INDEX

Explanations

mentions of houses and homes in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

stal

-0.16

ufs

-0.16

hone

-0.15

ioc

-0.14

agens

-0.14

 onCancel

-0.14

quo

-0.13

utzer

-0.13

ãĥ³ãĥĹ

-0.13

iyan

-0.13

POSITIVE LOGITS

 house

0.22

OKEN

0.20

å®ħ

0.19

 home

0.19

å®¶

0.18

å®¶çļĦ

0.17

/home

0.16

 hosts

0.15

çĻĤ

0.15

(home

0.15

Activations Density 0.235%

mentions of houses and homes in various contexts

No Comments

No Known Activations