INDEX

Explanations

references to classifications and categories of people, organizations, or items

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

issing

-0.14

iefs

-0.14

Fro

-0.14

annis

-0.13

rim

-0.13

 Twist

-0.13

lename

-0.13

isque

-0.13

ĶĦ

-0.13

áº§n

-0.12

POSITIVE LOGITS

 category

0.77

 categories

0.61

category

0.59

 Category

0.58

-category

0.54

Category

0.53

 CATEGORY

0.50

 ÐºÐ°ÑĤÐµÐ³Ð¾ÑĢ

0.49

.category

0.49

 categorie

0.48

Activations Density 0.177%

references to classifications and categories of people, organizations, or items

No Comments

No Known Activations