INDEX

Explanations

phrases indicating ease of use or convenience in processes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.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

kovÃ¡

-0.16

opak

-0.15

igm

-0.14

agara

-0.14

ivre

-0.14

asurer

-0.14

ander

-0.13

iores

-0.13

.overflow

-0.13

isoft

-0.13

POSITIVE LOGITS

tap

0.33

 press

0.32

 touch

0.31

 mere

0.31

 simple

0.29

 push

0.29

press

0.29

 taps

0.28

 presses

0.28

mere

0.27

Activations Density 0.115%

phrases indicating ease of use or convenience in processes

No Comments

No Known Activations