INDEX

Explanations

references to quantities or types of items, especially in a vague or generalized context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 22-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L22R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.22.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

for

-0.21

 untuk

-0.18

 fÃ¼r

-0.17

aro

-0.17

for

-0.16

 voor

-0.15

for

-0.15

ä¸ºäºĨ

-0.15

aub

-0.15

hen

-0.15

POSITIVE LOGITS

 purposes

0.69

 sake

0.60

 reasons

0.42

 purpose

0.35

 reason

0.35

reason

0.29

 PURPOSE

0.27

Purpose

0.26

purpose

0.26

Reason

0.25

Activations Density 0.188%

references to quantities or types of items, especially in a vague or generalized context

No Comments

No Known Activations

references to quantities or types of items, especially in a vague or generalized context

No Comments

No Known Activations