INDEX

Explanations

phrases encouraging action or response from the reader

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

mons

-0.16

Eat

-0.15

NSF

-0.15

irectional

-0.15

rub

-0.14

string

-0.14

unal

-0.14

 Advance

-0.14

Sid

-0.14

pol

-0.14

POSITIVE LOGITS

Gratis

0.17

privation

0.16

.scalablytyped

0.16

anse

0.15

CADE

0.15

Ð¾ÑģÑĥÐ´

0.15

 nues

0.15

agnar

0.15

/Dk

0.14

Activations Density 0.098%

phrases encouraging action or response from the reader

No Comments

No Known Activations