INDEX

Explanations

sequences of numbers and dates in the text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 31-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L31R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.31.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

caret

-0.17

 nunca

-0.17

 never

-0.16

imson

-0.16

-caret

-0.15

 spÃ¤ter

-0.15

 siempre

-0.15

 later

-0.14

ÑħÑĢÐ°Ð½

-0.14

ootball

-0.14

POSITIVE LOGITS

 firma

0.23

 fund

0.22

 firm

0.22

 inaug

0.21

 decide

0.20

 entra

0.19

 publish

0.19

 anunci

0.18

pub

0.18

 decides

0.18

Activations Density 0.040%

sequences of numbers and dates in the text

No Comments

No Known Activations