INDEX

Explanations

words and phrases related to specific characteristics or qualities

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

 upstream

-0.17

amus

-0.15

 ÑĥÐ³Ð¾Ð´

-0.15

suc

-0.15

à¸Ĺà¸£

-0.14

 spect

-0.14

hus

-0.14

 flank

-0.14

 datum

-0.14

ado

-0.14

POSITIVE LOGITS

 ÑįÑĦÑĦÐµÐº

0.18

 Angiosper

0.17

ÑģÑĤÐ²Ð¾ÑĢ

0.16

 ÐºÐ¾Ð»Ð¸ÑĩÐµ

0.15

 ÑģÐ¾ÑħÑĢÐ°

0.15

hoa

0.15

zim

0.15

 Ð¼Ð½Ð¾Ð¶Ðµ

0.15

Ø§Ø³Ø·Ø©

0.15

 ÑĥÐ¼ÐµÐ½ÑĮ

0.15

Activations Density 0.005%

words and phrases related to specific characteristics or qualities

No Comments

No Known Activations