INDEX

Explanations

references to past events or situations

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

endon

-0.18

lei

-0.17

annes

-0.17

åĻ¨

-0.17

fter

-0.15

iams

-0.14

viÄį

-0.14

ered

-0.14

thing

-0.14

phan

-0.14

POSITIVE LOGITS

ebin

0.37

imes

0.34

ime

0.32

iche

0.31

 tense

0.29

ries

0.28

/current

0.28

eur

0.27

ures

0.27

oral

0.25

Activations Density 0.032%

references to past events or situations

No Comments

No Known Activations