INDEX

Explanations

phrases related to notable achievements or prominent subjects

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 20-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L20R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.20.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

otten

-0.18

 Shapiro

-0.18

andom

-0.16

Parms

-0.15

aven

-0.15

 Yates

-0.14

angl

-0.14

 SETUP

-0.13

 disproportionate

-0.13

emez

-0.13

POSITIVE LOGITS

Ð¼Ð¾ÑĢ

0.16

 Colomb

0.15

ONO

0.14

arie

0.14

gem

0.14

KI

0.14

rem

0.14

aks

0.14

atrix

0.14

 Champ

0.13

Activations Density 0.001%

phrases related to notable achievements or prominent subjects

No Comments

No Known Activations