INDEX

Explanations

statements regarding existence or being in various states

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 21-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L21R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.21.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

Ã¤ter

-0.15

ENÃį

-0.15

asic

-0.14

aris

-0.14

arness

-0.14

TriState

-0.14

.TO

-0.14

renom

-0.14

 sido

-0.14

POSITIVE LOGITS

 able

0.30

 unable

0.23

 revealed

0.21

 born

0.18

 Unable

0.18

 replaced

0.18

 transformed

0.17

 Able

0.17

 activated

0.17

unable

0.17

Activations Density 0.425%

statements regarding existence or being in various states

No Comments

No Known Activations