INDEX

Explanations

phrases indicating time references or sequences

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 25-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L25R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.25.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

imest

-0.17

yw

-0.14

exampleInputEmail

-0.14

longleftrightarrow

-0.14

arest

-0.14

aturdays

-0.13

(TM

-0.13

.setViewport

-0.13

Ð¾Ð³Ð¾Ð´

-0.13

inja

-0.13

POSITIVE LOGITS

 following

1.16

following

1.01

 Following

0.96

Following

0.89

 siguiente

0.81

 siguientes

0.79

 seguint

0.79

 ÑģÐ»ÐµÐ´ÑĥÑİÑī

0.78

 suiv

0.75

 nÃ¡sledujÃŃcÃŃ

0.64

Activations Density 0.224%

phrases indicating time references or sequences

No Comments

No Known Activations