INDEX

Explanations

references to media items and their accompanying metadata

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 18-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L18R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.18.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

okud

-0.17

olu

-0.16

olik

-0.15

olla

-0.15

ttp

-0.14

okus

-0.14

ymes

-0.13

izzo

-0.13

usu

-0.13

isers

-0.13

POSITIVE LOGITS

è

0.16

noinspection

0.15

.getOwnProperty

0.15

connexion

0.14

 Åĺ

0.14

ahun

0.14

 Ð¿Ð¸ÑĤ

0.14

 Scho

0.13

 Macron

0.13

Î¿ÏĤ

0.13

Activations Density 0.050%

references to media items and their accompanying metadata

No Known Activations