INDEX

Explanations

programming-related terms and commands

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.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

rani

-0.17

 Origin

-0.15

ç«¥

-0.15

	UObject

-0.15

/commons

-0.14

CHO

-0.14

Origin

-0.14

 origin

-0.14

ÏĦÎŃ

-0.14

é¥

-0.14

POSITIVE LOGITS

 qualified

0.39

qualified

0.29

 Control

0.23

Qualified

0.23

{-

0.22

 qualification

0.21

 Foreign

0.20

GHC

0.20

foreign

0.20

Foreign

0.20

Activations Density 0.002%

programming-related terms and commands

No Comments

No Known Activations