INDEX

Explanations

instances of planning and future intentions

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

Ø§ÙĩÙħ

-0.15

à¸ļà¸ģ

-0.15

lesen

-0.15

ubo

-0.14

rez

-0.14

ottenham

-0.14

amework

-0.14

è¸

-0.14

lias

-0.14

agma

-0.14

POSITIVE LOGITS

 ahead

0.25

ahead

0.23

Ahead

0.22

 Ahead

0.22

-ahead

0.18

 accordingly

0.16

 eventually

0.16

 soon

0.16

 Accordingly

0.16

 eventual

0.15

Activations Density 0.019%

instances of planning and future intentions

No Comments

No Known Activations