INDEX

Explanations

words related to fading or diminishing effects

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 2-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L2R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.2.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

Å©i

-0.16

spÄĽ

-0.15

bib

-0.15

Hur

-0.15

à¥Ģà¤¬

-0.14

ander

-0.14

Ã¨o

-0.14

 cÃ©lib

-0.13

ëĭ¹

-0.13

ebi

-0.13

POSITIVE LOGITS

EMON

0.15

etic

0.15

-transitional

0.15

fade

0.15

/problem

0.15

lights

0.15

mÃ©

0.15

uating

0.14

arella

0.14

_TestCase

0.14

Activations Density 0.008%

words related to fading or diminishing effects

No Comments

No Known Activations