INDEX

Explanations

the word "just" and its variations, indicating an emphasis on simplicity or minimalism

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

Ð¾ÑĢÐ°Ð·

-0.17

oog

-0.15

.infinity

-0.14

_DEFINE

-0.14

anitize

-0.14

issen

-0.14

æ»

-0.13

jsc

-0.13

slo

-0.13

lant

-0.13

POSITIVE LOGITS

ajan

0.16

 Rodgers

0.15

ublic

0.14

èĪĪ

0.14

 simply

0.14

iffe

0.13

erli

0.13

OfType

0.13

ander

0.13

gone

0.13

Activations Density 0.013%

the word "just" and its variations, indicating an emphasis on simplicity or minimalism

No Comments

No Known Activations