cloudyu-chaiml-nemo-dpo-v2

developer_uid: cloudyu

submission_id: cloudyu-chaiml-nemo-dpo-v2_v1

model_name: cloudyu-chaiml-nemo-dpo-v2_v1

model_group: cloudyu/ChaiML-Nemo-DPO-

status: inactive

timestamp: 2024-11-24T04:55:53+00:00

num_battles: 22131

num_wins: 10625

celo_rating: 1245.5

family_friendly_score: 0.5724

family_friendly_standard_error: 0.006996545433283485

submission_type: basic

model_repo: cloudyu/ChaiML-Nemo-DPO-V2

model_architecture: MistralForCausalLM

model_num_parameters: 12772070400.0

best_of: 8

max_input_tokens: 1024

max_output_tokens: 64

latencies: [{'batch_size': 1, 'throughput': 0.608003308666687, 'latency_mean': 1.6446350753307342, 'latency_p50': 1.6356662511825562, 'latency_p90': 1.8237297534942627}, {'batch_size': 3, 'throughput': 1.118331383229901, 'latency_mean': 2.673145030736923, 'latency_p50': 2.6990216970443726, 'latency_p90': 2.966651964187622}, {'batch_size': 5, 'throughput': 1.3536730593116386, 'latency_mean': 3.683087363243103, 'latency_p50': 3.6976112127304077, 'latency_p90': 4.0862257242202755}, {'batch_size': 6, 'throughput': 1.4129673250514418, 'latency_mean': 4.225499775409698, 'latency_p50': 4.222665309906006, 'latency_p90': 4.718549513816834}, {'batch_size': 8, 'throughput': 1.4834557475241437, 'latency_mean': 5.369685448408127, 'latency_p50': 5.391619920730591, 'latency_p90': 6.131701064109802}, {'batch_size': 10, 'throughput': 1.5165684493104925, 'latency_mean': 6.5525931239128115, 'latency_p50': 6.5140604972839355, 'latency_p90': 7.347312378883362}]

