INDEX

Explanations

frequent discourse markers or transitional phrases

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

hint

-0.16

emouth

-0.15

anca

-0.15

à¹īà¸²à¸ĩ

-0.14

utter

-0.14

icon

-0.14

riz

-0.14

 darn

-0.14

hook

-0.14

oga

-0.14

POSITIVE LOGITS

 onto

0.37

onto

0.33

 enough

0.30

 Enough

0.27

Ont

0.26

Ont

0.24

 back

0.22

 moving

0.21

moving

0.21

Enough

0.21

Activations Density 0.092%

frequent discourse markers or transitional phrases

No Comments

No Known Activations