INDEX

Explanations

terms related to focus and attention in various contexts

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

 afirm

-0.14

Oi

-0.14

Î¯Î½Î±

-0.13

axe

-0.13

ANK

-0.13

,'#

-0.13

 konk

-0.13

 ServiceException

-0.13

thren

-0.13

POSITIVE LOGITS

queeze

0.17

ostringstream

0.16

rowse

0.15

Î¸Î¹

0.15

 SÄ±r

0.15

onda

0.15

áº¡n

0.15

anda

0.15

enheim

0.15

EMPL

0.14

Activations Density 0.005%

terms related to focus and attention in various contexts

No Comments

No Known Activations