INDEX

Explanations

references to specific geographic locations and cities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

Åĵ

-0.15

odega

-0.14

ä¸Ģçº§

-0.14

Ð¾Ð²ÐµÑĢ

-0.14

Ð¼ÐµÑĩ

-0.13

 Loren

-0.13

 aktu

-0.13

adro

-0.13

 trans

-0.13

çĬ

-0.13

POSITIVE LOGITS

tember

0.16

$MESS

0.15

iyel

0.15

 ADVISED

0.14

ÏĢÎ»

0.14

elm

0.14

 JetBrains

0.14

isten

0.14

itten

0.14

ustin

0.14

Activations Density 0.055%

references to specific geographic locations and cities

No Comments

No Known Activations