INDEX

Explanations

phrases related to actions of taking or similar activities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

uhe

-0.15

ucas

-0.15

.tc

-0.14

 Berger

-0.14

ippo

-0.14

edback

-0.14

opup

-0.14

uet

-0.14

-0.13

 ÐĵÑĢÐ¸

-0.13

POSITIVE LOGITS

aim

0.30

 center

0.26

 part

0.23

 centre

0.21

Aim

0.20

-to

0.19

 Center

0.19

 stage

0.19

 control

0.18

-over

0.18

Activations Density 0.034%

phrases related to actions of taking or similar activities

No Comments

No Known Activations