INDEX

Explanations

references to programming concepts and structures

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

uja

-0.16

phem

-0.15

loff

-0.15

Balancer

-0.15

SizePolicy

-0.15

rani

-0.15

CHO

-0.15

	UObject

-0.14

Shock

-0.14

 Origin

-0.14

POSITIVE LOGITS

 qualified

0.35

qualified

0.27

Qualified

0.21

 control

0.21

 Foreign

0.20

 Control

0.20

{-

0.19

 qualification

0.18

ontrol

0.18

Foreign

0.18

Activations Density 0.006%

references to programming concepts and structures

No Comments

No Known Activations