INDEX

Explanations

names of locations and geographical distinctions

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

soever

-0.15

Ã¡lt

-0.15

ocado

-0.13

.weixin

-0.13

erten

-0.13

gon

-0.13

enden

-0.13

fdc

-0.13

frage

-0.13

áº¯t

-0.13

POSITIVE LOGITS

 home

0.50

 where

0.45

 site

0.43

home

0.38

where

0.36

(home

0.33

site

0.32

 scene

0.32

/home

0.32

-home

0.30

Activations Density 0.201%

names of locations and geographical distinctions

No Comments

No Known Activations