INDEX

Explanations

punctuation marks that indicate a strong emotional response

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

-0.18

hs

-0.14

Ø±ÙĪÙħ

-0.14

icide

-0.13

trag

-0.13

ysi

-0.13

 Ð¿Ð¸ÑĤÐ°Ð½Ð¸Ñı

-0.13

ichi

-0.13

.tw

-0.13

responsive

-0.13

POSITIVE LOGITS

ernes

0.17

vos

0.15

 Leisure

0.15

Opens

0.14

enza

0.14

ffer

0.14

Khu

0.14

Ð¾Ð¶ÐµÑĤ

0.14

 éĺ

0.14

OUS

0.13

Activations Density 0.000%

punctuation marks that indicate a strong emotional response

No Comments

No Known Activations