INDEX

Explanations

coding-related terms and structures, particularly those associated with route mapping and configuration in software applications

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 VerfÃ¼g

-0.16

.Env

-0.15

Ð°ÑĢÑĩ

-0.15

RenderingContext

-0.14

RLF

-0.14

ollen

-0.14

beam

-0.13

resher

-0.13

orizontal

-0.13

echa

-0.13

POSITIVE LOGITS

 routes

0.29

 route

0.28

 Route

0.26

 Routes

0.25

Route

0.24

route

0.24

.route

0.22

Routes

0.22

routes

0.22

pattern

0.21

Activations Density 0.031%

coding-related terms and structures, particularly those associated with route mapping and configuration in software applications

No Comments

No Known Activations