INDEX

Explanations

references to Los Angeles

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 27-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L27R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.27.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

ocop

-0.17

eer

-0.16

iqueta

-0.15

tti

-0.15

backs

-0.14

uelle

-0.14

auss

-0.14

ault

-0.14

icina

-0.14

ateau

-0.14

POSITIVE LOGITS

 Angeles

0.41

ange

0.28

ANGE

0.27

 ange

0.26

-An

0.22

 Ange

0.21

lassen

0.20

 angles

0.20

ANGLES

0.20

 Alam

0.19

Activations Density 0.006%

references to Los Angeles

No Comments

No Known Activations