INDEX

Explanations

references to Lego sets and their components

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 16-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L16R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.16.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

orris

-0.16

iez

-0.15

ãĤµãĤ¤

-0.15

iban

-0.14

xcf

-0.14

çĦ¼

-0.14

èķ

-0.14

è½®

-0.14

lassen

-0.14

ordion

-0.13

POSITIVE LOGITS

 LEGO

0.20

å±

0.17

 Lego

0.16

 prec

0.16

enberg

0.16

untime

0.16

 Klan

0.14

ylan

0.14

 Stores

0.14

 bricks

0.14

Activations Density 0.016%

references to Lego sets and their components

No Comments

No Known Activations