INDEX

Explanations

scientific terminology and concepts related to research studies

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

º

-0.16

Ã½

-0.15

 Creat

-0.15

ÐµÑĢÐº

-0.14

icker

-0.14

uros

-0.14

 Lever

-0.14

æ£Ĵ

-0.14

.bind

-0.14

 Pract

-0.14

POSITIVE LOGITS

 obstruction

0.20

 obstruct

0.19

 functor

0.19

olit

0.17

 adel

0.17

 gadgets

0.17

adel

0.16

-equiv

0.16

 gadget

0.16

 ÄĮech

0.15

Activations Density 0.315%

scientific terminology and concepts related to research studies

No Comments

No Known Activations