INDEX

Explanations

phrases and actions related to planning and organizing activities or information

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

eldom

-0.15

uars

-0.14

osemite

-0.13

ologne

-0.13

ariant

-0.13

ucker

-0.12

ÐµÐ»ÑĮÐ·Ñı

-0.12

ÏĢÎ¿ÏĤ

-0.12

 volte

-0.12

.setViewport

-0.12

POSITIVE LOGITS

 some

1.05

some

0.85

 Some

0.81

Some

0.75

ä¸ĢäºĽ

0.71

 SOME

0.66

.some

0.65

_some

0.65

 einige

0.62

äºĽ

0.57

Activations Density 0.827%

phrases and actions related to planning and organizing activities or information

No Comments

No Known Activations