INDEX

Explanations

references to searching online, particularly using Google

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 12-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L12R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.12.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

cz

-0.15

-0.14

PropertyDescriptor

-0.14

 Sidd

-0.14

etail

-0.14

erdale

-0.13

itious

-0.13

 Sevilla

-0.13

ickle

-0.13

gle

-0.13

POSITIVE LOGITS

VML

0.16

à¤¿à¤¥

0.15

DIR

0.15

tings

0.15

VICE

0.14

 Durant

0.14

linik

0.14

 GENERIC

0.14

 Stateless

0.14

abinet

0.14

Activations Density 0.032%

references to searching online, particularly using Google

No Comments

No Known Activations