INDEX

Explanations

references to statistical analysis of time series data

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

 sting

-0.15

gra

-0.14

 protected

-0.14

 strokes

-0.14

aking

-0.14

 transc

-0.14

Rel

-0.13

coli

-0.13

 chances

-0.13

 ÑģÐ¾ÑĢ

-0.13

POSITIVE LOGITS

 ARCH

0.28

 station

0.27

aut

0.27

 Station

0.25

ARCH

0.24

Aut

0.24

Station

0.23

Aut

0.22

 autoc

0.22

-aut

0.22

Activations Density 0.026%

references to statistical analysis of time series data

No Comments

No Known Activations