gpu_counts: {'NVIDIA RTX A5000': 1}

display_name: cloudyu-chaiml-nemo-dpo-v2_v1

is_internal_developer: False

language_model: cloudyu/ChaiML-Nemo-DPO-V2

model_size: 13B

ranking_group: single

throughput_3p7s: 1.36

us_pacific_date: 2024-11-23

win_ratio: 0.48009579323121415

generation_params: {'temperature': 1.0, 'top_p': 1.0, 'min_p': 0.0, 'top_k': 40, '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-chaiml-nemo-dpo-v2-v1-mkmlizer
Waiting for job on cloudyu-chaiml-nemo-dpo-v2-v1-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
cloudyu-chaiml-nemo-dpo-v2-v1-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
cloudyu-chaiml-nemo-dpo-v2-v1-mkmlizer: ║     _____            __           __                                ║
cloudyu-chaiml-nemo-dpo-v2-v1-mkmlizer: ║    / _/ /_ ___    __/ /  ___ ___ / /                                ║
cloudyu-chaiml-nemo-dpo-v2-v1-mkmlizer: ║   / _/ / // / |/|/ / _ \/ -_) -_) /                                 ║
cloudyu-chaiml-nemo-dpo-v2-v1-mkmlizer: ║  /_//_/\_, /|__,__/_//_/\__/\__/_/                                  ║
cloudyu-chaiml-nemo-dpo-v2-v1-mkmlizer: ║       /___/                                                         ║
cloudyu-chaiml-nemo-dpo-v2-v1-mkmlizer: ║                                                                     ║
cloudyu-chaiml-nemo-dpo-v2-v1-mkmlizer: ║  Version: 0.11.12                                                   ║
cloudyu-chaiml-nemo-dpo-v2-v1-mkmlizer: ║  Copyright 2023 MK ONE TECHNOLOGIES Inc.                            ║
cloudyu-chaiml-nemo-dpo-v2-v1-mkmlizer: ║  https://mk1.ai                                                     ║
cloudyu-chaiml-nemo-dpo-v2-v1-mkmlizer: ║                                                                     ║
cloudyu-chaiml-nemo-dpo-v2-v1-mkmlizer: ║  The license key for the current software has been verified as      ║
cloudyu-chaiml-nemo-dpo-v2-v1-mkmlizer: ║  belonging to:                                                      ║
cloudyu-chaiml-nemo-dpo-v2-v1-mkmlizer: ║                                                                     ║
cloudyu-chaiml-nemo-dpo-v2-v1-mkmlizer: ║  Chai Research Corp.                                                ║
cloudyu-chaiml-nemo-dpo-v2-v1-mkmlizer: ║  Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f                   ║
cloudyu-chaiml-nemo-dpo-v2-v1-mkmlizer: ║  Expiration: 2025-01-15 23:59:59                                    ║
cloudyu-chaiml-nemo-dpo-v2-v1-mkmlizer: ║                                                                     ║
cloudyu-chaiml-nemo-dpo-v2-v1-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
cloudyu-chaiml-nemo-dpo-v2-v1-mkmlizer: Downloaded to shared memory in 49.315s
cloudyu-chaiml-nemo-dpo-v2-v1-mkmlizer: quantizing model to /dev/shm/model_cache, profile:s0, folder:/tmp/tmp7p1wdehv, device:0
cloudyu-chaiml-nemo-dpo-v2-v1-mkmlizer: Saving flywheel model at /dev/shm/model_cache
cloudyu-chaiml-nemo-dpo-v2-v1-mkmlizer: quantized model in 35.971s
cloudyu-chaiml-nemo-dpo-v2-v1-mkmlizer: Processed model cloudyu/ChaiML-Nemo-DPO-V2 in 85.286s
cloudyu-chaiml-nemo-dpo-v2-v1-mkmlizer: creating bucket guanaco-mkml-models
cloudyu-chaiml-nemo-dpo-v2-v1-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
cloudyu-chaiml-nemo-dpo-v2-v1-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/cloudyu-chaiml-nemo-dpo-v2-v1
cloudyu-chaiml-nemo-dpo-v2-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/cloudyu-chaiml-nemo-dpo-v2-v1/tokenizer.json
cloudyu-chaiml-nemo-dpo-v2-v1-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/cloudyu-chaiml-nemo-dpo-v2-v1/flywheel_model.0.safetensors
cloudyu-chaiml-nemo-dpo-v2-v1-mkmlizer: 
Loading 0:   0%|          | 0/363 [00:00<?, ?it/s]
Loading 0:   1%|▏         | 5/363 [00:00<00:12, 27.72it/s]
Loading 0:   4%|▎         | 13/363 [00:00<00:07, 48.21it/s]
Loading 0:   5%|▌         | 19/363 [00:00<00:07, 45.93it/s]
Loading 0:   7%|▋         | 24/363 [00:00<00:07, 43.89it/s]
Loading 0:   9%|▊         | 31/363 [00:00<00:06, 50.10it/s]
Loading 0:  10%|█         | 37/363 [00:00<00:06, 46.66it/s]
Loading 0:  12%|█▏        | 42/363 [00:00<00:07, 44.34it/s]
Loading 0:  13%|█▎        | 49/363 [00:01<00:06, 48.58it/s]
Loading 0:  15%|█▍        | 54/363 [00:01<00:06, 48.17it/s]
Loading 0:  17%|█▋        | 60/363 [00:01<00:06, 43.55it/s]
Loading 0:  18%|█▊        | 65/363 [00:01<00:10, 29.25it/s]
Loading 0:  20%|█▉        | 72/363 [00:01<00:08, 35.71it/s]
Loading 0:  21%|██▏       | 78/363 [00:01<00:07, 37.72it/s]
Loading 0:  23%|██▎       | 83/363 [00:02<00:07, 39.16it/s]
Loading 0:  25%|██▍       | 90/363 [00:02<00:06, 45.18it/s]
Loading 0:  26%|██▋       | 96/363 [00:02<00:06, 44.20it/s]
Loading 0:  28%|██▊       | 101/363 [00:02<00:05, 43.68it/s]
Loading 0:  30%|███       | 109/363 [00:02<00:04, 51.99it/s]
Loading 0:  32%|███▏      | 115/363 [00:02<00:05, 47.28it/s]
Loading 0:  33%|███▎      | 121/363 [00:02<00:05, 44.97it/s]
Loading 0:  35%|███▍      | 127/363 [00:02<00:05, 41.62it/s]
Loading 0:  37%|███▋      | 135/363 [00:03<00:04, 48.40it/s]
Loading 0:  39%|███▉      | 141/363 [00:03<00:04, 45.88it/s]
Loading 0:  40%|████      | 146/363 [00:03<00:06, 33.02it/s]
Loading 0:  41%|████▏     | 150/363 [00:03<00:06, 32.67it/s]
Loading 0:  43%|████▎     | 157/363 [00:03<00:05, 38.55it/s]
Loading 0:  45%|████▍     | 162/363 [00:03<00:04, 40.94it/s]
Loading 0:  46%|████▌     | 167/363 [00:04<00:05, 36.54it/s]
Loading 0:  48%|████▊     | 175/363 [00:04<00:04, 44.78it/s]
Loading 0:  50%|████▉     | 180/363 [00:04<00:03, 45.99it/s]
Loading 0:  51%|█████     | 185/363 [00:04<00:04, 38.52it/s]
Loading 0:  53%|█████▎    | 193/363 [00:04<00:03, 46.39it/s]
Loading 0:  55%|█████▍    | 199/363 [00:04<00:03, 44.83it/s]
Loading 0:  56%|█████▌    | 204/363 [00:04<00:03, 43.50it/s]
Loading 0:  58%|█████▊    | 211/363 [00:04<00:03, 48.41it/s]
Loading 0:  60%|█████▉    | 217/363 [00:05<00:03, 45.44it/s]
Loading 0:  61%|██████▏   | 223/363 [00:05<00:04, 34.97it/s]
Loading 0:  63%|██████▎   | 227/363 [00:05<00:03, 34.85it/s]
Loading 0:  64%|██████▎   | 231/363 [00:05<00:03, 33.79it/s]
Loading 0:  66%|██████▌   | 238/363 [00:05<00:03, 40.18it/s]
Loading 0:  67%|██████▋   | 244/363 [00:05<00:02, 39.93it/s]
Loading 0:  69%|██████▊   | 249/363 [00:05<00:02, 40.45it/s]
Loading 0:  71%|███████   | 256/363 [00:06<00:02, 45.46it/s]
Loading 0:  72%|███████▏  | 261/363 [00:06<00:02, 45.96it/s]
Loading 0:  73%|███████▎  | 266/363 [00:06<00:02, 38.89it/s]
Loading 0:  75%|███████▌  | 274/363 [00:06<00:01, 46.51it/s]
Loading 0:  77%|███████▋  | 280/363 [00:06<00:01, 45.14it/s]
Loading 0:  79%|███████▊  | 285/363 [00:06<00:01, 44.52it/s]
Loading 0:  80%|████████  | 292/363 [00:06<00:01, 49.32it/s]
Loading 0:  82%|████████▏ | 298/363 [00:07<00:01, 45.58it/s]
Loading 0:  84%|████████▎ | 304/363 [00:13<00:20,  2.85it/s]
Loading 0:  85%|████████▍ | 308/363 [00:13<00:15,  3.56it/s]
Loading 0:  86%|████████▌ | 312/363 [00:14<00:11,  4.53it/s]
Loading 0:  88%|████████▊ | 320/363 [00:14<00:05,  7.32it/s]
Loading 0:  90%|████████▉ | 326/363 [00:14<00:03,  9.73it/s]
Loading 0:  91%|█████████ | 331/363 [00:14<00:02, 12.27it/s]
Loading 0:  93%|█████████▎| 338/363 [00:14<00:01, 17.08it/s]
Loading 0:  95%|█████████▍| 344/363 [00:14<00:00, 20.66it/s]
Loading 0:  96%|█████████▌| 349/363 [00:14<00:00, 23.87it/s]
Loading 0:  98%|█████████▊| 356/363 [00:14<00:00, 30.26it/s]
Loading 0: 100%|█████████▉| 362/363 [00:15<00:00, 32.46it/s]
                                                            
