INDEX

Explanations

locations and physical spaces within the narrative

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

 places

-0.15

Ã¥n

-0.15

coh

-0.15

 cole

-0.14

ÑĪÐ¸

-0.14

 everywhere

-0.14

 Places

-0.14

 eiusmod

-0.14

zb

-0.13

assi

-0.13

POSITIVE LOGITS

 area

0.27

 complex

0.26

 compound

0.23

 where

0.22

-area

0.20

 entrance

0.20

 grounds

0.19

compound

0.19

-turned

0.19

(where

0.18

Activations Density 0.361%

locations and physical spaces within the narrative

No Comments

No Known Activations