INDEX

Explanations

recognition and reference to various languages, particularly South Asian ones

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 12-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L12R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.12.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

ripe

-0.17

INI

-0.15

alse

-0.14

achel

-0.14

Ð°Ð½Ð¸ÑĨ

-0.14

istol

-0.14

hsi

-0.14

aliz

-0.14

ellas

-0.14

avic

-0.14

POSITIVE LOGITS

-speaking

0.26

-language

0.19

 Version

0.16

 versions

0.16

 version

0.16

 verze

0.15

_version

0.15

keit

0.15

-American

0.15

cent

0.14

Activations Density 0.058%

recognition and reference to various languages, particularly South Asian ones

No Comments

No Known Activations