cloudyu-nemo-dpo-v22

developer_uid: cloudyu

submission_id: cloudyu-nemo-dpo-v22_v5

model_name: cloudyu-nemo-dpo-v22_v1

model_group: cloudyu/Nemo-DPO-V22

status: inactive

timestamp: 2024-12-18T14:06:01+00:00

num_battles: 18709

num_wins: 9869

celo_rating: 1281.5

family_friendly_score: 0.56

family_friendly_standard_error: 0.007019971509913698

submission_type: basic

model_repo: cloudyu/Nemo-DPO-V22

model_architecture: MistralForCausalLM

model_num_parameters: 12772070400.0

best_of: 8

max_input_tokens: 1024

max_output_tokens: 64

latencies: [{'batch_size': 1, 'throughput': 0.6109791628699919, 'latency_mean': 1.6366224789619446, 'latency_p50': 1.627211570739746, 'latency_p90': 1.8205772161483764}, {'batch_size': 3, 'throughput': 1.1151977683090772, 'latency_mean': 2.6789412009716034, 'latency_p50': 2.6999326944351196, 'latency_p90': 2.995386028289795}, {'batch_size': 5, 'throughput': 1.3498645447638864, 'latency_mean': 3.6884069454669954, 'latency_p50': 3.6967523097991943, 'latency_p90': 4.146683764457703}, {'batch_size': 6, 'throughput': 1.4239437382959645, 'latency_mean': 4.204284677505493, 'latency_p50': 4.2069783210754395, 'latency_p90': 4.663272786140442}, {'batch_size': 8, 'throughput': 1.4736965298295552, 'latency_mean': 5.396490411758423, 'latency_p50': 5.417945027351379, 'latency_p90': 6.07870466709137}, {'batch_size': 10, 'throughput': 1.5156729341373307, 'latency_mean': 6.52959664106369, 'latency_p50': 6.563353896141052, 'latency_p90': 7.435246849060059}]

gpu_counts: {'NVIDIA RTX A5000': 1}

display_name: cloudyu-nemo-dpo-v22_v1

is_internal_developer: False

language_model: cloudyu/Nemo-DPO-V22

model_size: 13B

ranking_group: single

throughput_3p7s: 1.36

us_pacific_date: 2024-12-18

win_ratio: 0.5275001336255278

generation_params: {'temperature': 1.25, 'top_p': 0.95, 'min_p': 0.05, 'top_k': 80, 'presence_penalty': 0.0, 'frequency_penalty': 0.0, 'stopping_words': ['\n'], 'max_input_tokens': 1024, 'best_of': 8, 'max_output_tokens': 64}

formatter: {'memory_template': "{bot_name}'s Persona: {memory}\n####\n", 'prompt_template': '{prompt}\n<START>\n', 'bot_template': '{bot_name}: {message}\n', 'user_template': '{user_name}: {message}\n', 'response_template': '{bot_name}:', 'truncate_by_message': False}

