INDEX

Explanations

references to physical attributes and characteristics of people and things

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 27-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L27R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.27.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

.criteria

-0.17

-0.15

.cat

-0.15

ODY

-0.15

Ļ

-0.15

ãĥ¼ãĤ¸

-0.14

alom

-0.14

è±Ĩ

-0.14

èª¿

-0.14

 proximity

-0.14

POSITIVE LOGITS

 sport

0.36

 sporting

0.33

sport

0.32

 sports

0.30

Sport

0.28

 wears

0.27

 SPORT

0.27

 Sport

0.26

sports

0.24

 deck

0.24

Activations Density 0.022%

references to physical attributes and characteristics of people and things

No Comments

No Known Activations