INDEX

Explanations

phrases relating to permission and consent

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

avy

-0.15

lue

-0.15

pz

-0.14

znik

-0.14

pell

-0.14

Ä©

-0.14

arshal

-0.14

olk

-0.14

KI

-0.14

ena

-0.13

POSITIVE LOGITS

 permission

0.73

permission

0.57

 Permission

0.57

 permissions

0.52

 consent

0.51

 PERMISSION

0.49

Permission

0.48

 approval

0.46

_permission

0.44

 Permissions

0.44

Activations Density 0.139%

phrases relating to permission and consent

No Comments

No Known Activations