INDEX

Explanations

references to prominent individuals or celebrities 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 @ 13-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L13R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.13.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

kart

-0.19

ors

-0.19

als

-0.18

ext

-0.17

extern

-0.16

ester

-0.16

naire

-0.16

eer

-0.16

ering

-0.15

eways

-0.15

POSITIVE LOGITS

red

0.20

fish

0.19

ship

0.19

light

0.17

vation

0.17

let

0.16

0.15

hyth

0.15

lit

0.14

nton

0.14

Activations Density 0.040%

references to prominent individuals or celebrities in various contexts

No Comments

No Known Activations