INDEX

Explanations

references to geographic locations and regions

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

ØºÙĦ

-0.13

bout

-0.13

reta

-0.12

peria

-0.12

POCH

-0.12

ording

-0.12

 Downtown

-0.12

kel

-0.12

ory

-0.12

 mÄ±

-0.12

POSITIVE LOGITS

 cities

0.35

 regions

0.33

 communities

0.33

 counties

0.31

-wide

0.30

/world

0.29

 markets

0.29

 region

0.29

 towns

0.29

 suburbs

0.28

Activations Density 0.208%

references to geographic locations and regions

No Comments

No Known Activations