INDEX

Explanations

phrases inviting engagement or action from the audience

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 13-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L13R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.13.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

umba

-0.15

ritis

-0.15

èĽ

-0.14

ÙĨ

-0.14

mpl

-0.14

ilton

-0.14

ass

-0.14

inning

-0.13

native

-0.13

idor

-0.13

POSITIVE LOGITS

 anytime

0.18

any

0.18

esine

0.17

 freely

0.17

.fre

0.15

hausen

0.15

 Gratis

0.14

.any

0.14

 ìŀĲìľł

0.14

Activations Density 0.028%

phrases inviting engagement or action from the audience

No Comments

No Known Activations