INDEX

Explanations

technical terms and parameters related to programming and layout management

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

angling

-0.14

ode

-0.14

iah

-0.14

Wunused

-0.14

round

-0.14

 Cord

-0.13

 stead

-0.13

å´İ

-0.13

cap

-0.13

ylie

-0.13

POSITIVE LOGITS

Stretch

0.19

Spacer

0.18

 Stretch

0.18

 anch

0.18

 Fill

0.18

 fill

0.17

 stretch

0.17

Fill

0.17

 fills

0.17

fills

0.17

Activations Density 0.039%

technical terms and parameters related to programming and layout management

No Comments

No Known Activations