INDEX

Explanations

references to indicators of strength and evidence of conditions or phenomena

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.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

Visibility

-0.15

ãģĤãģĴ

-0.15

èªī

-0.15

OTTOM

-0.14

Ùĩ

-0.14

 Silence

-0.14

 ÑĩÐ¸ÑģÐ»Ðµ

-0.14

variants

-0.13

 Visibility

-0.13

Retention

-0.13

POSITIVE LOGITS

how

0.31

why

0.27

 intent

0.22

how

0.21

 where

0.21

 something

0.21

omething

0.20

 commitment

0.20

why

0.19

 either

0.18

Activations Density 0.171%

references to indicators of strength and evidence of conditions or phenomena

No Comments

No Known Activations