INDEX

Explanations

negations and questions about preferences or intentions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

.central

-0.14

Gatt

-0.14

GV

-0.14

 ÐºÐ¾ÑģÑĤ

-0.14

.scalablytyped

-0.14

canf

-0.14

atis

-0.14

uke

-0.14

ove

-0.14

inati

-0.14

POSITIVE LOGITS

 wouldn

0.23

 want

0.20

 mind

0.20

 dream

0.20

 rather

0.19

mind

0.17

 trade

0.17

nt

0.17

 Dream

0.17

æ¢¦

0.17

Activations Density 0.069%

negations and questions about preferences or intentions

No Comments

No Known Activations