Push Phase 3 empire-embed sentence transformer

7c2a4d6 verified 3 months ago

19.2 kB

tags:
  - sentence-transformers
  - sentence-similarity
  - feature-extraction
  - dense
  - generated_from_trainer
  - dataset_size:333
  - loss:CosineSimilarityLoss
base_model: sentence-transformers/all-MiniLM-L6-v2
widget:
  - source_sentence: brain scan
    sentences:
      - >-
        Full WordPress Site Health Audit. Check plugins — installed, active, and
        update status command Verify SEO configuration command Test page speed
        with Lighthouse command Security check — WordFence scan and login
        protection check Test REST API connectivity and credentials command
        Check Google Search Console for crawl errors and ranking prompt. Tags:
        wordpress, audit, seo, security, performance, plugins, health
      - >-
        Bulk Article Creation for a Site. Research topics and identify content
        gaps command Generate content outlines and briefs command Write content
        using ZimmWriter or scripts prompt Generate featured images for all
        articles command Publish articles and set featured images command. Tags:
        content, articles, bulk, wordpress, seo, publishing, zimmwriter
      - >-
        Run EMPIRE-BRAIN Scan and Intelligence Cycle. Run full empire brain scan
        command Generate intelligence briefing command Check brain stats and
        performance metrics command Run evolution cycle command Verify Sentinel
        service monitoring status command. Tags: empire, brain, scan,
        intelligence, monitoring, evolution, briefing
  - source_sentence: check vps
    sentences:
      - >-
        Cross-Site Internal Linking Strategy. Identify cluster and linking
        opportunities check Find matching content between cluster sites command
        Generate and inject cross-links command Verify and monitor link health
        command. Tags: seo, internal-links, cross-site, clusters, link-whisper,
        revenue
      - >-
        Full WordPress Site Health Audit. Check plugins — installed, active, and
        update status command Verify SEO configuration command Test page speed
        with Lighthouse command Security check — WordFence scan and login
        protection check Test REST API connectivity and credentials command
        Check Google Search Console for crawl errors and ranking prompt. Tags:
        wordpress, audit, seo, security, performance, plugins, health
      - >-
        Full WordPress Site Health Audit. Check plugins — installed, active, and
        update status command Verify SEO configuration command Test page speed
        with Lighthouse command Security check — WordFence scan and login
        protection check Test REST API connectivity and credentials command
        Check Google Search Console for crawl errors and ranking prompt. Tags:
        wordpress, audit, seo, security, performance, plugins, health
  - source_sentence: performance optim
    sentences:
      - >-
        Run EMPIRE-BRAIN Scan and Intelligence Cycle. Run full empire brain scan
        command Generate intelligence briefing command Check brain stats and
        performance metrics command Run evolution cycle command Verify Sentinel
        service monitoring status command. Tags: empire, brain, scan,
        intelligence, monitoring, evolution, briefing
      - >-
        Site Speed & Core Web Vitals Optimization. Run Lighthouse audit command
        Configure LiteSpeed Cache (all sites use it) command Image optimization
        check Font and render-blocking resource optimization check Verify
        improvements and monitor command. Tags: seo, speed, performance,
        core-web-vitals, litespeed, images, caching
      - >-
        WordPress Plugin Deployment to Sites. Check if plugin is available on
        WorldPressIT check Install via WordPress REST API or WP-CLI command
        Configure plugin settings prompt Verify no conflicts with existing
        plugins check Deploy fleet-wide if applicable prompt. Tags: wordpress,
        plugin, deploy, fleet
  - source_sentence: make pinterest pin
    sentences:
      - >-
        Generate Article Featured Images. Run article_images_pipeline.py with
        correct arguments command Verify the featured image was set check. Tags:
        content, images, wordpress, featured-image
      - >-
        Social Media Post Generation & Scheduling. Generate platform-specific
        images command Write platform-specific copy prompt Schedule or post via
        automation command. Tags: content, social, pinterest, instagram,
        facebook, twitter, scheduling
      - >-
        Bootstrap a New Empire Project. Create project directory and initialize
        git command Create PROJECT_DNA.md command Create CLAUDE.md with
        project-specific config prompt Set up Python environment command Create
        initial git commit and push to GitHub command Register project in
        EMPIRE-BRAIN command. Tags: empire, project, bootstrap, setup, new
  - source_sentence: batch articles
    sentences:
      - >-
        Bulk Article Creation for a Site. Research topics and identify content
        gaps command Generate content outlines and briefs command Write content
        using ZimmWriter or scripts prompt Generate featured images for all
        articles command Publish articles and set featured images command. Tags:
        content, articles, bulk, wordpress, seo, publishing, zimmwriter
      - >-
        Site Speed & Core Web Vitals Optimization. Run Lighthouse audit command
        Configure LiteSpeed Cache (all sites use it) command Image optimization
        check Font and render-blocking resource optimization check Verify
        improvements and monitor command. Tags: seo, speed, performance,
        core-web-vitals, litespeed, images, caching
      - >-
        SSL Certificate & Domain Management. Check domain/SSL status via
        Hostinger command Manage DNS records command SSL certificate management
        manual Domain security audit command. Tags: ssl, domain, certificate,
        dns, hostinger, renewal
pipeline_tag: sentence-similarity
library_name: sentence-transformers
metrics:
  - pearson_cosine
  - spearman_cosine
model-index:
  - name: SentenceTransformer based on sentence-transformers/all-MiniLM-L6-v2
    results:
      - task:
          type: semantic-similarity
          name: Semantic Similarity
        dataset:
          name: empire eval
          type: empire-eval
        metrics:
          - type: pearson_cosine
            value: 0.5531234570199464
            name: Pearson Cosine
          - type: spearman_cosine
            value: 0.5320495924169611
            name: Spearman Cosine

