INDEX

Explanations

programming keywords and structures related to data handling, particularly in dictionary-like contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 10-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L10R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.10.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

egin

-0.16

itia

-0.15

.modules

-0.15

 Maher

-0.15

oogle

-0.15

Dip

-0.14

Ep

-0.14

hammer

-0.14

 gratuita

-0.14

ate

-0.14

POSITIVE LOGITS

iece

0.20

nero

0.16

ocator

0.15

owi

0.15

ailer

0.15

eks

0.15

erner

0.15

 DbType

0.15

ÑŁ

0.14

uales

0.14

Activations Density 0.140%

programming keywords and structures related to data handling, particularly in dictionary-like contexts

No Comments

No Known Activations