INDEX

Explanations

programming-related keywords and structures in code

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

Ĥæķ°

-0.16

Ã¡zi

-0.16

arus

-0.14

ãģ¨ãĤĤ

-0.14

ileo

-0.14

ptions

-0.14

mÃ¼r

-0.14

olicited

-0.14

odzi

-0.14

POSITIVE LOGITS

Ø§

0.15

uter

0.14

 recip

0.14

 Appears

0.14

 circ

0.13

DetailsService

0.13

Ð¸Ð½Ð¾Ð²

0.13

PCS

0.13

ï½Ĵ

0.13

 Benson

0.13

Activations Density 0.063%

programming-related keywords and structures in code

No Comments

No Known Activations