INDEX

Explanations

references to the concept of free speech

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.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

stro

-0.17

ivet

-0.16

stag

-0.15

errupted

-0.15

drv

-0.15

lights

-0.15

urally

-0.15

æĺĩ

-0.15

çĶ²

-0.14

riad

-0.14

POSITIVE LOGITS

-wheel

0.27

fall

0.25

edom

0.25

 speech

0.24

boot

0.24

floating

0.24

 enterprise

0.23

-enter

0.22

-market

0.22

hold

0.22

Activations Density 0.025%

references to the concept of free speech

No Comments

No Known Activations