Resubmit model

Shutdown handler not registered because Python interpreter is not running in the main thread
run pipeline %s
run pipeline stage %s
Running pipeline stage MKMLizer
Starting job with name cloudyu-nemo-dpo-v22-v5-mkmlizer
Waiting for job on cloudyu-nemo-dpo-v22-v5-mkmlizer to finish
Connection pool is full, discarding connection: %s. Connection pool size: %s
Connection pool is full, discarding connection: %s. Connection pool size: %s
Connection pool is full, discarding connection: %s. Connection pool size: %s
Connection pool is full, discarding connection: %s. Connection pool size: %s
Connection pool is full, discarding connection: %s. Connection pool size: %s
Connection pool is full, discarding connection: %s. Connection pool size: %s
Connection pool is full, discarding connection: %s. Connection pool size: %s
Connection pool is full, discarding connection: %s. Connection pool size: %s
cloudyu-nemo-dpo-v22-v5-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
cloudyu-nemo-dpo-v22-v5-mkmlizer: ║     _____            __           __                                ║
cloudyu-nemo-dpo-v22-v5-mkmlizer: ║    / _/ /_ ___    __/ /  ___ ___ / /                                ║
cloudyu-nemo-dpo-v22-v5-mkmlizer: ║   / _/ / // / |/|/ / _ \/ -_) -_) /                                 ║
cloudyu-nemo-dpo-v22-v5-mkmlizer: ║  /_//_/\_, /|__,__/_//_/\__/\__/_/                                  ║
cloudyu-nemo-dpo-v22-v5-mkmlizer: ║       /___/                                                         ║
cloudyu-nemo-dpo-v22-v5-mkmlizer: ║                                                                     ║
cloudyu-nemo-dpo-v22-v5-mkmlizer: ║  Version: 0.11.12                                                   ║
cloudyu-nemo-dpo-v22-v5-mkmlizer: ║  Copyright 2023 MK ONE TECHNOLOGIES Inc.                            ║
cloudyu-nemo-dpo-v22-v5-mkmlizer: ║  https://mk1.ai                                                     ║
cloudyu-nemo-dpo-v22-v5-mkmlizer: ║                                                                     ║
cloudyu-nemo-dpo-v22-v5-mkmlizer: ║  The license key for the current software has been verified as      ║
cloudyu-nemo-dpo-v22-v5-mkmlizer: ║  belonging to:                                                      ║
cloudyu-nemo-dpo-v22-v5-mkmlizer: ║                                                                     ║
cloudyu-nemo-dpo-v22-v5-mkmlizer: ║  Chai Research Corp.                                                ║
cloudyu-nemo-dpo-v22-v5-mkmlizer: ║  Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f                   ║
cloudyu-nemo-dpo-v22-v5-mkmlizer: ║  Expiration: 2025-01-15 23:59:59                                    ║
cloudyu-nemo-dpo-v22-v5-mkmlizer: ║                                                                     ║
cloudyu-nemo-dpo-v22-v5-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
Connection pool is full, discarding connection: %s. Connection pool size: %s
Connection pool is full, discarding connection: %s. Connection pool size: %s
Connection pool is full, discarding connection: %s. Connection pool size: %s
Connection pool is full, discarding connection: %s. Connection pool size: %s
Connection pool is full, discarding connection: %s. Connection pool size: %s
Connection pool is full, discarding connection: %s. Connection pool size: %s
Connection pool is full, discarding connection: %s. Connection pool size: %s
Connection pool is full, discarding connection: %s. Connection pool size: %s
Connection pool is full, discarding connection: %s. Connection pool size: %s
cloudyu-nemo-dpo-v22-v5-mkmlizer: Downloaded to shared memory in 32.487s
cloudyu-nemo-dpo-v22-v5-mkmlizer: quantizing model to /dev/shm/model_cache, profile:s0, folder:/tmp/tmp1cquhrp_, device:0
cloudyu-nemo-dpo-v22-v5-mkmlizer: Saving flywheel model at /dev/shm/model_cache
cloudyu-nemo-dpo-v22-v5-mkmlizer: quantized model in 35.914s
cloudyu-nemo-dpo-v22-v5-mkmlizer: Processed model cloudyu/Nemo-DPO-V22 in 68.402s
cloudyu-nemo-dpo-v22-v5-mkmlizer: creating bucket guanaco-mkml-models
cloudyu-nemo-dpo-v22-v5-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
cloudyu-nemo-dpo-v22-v5-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/cloudyu-nemo-dpo-v22-v5
cloudyu-nemo-dpo-v22-v5-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/cloudyu-nemo-dpo-v22-v5/config.json
