INDEX

Explanations

references to the letter "Y" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.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

ç¼ĺ

-0.18

inoa

-0.15

aled

-0.14

alement

-0.14

 quota

-0.14

ceptar

-0.14

 Threads

-0.14

ear

-0.14

 Kron

-0.14

Cup

-0.14

POSITIVE LOGITS

gles

0.22

bane

0.20

dens

0.20

ilm

0.20

bar

0.19

phant

0.18

eam

0.18

zag

0.18

ank

0.17

unker

0.17

Activations Density 0.015%

references to the letter "Y" in various contexts

No Comments

No Known Activations