INDEX

Explanations

phrases and indicators related to updates and modifications in content

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 12-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L12R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.12.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

agram

-0.17

esti

-0.16

ãĥĥãĥĦ

-0.15

inst

-0.15

rav

-0.14

osing

-0.14

_alias

-0.14

ulp

-0.14

aurant

-0.13

Iss

-0.13

POSITIVE LOGITS

DDevice

0.16

istrovstvÃŃ

0.15

á»Ĳ

0.15

asje

0.15

RuntimeObject

0.14

 Alexand

0.14

dÄ±ÄŁÄ±

0.14

nish

0.14

ìĿ´ìĸ´

0.14

Activations Density 0.005%

phrases and indicators related to updates and modifications in content

No Comments

No Known Activations