INDEX

Explanations

the word "newly" and its variations, indicating a focus on recent developments or changes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

reach

-0.15

eward

-0.15

ulumi

-0.14

/Dk

-0.14

EMPLARY

-0.14

venture

-0.14

íħĶ

-0.14

é¡¿

-0.13

lacak

-0.13

à¤¾à¤§

-0.13

POSITIVE LOGITS

 mint

0.38

wed

0.28

mint

0.28

 Mint

0.22

 formed

0.20

 christ

0.19

 arrived

0.18

-formed

0.18

wed

0.17

 created

0.17

Activations Density 0.011%

the word "newly" and its variations, indicating a focus on recent developments or changes

No Comments

No Known Activations