INDEX

Explanations

instances of combining or blending different elements or concepts

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

quete

-0.16

æ¼

-0.16

gba

-0.15

alytics

-0.15

lung

-0.15

hof

-0.14

SearchTree

-0.14

ÏħÎ³

-0.14

ltk

-0.14

iode

-0.14

POSITIVE LOGITS

 elements

0.27

 tradition

0.24

 traditional

0.23

elements

0.22

old

0.20

 together

0.20

Old

0.19

 aspects

0.18

 conventional

0.18

 Elements

0.18

Activations Density 0.123%

instances of combining or blending different elements or concepts

No Comments

No Known Activations