INDEX

Explanations

phrases that express varying degrees of frustration and complexity in outcomes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 25-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L25R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.25.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

 improbable

-0.16

 unthinkable

-0.15

 unimagin

-0.14

abus

-0.14

elez

-0.14

 impl

-0.14

echan

-0.13

 Ð¼ÐµÑĤÐ°Ð»Ð»Ð¸

-0.13

ÑİÐ´Ð¶

-0.13

 horrified

-0.13

POSITIVE LOGITS

 interesting

0.34

interesting

0.30

-interest

0.29

 enlight

0.29

 grat

0.29

Interesting

0.29

 instruct

0.28

 interes

0.28

 Fasc

0.28

 Interesting

0.28

Activations Density 0.563%

phrases that express varying degrees of frustration and complexity in outcomes

No Comments

No Known Activations