INDEX

Explanations

addresses or location-related information

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 20-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L20R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

NZ

-0.16

vvm

-0.15

Australian

-0.14

 Australian

-0.14

 Australians

-0.14

.sax

-0.14

NZ

-0.14

ãĥ¬ãĥ¼

-0.14

Han

-0.14

Ze

-0.14

POSITIVE LOGITS

 Montreal

0.19

 Labrador

0.17

 McGill

0.17

ìĪ

0.16

 Quebec

0.16

rong

0.16

.VisualBasic

0.16

 VÅ¡

0.15

CHIP

0.15

ħ§

0.15

Activations Density 0.242%

addresses or location-related information

No Known Activations