INDEX

Explanations

phrases related to earning and rewards

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

 Mirror

-0.15

pline

-0.14

Ä¾

-0.14

eldom

-0.14

Mirror

-0.14

uess

-0.13

 Sharon

-0.13

edata

-0.13

tes

-0.13

Ã±a

-0.13

POSITIVE LOGITS

 Earn

0.17

:request

0.16

 earned

0.16

Earn

0.15

enance

0.15

ORITY

0.15

Ã©gor

0.15

ussy

0.15

 earn

0.14

earned

0.14

Activations Density 0.029%

phrases related to earning and rewards

No Comments

No Known Activations