INDEX

Explanations

words and phrases that indicate speakers, roles, or significant individuals in a context

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

enus

-0.15

aseline

-0.14

 ranging

-0.14

odos

-0.14

NOWLED

-0.13

zcze

-0.13

·»

-0.13

lul

-0.13

 Ð¿ÑĢÐ¸Ñĩ

-0.13

POSITIVE LOGITS

çļĦæĺ¯

0.23

å°±æĺ¯

0.21

 Ø¹Ø¨Ø§Ø±Øª

0.20

 include

0.19

 happens

0.18

include

0.17

is

0.17

 adalah

0.16

ãģ®ãģĮ

0.16

Ð¼Ð¾ÑĢ

0.15

Activations Density 0.119%

words and phrases that indicate speakers, roles, or significant individuals in a context

No Comments

No Known Activations