INDEX

Explanations

references to software functionalities and programming tasks related to autocomplete features

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 23-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L23R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.23.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

ucc

-0.14

nez

-0.14

rena

-0.14

ÑĥÐºÐ¸

-0.14

urvey

-0.14

 Ø´Ùģ

-0.14

ÅĻik

-0.14

burgh

-0.13

.bid

-0.13

allee

-0.13

POSITIVE LOGITS

 suggestions

0.41

 suggestion

0.40

Suggestions

0.34

 Suggestions

0.34

 suggest

0.32

 suggesting

0.31

Aut

0.29

suggest

0.29

 suggested

0.29

 completion

0.28

Activations Density 0.031%

references to software functionalities and programming tasks related to autocomplete features

No Comments

No Known Activations