INDEX

Explanations

expressions of success or achievement

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

elas

-0.15

Ã¡du

-0.13

èĪį

-0.13

Ø§Ø³Ø§ÙĨ

-0.12

ARAM

-0.12

ighton

-0.12

 Shoe

-0.12

eligible

-0.12

UnderTest

-0.12

ÐµÐ»Ð¸Ñĩ

-0.12

POSITIVE LOGITS

 success

1.03

 successful

1.00

æĪĲåĬŁ

0.93

 succes

0.87

 Successful

0.87

success

0.85

successful

0.85

 Success

0.85

-success

0.85

 succeed

0.81

Activations Density 0.504%

expressions of success or achievement

No Comments

No Known Activations

expressions of success or achievement

No Comments

No Known Activations