SentenceTransformer based on sentence-transformers/all-MiniLM-L6-v2

This is a sentence-transformers model finetuned from sentence-transformers/all-MiniLM-L6-v2. It maps sentences & paragraphs to a 384-dimensional dense vector space and can be used for semantic textual similarity, semantic search, paraphrase mining, text classification, clustering, and more.

Model Details

Model Description

Model Type: Sentence Transformer
Base model: sentence-transformers/all-MiniLM-L6-v2
Maximum Sequence Length: 256 tokens
Output Dimensionality: 384 dimensions
Similarity Function: Cosine Similarity

Model Sources

Documentation: Sentence Transformers Documentation
Repository: Sentence Transformers on GitHub
Hugging Face: Sentence Transformers on Hugging Face

Full Model Architecture

SentenceTransformer(
  (0): Transformer({'max_seq_length': 256, 'do_lower_case': False, 'architecture': 'BertModel'})
  (1): Pooling({'word_embedding_dimension': 384, 'pooling_mode_cls_token': False, 'pooling_mode_mean_tokens': True, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': False, 'include_prompt': True})
  (2): Normalize()
)

Usage

Direct Usage (Sentence Transformers)

First install the Sentence Transformers library:

pip install -U sentence-transformers

Then you can load this model and run inference.

from sentence_transformers import SentenceTransformer

# Download from the 🤗 Hub
model = SentenceTransformer("WealthFromAI/empire-embed")
# Run inference
sentences = [
    'batch articles',
    'Bulk Article Creation for a Site. Research topics and identify content gaps command Generate content outlines and briefs command Write content using ZimmWriter or scripts prompt Generate featured images for all articles command Publish articles and set featured images command. Tags: content, articles, bulk, wordpress, seo, publishing, zimmwriter',
    'Site Speed & Core Web Vitals Optimization. Run Lighthouse audit command Configure LiteSpeed Cache (all sites use it) command Image optimization check Font and render-blocking resource optimization check Verify improvements and monitor command. Tags: seo, speed, performance, core-web-vitals, litespeed, images, caching',
]
embeddings = model.encode(sentences)
print(embeddings.shape)
# [3, 384]

# Get the similarity scores for the embeddings
similarities = model.similarity(embeddings, embeddings)
print(similarities)
# tensor([[1.0000, 0.7041, 0.2289],
#         [0.7041, 1.0000, 0.3787],
#         [0.2289, 0.3787, 1.0000]])

Evaluation

Metrics

Semantic Similarity

Dataset: empire-eval
Evaluated with EmbeddingSimilarityEvaluator

Metric	Value
pearson_cosine	0.5531
spearman_cosine	0.532

Training Details

Training Dataset

Unnamed Dataset

Size: 333 training samples
Columns: sentence_0, sentence_1, and label
Approximate statistics based on the first 333 samples:
sentence_0 sentence_1 label
type string string float
details
min: 3 tokens
mean: 5.24 tokens
max: 33 tokens

min: 6 tokens
mean: 62.94 tokens
max: 141 tokens

min: 0.0
mean: 0.57
max: 1.0

	sentence_0	sentence_1	label
type	string	string	float
details	min: 3 tokens mean: 5.24 tokens max: 33 tokens	min: 6 tokens mean: 62.94 tokens max: 141 tokens	min: 0.0 mean: 0.57 max: 1.0