cloudyu-nemo-dpo-v22-v5-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/cloudyu-nemo-dpo-v22-v5/special_tokens_map.json
cloudyu-nemo-dpo-v22-v5-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/cloudyu-nemo-dpo-v22-v5/tokenizer_config.json
cloudyu-nemo-dpo-v22-v5-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/cloudyu-nemo-dpo-v22-v5/tokenizer.json
cloudyu-nemo-dpo-v22-v5-mkmlizer: 
Loading 0:   0%|          | 0/363 [00:00<?, ?it/s]
Loading 0:   1%|▏         | 5/363 [00:00<00:12, 28.62it/s]
Loading 0:   4%|▎         | 13/363 [00:00<00:07, 48.99it/s]
Loading 0:   5%|▌         | 19/363 [00:00<00:07, 44.75it/s]
Loading 0:   7%|▋         | 24/363 [00:00<00:07, 43.47it/s]
Loading 0:   9%|▊         | 31/363 [00:00<00:06, 48.83it/s]
Loading 0:  10%|█         | 37/363 [00:00<00:07, 45.72it/s]
Loading 0:  12%|█▏        | 42/363 [00:00<00:07, 44.65it/s]
Loading 0:  13%|█▎        | 49/363 [00:01<00:06, 48.90it/s]
Loading 0:  15%|█▌        | 55/363 [00:01<00:06, 44.92it/s]
Loading 0:  17%|█▋        | 60/363 [00:01<00:06, 45.43it/s]
Loading 0:  18%|█▊        | 65/363 [00:01<00:09, 30.50it/s]
Loading 0:  20%|█▉        | 72/363 [00:01<00:07, 36.82it/s]
Loading 0:  21%|██▏       | 78/363 [00:01<00:07, 38.13it/s]
Loading 0:  23%|██▎       | 83/363 [00:02<00:07, 39.12it/s]
Loading 0:  25%|██▍       | 90/363 [00:02<00:06, 43.51it/s]
Loading 0:  26%|██▋       | 96/363 [00:02<00:06, 42.41it/s]
Loading 0:  28%|██▊       | 101/363 [00:02<00:06, 42.13it/s]
Loading 0:  29%|██▉       | 106/363 [00:02<00:05, 43.21it/s]
Loading 0:  31%|███       | 113/363 [00:02<00:05, 42.55it/s]
Loading 0:  33%|███▎      | 118/363 [00:02<00:05, 42.36it/s]
Loading 0:  35%|███▍      | 126/363 [00:02<00:04, 49.95it/s]
Loading 0:  36%|███▋      | 132/363 [00:03<00:04, 46.50it/s]
Loading 0:  38%|███▊      | 137/363 [00:03<00:04, 45.86it/s]
Loading 0:  39%|███▉      | 142/363 [00:03<00:06, 35.54it/s]
Loading 0:  40%|████      | 146/363 [00:03<00:06, 36.13it/s]
Loading 0:  41%|████▏     | 150/363 [00:03<00:06, 33.22it/s]
Loading 0:  43%|████▎     | 156/363 [00:03<00:05, 38.52it/s]
Loading 0:  44%|████▍     | 161/363 [00:03<00:05, 39.01it/s]
Loading 0:  46%|████▌     | 166/363 [00:04<00:04, 40.96it/s]
Loading 0:  47%|████▋     | 172/363 [00:04<00:04, 40.30it/s]
Loading 0:  49%|████▉     | 177/363 [00:04<00:04, 41.19it/s]
Loading 0:  50%|█████     | 183/363 [00:04<00:03, 45.85it/s]
Loading 0:  52%|█████▏    | 188/363 [00:04<00:03, 45.57it/s]
Loading 0:  53%|█████▎    | 193/363 [00:04<00:03, 46.37it/s]
Loading 0:  55%|█████▍    | 199/363 [00:04<00:03, 42.88it/s]
Loading 0:  56%|█████▌    | 204/363 [00:04<00:03, 41.87it/s]
Loading 0:  58%|█████▊    | 210/363 [00:04<00:03, 45.90it/s]
Loading 0:  59%|█████▉    | 215/363 [00:05<00:03, 44.43it/s]
Loading 0:  61%|██████    | 220/363 [00:05<00:03, 45.02it/s]
Loading 0:  62%|██████▏   | 225/363 [00:05<00:04, 27.92it/s]
Loading 0:  63%|██████▎   | 230/363 [00:05<00:04, 30.42it/s]
Loading 0:  65%|██████▌   | 237/363 [00:05<00:03, 38.12it/s]
Loading 0:  67%|██████▋   | 242/363 [00:05<00:03, 39.48it/s]
Loading 0:  68%|██████▊   | 247/363 [00:06<00:02, 40.62it/s]
Loading 0:  69%|██████▉   | 252/363 [00:06<00:02, 42.67it/s]
Loading 0:  71%|███████   | 257/363 [00:06<00:03, 34.49it/s]
Loading 0:  73%|███████▎  | 265/363 [00:06<00:02, 42.88it/s]
Loading 0:  75%|███████▍  | 271/363 [00:06<00:02, 41.76it/s]
Loading 0:  76%|███████▌  | 276/363 [00:06<00:02, 41.24it/s]
Loading 0:  78%|███████▊  | 282/363 [00:06<00:01, 45.02it/s]
Loading 0:  79%|███████▉  | 287/363 [00:06<00:01, 44.36it/s]
Loading 0:  80%|████████  | 292/363 [00:07<00:01, 44.29it/s]
Loading 0:  82%|████████▏ | 298/363 [00:07<00:01, 41.08it/s]
Loading 0:  83%|████████▎ | 303/363 [00:07<00:01, 43.16it/s]
Loading 0:  85%|████████▍ | 308/363 [00:14<00:22,  2.49it/s]
Loading 0:  86%|████████▌ | 312/363 [00:14<00:15,  3.22it/s]
Loading 0:  88%|████████▊ | 320/363 [00:14<00:08,  5.31it/s]
Loading 0:  90%|████████▉ | 326/363 [00:14<00:05,  7.22it/s]
Loading 0:  91%|█████████ | 331/363 [00:14<00:03,  9.26it/s]
Loading 0:  93%|█████████▎| 337/363 [00:14<00:02, 12.56it/s]
Loading 0:  94%|█████████▍| 342/363 [00:14<00:01, 15.56it/s]
Loading 0:  96%|█████████▌| 347/363 [00:14<00:00, 19.13it/s]
Loading 0:  97%|█████████▋| 353/363 [00:15<00:00, 22.74it/s]
Loading 0:  99%|█████████▊| 358/363 [00:15<00:00, 26.10it/s]
                                                            
