INDEX

Explanations

phrases related to maintaining or preserving something

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 4-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L4R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.4.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

aub

-0.21

ÅĤu

-0.15

antt

-0.15

å°Ĥ

-0.15

erie

-0.15

MBED

-0.14

emed

-0.14

antino

-0.14

Ð´Ð°Ð¿

-0.13

itar

-0.13

POSITIVE LOGITS

ngle

0.15

èµ

0.15

ellipsis

0.15

 Vital

0.15

idge

0.14

lash

0.14

idges

0.14

à¤¾à¤Ĺà¤°

0.14

endor

0.14

Catch

0.14

Activations Density 0.029%

phrases related to maintaining or preserving something

No Comments

No Known Activations