INDEX

Explanations

names of individuals and references related to personal information or achievements

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

Å®

-0.14

Ã¡bado

-0.14

-0.13

Ì§

-0.12

ushima

-0.12

ÇĲ

-0.12

ubbo

-0.12

detalle

-0.11

ï»¿

-0.11

postalcode

-0.11

POSITIVE LOGITS

/***************************************************************************↵

0.12

Slf

0.12

/****************************************************************************↵

0.12

/*****************************************************************************↵

0.12

Ð°ÑĢÐ°Ð¼

0.10

 treff

0.10

 prostitu

0.10

IsRequired

0.10

 Mitar

0.10

 trai

0.10

Activations Density 0.923%

names of individuals and references related to personal information or achievements

No Comments

No Known Activations