INDEX

Explanations

references to the concept of "common" or shared elements across different contexts or subjects

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 23-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L23R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.23.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

 Shack

-0.17

amer

-0.16

idel

-0.15

yr

-0.15

oul

-0.15

aping

-0.14

éĻĲå®ļ

-0.14

ged

-0.14

ItemClick

-0.14

à¸«à¸Ļ

-0.14

POSITIVE LOGITS

wealth

0.52

ality

0.38

 denominator

0.34

 sense

0.30

sense

0.29

alty

0.25

places

0.25

place

0.25

 Sense

0.25

alties

0.24

Activations Density 0.030%

references to the concept of "common" or shared elements across different contexts or subjects

No Comments

No Known Activations