INDEX

Explanations

information related to geographic locations and street addresses

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

Chicago

-0.18

 chicago

-0.17

.EventHandler

-0.16

 Chicago

-0.16

stanbul

-0.15

 Heller

-0.15

 Tulsa

-0.15

yu

-0.15

raÅ¾

-0.15

inho

-0.14

POSITIVE LOGITS

NB

0.38

 Freder

0.31

NB

0.31

 Edmund

0.28

 Shed

0.26

New

0.25

nb

0.25

 Saint

0.25

_NB

0.25

nb

0.25

Activations Density 0.058%

information related to geographic locations and street addresses

No Comments

No Known Activations