INDEX

Explanations

terms related to radar technology and beamforming techniques

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

_bias

-0.15

 absorbed

-0.14

Ø±Ø®

-0.14

 mobil

-0.14

 coward

-0.14

ç¿»

-0.13

asl

-0.13

kazy

-0.13

 Flip

-0.13

gap

-0.13

POSITIVE LOGITS

 Fres

0.25

 kino

0.25

 diff

0.25

 Shack

0.24

dif

0.23

dif

0.22

 phase

0.21

 aberr

0.20

 pupil

0.20

upil

0.20

Activations Density 0.086%

terms related to radar technology and beamforming techniques

No Comments

No Known Activations