INDEX

Explanations

references to law enforcement and missing person investigations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.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

 Daytona

-0.17

orida

-0.16

 Baltimore

-0.16

SEG

-0.16

 Charlottesville

-0.15

gart

-0.15

 Tampa

-0.15

Bam

-0.15

seg

-0.14

 Southampton

-0.14

POSITIVE LOGITS

 Nebraska

0.59

 Omaha

0.57

Om

0.39

braska

0.37

Hus

0.36

 Lincoln

0.35

Neb

0.34

hus

0.30

om

0.29

Ne

0.29

Activations Density 0.036%

references to law enforcement and missing person investigations

No Comments

No Known Activations