INDEX

Explanations

references to class components in a programming context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.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

Ã³ng

-0.16

ccoli

-0.15

imp

-0.15

nee

-0.15

à¥įà¤µà¤µ

-0.15

bung

-0.14

mada

-0.14

andez

-0.14

vik

-0.14

Ð°ÐºÐ°

-0.14

POSITIVE LOGITS

Arena

0.17

whe

0.14

ãģ°ãģĭãĤĬ

0.14

åĪļæīį

0.14

-scalable

0.14

 solic

0.14

laÄį

0.13

Pod

0.13

 Solic

0.13

aben

0.13

Activations Density 0.001%

references to class components in a programming context

No Comments

No Known Activations