INDEX

Explanations

programming and coding-related keywords, particularly related to input handling and web development

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

RAR

-0.16

urge

-0.16

kaar

-0.15

.vocab

-0.15

.appspot

-0.15

Raised

-0.15

indow

-0.14

stÃŃ

-0.14

positor

-0.14

 è¨Ģ

-0.14

POSITIVE LOGITS

alex

0.14

omatic

0.14

 Mayo

0.13

Ã¡s

0.13

.createFrom

0.13

orman

0.13

à¸ªà¸¥

0.13

ound

0.13

 Venez

0.13

/by

0.13

Activations Density 0.052%

programming and coding-related keywords, particularly related to input handling and web development

No Comments

No Known Activations