Job cloudyu-chaiml-nemo-dpo-v2-v1-mkmlizer completed after 103.88s with status: succeeded
Stopping job with name cloudyu-chaiml-nemo-dpo-v2-v1-mkmlizer
Pipeline stage MKMLizer completed in 104.37s
run pipeline stage %s
Running pipeline stage MKMLTemplater
Pipeline stage MKMLTemplater completed in 0.15s
run pipeline stage %s
Running pipeline stage MKMLDeployer
Creating inference service cloudyu-chaiml-nemo-dpo-v2-v1
Waiting for inference service cloudyu-chaiml-nemo-dpo-v2-v1 to be ready
Inference service cloudyu-chaiml-nemo-dpo-v2-v1 ready after 110.43642091751099s
Pipeline stage MKMLDeployer completed in 110.97s
run pipeline stage %s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.3992018699645996s
Received healthy response to inference request in 1.7128934860229492s
Received healthy response to inference request in 2.042638063430786s
Received healthy response to inference request in 1.8351788520812988s
Received healthy response to inference request in 2.212700366973877s
5 requests
0 failed requests
5th percentile: 1.7373505592346192
10th percentile: 1.761807632446289
20th percentile: 1.8107217788696288
30th percentile: 1.8766706943511964
40th percentile: 1.9596543788909913
50th percentile: 2.042638063430786
60th percentile: 2.1106629848480223
70th percentile: 2.178687906265259
80th percentile: 2.2500006675720217
90th percentile: 2.3246012687683106
95th percentile: 2.361901569366455
99th percentile: 2.3917418098449708
mean time: 2.0405225276947023
Pipeline stage StressChecker completed in 11.54s
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 2.39s
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 1.99s
Shutdown handler de-registered
cloudyu-chaiml-nemo-dpo-v2_v1 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 2793.94s
Shutdown handler de-registered
cloudyu-chaiml-nemo-dpo-v2_v1 status is now inactive due to auto deactivation removed underperforming models