INDEX

Explanations

references to individuals with the name Julian or Adrian

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 18-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L18R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.18.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

uffman

-0.19

edly

-0.16

upe

-0.15

kie

-0.14

 OTHERWISE

-0.14

itionally

-0.14

ULAR

-0.14

reek

-0.13

ãģĦãģ¦

-0.13

IOR

-0.13

POSITIVE LOGITS

ne

0.21

ople

0.20

apolis

0.20

ities

0.19

opsis

0.17

elli

0.17

opol

0.16

thus

0.16

ovsky

0.15

isiert

0.15

Activations Density 0.024%

references to individuals with the name Julian or Adrian

No Comments

No Known Activations