

0.14

å´İ

0.14

Zy

0.14

Activations Density 0.202%

references to geographic locations or administrative divisions

No Comments

No Known Activations