INDEX

Explanations

references to contests, competitions, and prizes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 21-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L21R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.21.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

owi

-0.15

ocks

-0.14

unga

-0.14

buat

-0.14

(Paint

-0.13

æ³¢

-0.13

exe

-0.13

nist

-0.13

inker

-0.13

ould

-0.13

POSITIVE LOGITS

 prize

0.66

 prizes

0.65

 Prize

0.56

 award

0.44

 awards

0.44

å¥ĸ

0.41

è³ŀ

0.40

 Award

0.40

 rewards

0.38

 reward

0.38

Activations Density 0.176%

references to contests, competitions, and prizes

No Comments

No Known Activations