INDEX

Explanations

concepts related to understanding, analysis, and systematic exploration of issues

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

caff

-0.14

-wow

-0.14

 eiusmod

-0.13

ytt

-0.13

è¿Ļæł·çļĦ

-0.13

ANY

-0.13

 Ð´ÐµÐ»Ð¾

-0.12

IFA

-0.12

ä¸Ģå®ļ

-0.12

raki

-0.12

POSITIVE LOGITS

how

0.71

why

0.68

how

0.54

why

0.50

 cÃ³mo

0.43

 what

0.41

å¦Ĥä½ķ

0.41

ä¸ºä»Ģä¹Ī

0.41

 exactly

0.40

 where

0.39

Activations Density 0.483%

concepts related to understanding, analysis, and systematic exploration of issues

No Comments

No Known Activations