Samples:

sentence_0	sentence_1	label
`bootstrap project`	`Bootstrap a New Empire Project. Create project directory and initialize git command Create PROJECT_DNA.md command Create CLAUDE.md with project-specific config prompt Set up Python environment command Create initial git commit and push to GitHub command Register project in EMPIRE-BRAIN command. Tags: empire, project, bootstrap, setup, new`	`1.0`
`update container`	`WordPress Site SEO Setup & Configuration. Verify RankMath SEO plugin is installed and activated check Configure RankMath general settings prompt Set up Schema markup patterns per post type prompt Configure robots.txt and sitemap command Set up internal linking structure prompt Configure affiliate link handling prompt. Tags: wordpress, seo, rankmath, schema, setup`	`0.0`
`restart service`	`Full WordPress Site Health Audit. Check plugins — installed, active, and update status command Verify SEO configuration command Test page speed with Lighthouse command Security check — WordFence scan and login protection check Test REST API connectivity and credentials command Check Google Search Console for crawl errors and ranking prompt. Tags: wordpress, audit, seo, security, performance, plugins, health`	`0.0`

Loss: CosineSimilarityLoss with these parameters:

{
    "loss_fct": "torch.nn.modules.loss.MSELoss"
}

Training Hyperparameters

Non-Default Hyperparameters

per_device_train_batch_size: 16
num_train_epochs: 5
eval_strategy: steps
per_device_eval_batch_size: 16
multi_dataset_batch_sampler: round_robin

All Hyperparameters

Click to expand

per_device_train_batch_size: 16
num_train_epochs: 5
max_steps: -1
learning_rate: 5e-05
lr_scheduler_type: linear
lr_scheduler_kwargs: None
warmup_steps: 0
optim: adamw_torch_fused
optim_args: None
weight_decay: 0.0
adam_beta1: 0.9
adam_beta2: 0.999
adam_epsilon: 1e-08
optim_target_modules: None
gradient_accumulation_steps: 1
average_tokens_across_devices: True
max_grad_norm: 1
label_smoothing_factor: 0.0
bf16: False
fp16: False
bf16_full_eval: False
fp16_full_eval: False
tf32: None
gradient_checkpointing: False
gradient_checkpointing_kwargs: None
torch_compile: False
torch_compile_backend: None
torch_compile_mode: None
use_liger_kernel: False
liger_kernel_config: None
use_cache: False
neftune_noise_alpha: None
torch_empty_cache_steps: None
auto_find_batch_size: False
log_on_each_node: True
logging_nan_inf_filter: True
include_num_input_tokens_seen: no
log_level: passive
log_level_replica: warning
disable_tqdm: False
project: huggingface
trackio_space_id: trackio
eval_strategy: steps
per_device_eval_batch_size: 16
prediction_loss_only: True
eval_on_start: False
eval_do_concat_batches: True
eval_use_gather_object: False
eval_accumulation_steps: None
include_for_metrics: []
batch_eval_metrics: False
save_only_model: False
save_on_each_node: False
enable_jit_checkpoint: False
push_to_hub: False
hub_private_repo: None
hub_model_id: None
hub_strategy: every_save
hub_always_push: False
hub_revision: None
load_best_model_at_end: False
ignore_data_skip: False
restore_callback_states_from_checkpoint: False
full_determinism: False
seed: 42
data_seed: None
use_cpu: False
accelerator_config: {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}
parallelism_config: None
dataloader_drop_last: False
dataloader_num_workers: 0
dataloader_pin_memory: True
dataloader_persistent_workers: False
dataloader_prefetch_factor: None
remove_unused_columns: True
label_names: None
train_sampling_strategy: random
length_column_name: length
ddp_find_unused_parameters: None
ddp_bucket_cap_mb: None
ddp_broadcast_buffers: False
ddp_backend: None
ddp_timeout: 1800
fsdp: []
fsdp_config: {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}
deepspeed: None
debug: []
skip_memory_metrics: True
do_predict: False
resume_from_checkpoint: None
warmup_ratio: None
local_rank: -1
prompts: None
batch_sampler: batch_sampler
multi_dataset_batch_sampler: round_robin
router_mapping: {}
learning_rate_mapping: {}

Training Logs

Epoch	Step	empire-eval_spearman_cosine
1.0	21	0.5320

Framework Versions

Python: 3.12.10
Sentence Transformers: 5.3.0
Transformers: 5.4.0
PyTorch: 2.11.0+cpu
Accelerate: 1.13.0
Datasets: 4.8.4
Tokenizers: 0.22.2

Citation

BibTeX

Sentence Transformers

@inproceedings{reimers-2019-sentence-bert,
    title = "Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks",
    author = "Reimers, Nils and Gurevych, Iryna",
    booktitle = "Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing",
    month = "11",
    year = "2019",
    publisher = "Association for Computational Linguistics",
    url = "https://arxiv.org/abs/1908.10084",
}