INDEX

Explanations

words and phrases related to granting permissions or rights

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

yun

-0.16

oster

-0.16

iao

-0.15

_mac

-0.15

aida

-0.14

_MAC

-0.14

legen

-0.14

mina

-0.14

ksi

-0.14

è³Ģ

-0.14

POSITIVE LOGITS

ee

0.19

 permission

0.18

 access

0.17

REE

0.16

antee

0.16

ree

0.15

 kepada

0.15

Gur

0.14

permission

0.14

itions

0.14

Activations Density 0.046%

words and phrases related to granting permissions or rights

No Comments

No Known Activations