INDEX

Explanations

roles and characters in performances or productions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.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

iferay

-0.15

ung

-0.15

ilyn

-0.15

Truthy

-0.14

idth

-0.13

arte

-0.13

ewish

-0.13

ÑĥÑĢÐ¸

-0.13

 Cush

-0.13

mail

-0.13

POSITIVE LOGITS

ãĥ¼ãĥª

0.15

RING

0.15

aggio

0.15

cxx

0.14

etÃŃ

0.14

regn

0.14

tout

0.14

verts

0.14

 circulation

0.14

ynchronized

0.13

Activations Density 0.011%

roles and characters in performances or productions

No Comments

No Known Activations