INDEX

Explanations

questions and hypothetical scenarios posed with the word "would."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 2-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L2R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.2.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

 piger

-0.17

ieri

-0.15

-schema

-0.15

innen

-0.15

igi

-0.14

iei

-0.14

erti

-0.14

ozem

-0.14

ramer

-0.14

iar

-0.14

POSITIVE LOGITS

 isol

0.16

alone

0.16

erland

0.15

YL

0.15

 alone

0.14

 Alone

0.14

Î¹ÏĥÏĦÎ¿

0.14

 Ø·Ø±ÙĬÙĤ

0.14

CF

0.14

offline

0.13

Activations Density 0.024%

questions and hypothetical scenarios posed with the word "would."

No Comments

No Known Activations