INDEX

Explanations

place names and geographical references

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

.nz

-0.20

rlen

-0.17

RVA

-0.17

NZ

-0.15

Ī

-0.15

áº¤

-0.15

NZ

-0.15

okie

-0.15

uteÄį

-0.15

ç¼ĺ

-0.14

POSITIVE LOGITS

 California

0.40

 Texas

0.35

 Arizona

0.35

 Colorado

0.34

 Florida

0.34

 Pennsylvania

0.33

 Illinois

0.33

 Minnesota

0.32

 Michigan

0.31

California

0.30

Activations Density 0.145%

place names and geographical references

No Comments

No Known Activations

place names and geographical references

No Comments

No Known Activations