INDEX

Explanations

features related to user customization and content sharing capabilities in applications

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 16-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L16R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.16.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

sher

-0.16

erland

-0.14

roe

-0.14

utory

-0.13

 synchronization

-0.13

ceptions

-0.13

 Braun

-0.13

gan

-0.13

ynchronization

-0.13

lette

-0.12

POSITIVE LOGITS

 easily

0.19

Ã³t

0.16

 anywhere

0.15

pa

0.15

 specify

0.15

 anytime

0.15

rud

0.15

 specifying

0.14

 choose

0.14

 Vict

0.14

Activations Density 0.135%

features related to user customization and content sharing capabilities in applications

No Known Activations