Job cloudyu-nemo-dpo-v22-v5-mkmlizer completed after 94.63s with status: succeeded
Stopping job with name cloudyu-nemo-dpo-v22-v5-mkmlizer
Pipeline stage MKMLizer completed in 95.15s
run pipeline stage %s
Running pipeline stage MKMLTemplater
Pipeline stage MKMLTemplater completed in 0.16s
run pipeline stage %s
Running pipeline stage MKMLDeployer
Creating inference service cloudyu-nemo-dpo-v22-v5
Waiting for inference service cloudyu-nemo-dpo-v22-v5 to be ready
Connection pool is full, discarding connection: %s. Connection pool size: %s
Connection pool is full, discarding connection: %s. Connection pool size: %s
Connection pool is full, discarding connection: %s. Connection pool size: %s
Connection pool is full, discarding connection: %s. Connection pool size: %s
Connection pool is full, discarding connection: %s. Connection pool size: %s
Connection pool is full, discarding connection: %s. Connection pool size: %s
Connection pool is full, discarding connection: %s. Connection pool size: %s
Connection pool is full, discarding connection: %s. Connection pool size: %s
Connection pool is full, discarding connection: %s. Connection pool size: %s
Inference service cloudyu-nemo-dpo-v22-v5 ready after 241.23659777641296s
Pipeline stage MKMLDeployer completed in 241.73s
run pipeline stage %s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.1274781227111816s
Received healthy response to inference request in 1.7817444801330566s
Received healthy response to inference request in 1.6878223419189453s
Received healthy response to inference request in 1.7940349578857422s
Received healthy response to inference request in 2.4435789585113525s
5 requests
0 failed requests
5th percentile: 1.7066067695617675
10th percentile: 1.7253911972045899
20th percentile: 1.7629600524902345
30th percentile: 1.7842025756835938
40th percentile: 1.789118766784668
50th percentile: 1.7940349578857422
60th percentile: 1.927412223815918
70th percentile: 2.0607894897460937
80th percentile: 2.1906982898712157
90th percentile: 2.3171386241912844
95th percentile: 2.3803587913513184
99th percentile: 2.4309349250793457
mean time: 1.9669317722320556
Pipeline stage StressChecker completed in 11.37s
run pipeline stage %s
Running pipeline stage OfflineFamilyFriendlyTriggerPipeline
run_pipeline:run_in_cloud %s
starting trigger_guanaco_pipeline args=%s
triggered trigger_guanaco_pipeline args=%s
Pipeline stage OfflineFamilyFriendlyTriggerPipeline completed in 2.43s
run pipeline stage %s
Running pipeline stage TriggerMKMLProfilingPipeline
run_pipeline:run_in_cloud %s
starting trigger_guanaco_pipeline args=%s
triggered trigger_guanaco_pipeline args=%s
Pipeline stage TriggerMKMLProfilingPipeline completed in 2.37s
Shutdown handler de-registered
cloudyu-nemo-dpo-v22_v5 status is now deployed due to DeploymentManager action
Shutdown handler registered
run pipeline %s
run pipeline stage %s
Running pipeline stage OfflineFamilyFriendlyScorer
Evaluating %s Family Friendly Score with %s threads
Pipeline stage OfflineFamilyFriendlyScorer completed in 3840.15s
Shutdown handler de-registered
cloudyu-nemo-dpo-v22_v5 status is now inactive due to auto deactivation removed underperforming models