INDEX

Explanations

variations of the word "perturb" and related terms

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 27-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L27R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.27.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

erland

-0.23

iesz

-0.20

andan

-0.17

alet

-0.16

ctp

-0.15

yne

-0.15

lopen

-0.15

erate

-0.15

iffe

-0.15

osten

-0.15

POSITIVE LOGITS

urb

0.44

urbed

0.41

urbation

0.41

 pert

0.38

 Pert

0.36

pert

0.30

uss

0.25

ur

0.24

ub

0.23

 distur

0.22

Activations Density 0.004%

variations of the word "perturb" and related terms

No Comments

No Known Activations