INDEX

Explanations

elements related to episodic content and titles from television shows

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 16-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L16R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.16.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

ActionCreators

-0.16

reo

-0.15

elerik

-0.15

SSERT

-0.15

 Äįer

-0.14

à¸ªà¸Ķ

-0.14

à¸ļà¸²à¸Ĺ

-0.14

.lt

-0.14

=-=-=-=-=-=-=-=-

-0.14

lio

-0.14

POSITIVE LOGITS

ilton

0.17

 Stanton

0.16

0.15

0.14

errar

0.14

ysi

0.14

åŁİ

0.14

 Calcul

0.14

Activations Density 0.075%

elements related to episodic content and titles from television shows

No Comments

No Known Activations