sao10k-mn-12b-lyra-v4a2

developer_uid: sao10k

submission_id: sao10k-mn-12b-lyra-v4a2_v1

model_name: lyra42

model_group: Sao10K/MN-12B-Lyra-v4a2

status: torndown

timestamp: 2024-09-05T15:46:39+00:00

num_battles: 13225

num_wins: 6930

celo_rating: 1258.97

family_friendly_score: 0.0

submission_type: basic

model_repo: Sao10K/MN-12B-Lyra-v4a2

model_architecture: MistralForCausalLM

model_num_parameters: 12772070400.0

best_of: 8

max_input_tokens: 512

max_output_tokens: 64

latencies: [{'batch_size': 1, 'throughput': 0.6853625498497917, 'latency_mean': 1.4590109074115754, 'latency_p50': 1.4615437984466553, 'latency_p90': 1.6178038835525512}, {'batch_size': 3, 'throughput': 1.3275431789234517, 'latency_mean': 2.2534703612327576, 'latency_p50': 2.2367494106292725, 'latency_p90': 2.5075491189956667}, {'batch_size': 5, 'throughput': 1.5387205187347217, 'latency_mean': 3.2398816430568695, 'latency_p50': 3.234255790710449, 'latency_p90': 3.6099015235900875}, {'batch_size': 6, 'throughput': 1.5893654075401775, 'latency_mean': 3.7541729605197904, 'latency_p50': 3.7732486724853516, 'latency_p90': 4.253945231437683}, {'batch_size': 8, 'throughput': 1.6026308170611085, 'latency_mean': 4.962867916822433, 'latency_p50': 5.003692626953125, 'latency_p90': 5.673020887374878}, {'batch_size': 10, 'throughput': 1.5139866971833797, 'latency_mean': 6.554821333885193, 'latency_p50': 6.587665557861328, 'latency_p90': 7.462587118148804}]

gpu_counts: {'NVIDIA RTX A5000': 1}

display_name: lyra42

is_internal_developer: False

language_model: Sao10K/MN-12B-Lyra-v4a2

model_size: 13B

ranking_group: single

throughput_3p7s: 1.59

us_pacific_date: 2024-09-05

win_ratio: 0.5240075614366729

generation_params: {'temperature': 0.75, 'top_p': 1.0, 'min_p': 0.1, 'top_k': 40, 'presence_penalty': 0.0, 'frequency_penalty': 0.0, 'stopping_words': ['\n', '\n\n', '\nYou:', '[/INST]', '<|im_end|>', '</s>'], 'max_input_tokens': 512, 'best_of': 8, 'max_output_tokens': 64}

formatter: {'memory_template': '<|im_start|>system\n{memory}[/INST]\n', 'prompt_template': '<|im_start|>user\n{prompt}[/INST]\n', 'bot_template': '<|im_start|>assistant\n{bot_name}: {message}[/INST]\n', 'user_template': '<|im_start|>user\n{user_name}: {message}[/INST]\n', 'response_template': '[INST]assistant\n{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 sao10k-mn-12b-lyra-v4a2-v1-mkmlizer
Waiting for job on sao10k-mn-12b-lyra-v4a2-v1-mkmlizer to finish
sao10k-mn-12b-lyra-v4a2-v1-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
sao10k-mn-12b-lyra-v4a2-v1-mkmlizer: ║     _____            __           __                                ║
sao10k-mn-12b-lyra-v4a2-v1-mkmlizer: ║    / _/ /_ ___    __/ /  ___ ___ / /                                ║
sao10k-mn-12b-lyra-v4a2-v1-mkmlizer: ║   / _/ / // / |/|/ / _ \/ -_) -_) /                                 ║
sao10k-mn-12b-lyra-v4a2-v1-mkmlizer: ║  /_//_/\_, /|__,__/_//_/\__/\__/_/                                  ║
sao10k-mn-12b-lyra-v4a2-v1-mkmlizer: ║       /___/                                                         ║
sao10k-mn-12b-lyra-v4a2-v1-mkmlizer: ║                                                                     ║
sao10k-mn-12b-lyra-v4a2-v1-mkmlizer: ║  Version: 0.10.1                                                    ║
sao10k-mn-12b-lyra-v4a2-v1-mkmlizer: ║  Copyright 2023 MK ONE TECHNOLOGIES Inc.                            ║
sao10k-mn-12b-lyra-v4a2-v1-mkmlizer: ║  https://mk1.ai                                                     ║
sao10k-mn-12b-lyra-v4a2-v1-mkmlizer: ║                                                                     ║
sao10k-mn-12b-lyra-v4a2-v1-mkmlizer: ║  The license key for the current software has been verified as      ║
sao10k-mn-12b-lyra-v4a2-v1-mkmlizer: ║  belonging to:                                                      ║
sao10k-mn-12b-lyra-v4a2-v1-mkmlizer: ║                                                                     ║
sao10k-mn-12b-lyra-v4a2-v1-mkmlizer: ║  Chai Research Corp.                                                ║
sao10k-mn-12b-lyra-v4a2-v1-mkmlizer: ║  Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f                   ║
sao10k-mn-12b-lyra-v4a2-v1-mkmlizer: ║  Expiration: 2024-10-15 23:59:59                                    ║
sao10k-mn-12b-lyra-v4a2-v1-mkmlizer: ║                                                                     ║
sao10k-mn-12b-lyra-v4a2-v1-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
sao10k-mn-12b-lyra-v4a2-v1-mkmlizer: Downloaded to shared memory in 49.324s
sao10k-mn-12b-lyra-v4a2-v1-mkmlizer: quantizing model to /dev/shm/model_cache, profile:s0, folder:/tmp/tmpn5lq6an2, device:0
sao10k-mn-12b-lyra-v4a2-v1-mkmlizer: Saving flywheel model at /dev/shm/model_cache
sao10k-mn-12b-lyra-v4a2-v1-mkmlizer: quantized model in 35.979s
sao10k-mn-12b-lyra-v4a2-v1-mkmlizer: Processed model Sao10K/MN-12B-Lyra-v4a2 in 85.303s
sao10k-mn-12b-lyra-v4a2-v1-mkmlizer: creating bucket guanaco-mkml-models
sao10k-mn-12b-lyra-v4a2-v1-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
sao10k-mn-12b-lyra-v4a2-v1-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/sao10k-mn-12b-lyra-v4a2-v1
sao10k-mn-12b-lyra-v4a2-v1-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/sao10k-mn-12b-lyra-v4a2-v1/special_tokens_map.json
sao10k-mn-12b-lyra-v4a2-v1-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/sao10k-mn-12b-lyra-v4a2-v1/config.json
sao10k-mn-12b-lyra-v4a2-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/sao10k-mn-12b-lyra-v4a2-v1/tokenizer_config.json
sao10k-mn-12b-lyra-v4a2-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/sao10k-mn-12b-lyra-v4a2-v1/tokenizer.json
sao10k-mn-12b-lyra-v4a2-v1-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/sao10k-mn-12b-lyra-v4a2-v1/flywheel_model.0.safetensors
sao10k-mn-12b-lyra-v4a2-v1-mkmlizer: 
Loading 0:   0%|          | 0/363 [00:00<?, ?it/s]
Loading 0:   1%|          | 2/363 [00:06<18:06,  3.01s/it]
Loading 0:   2%|▏         | 6/363 [00:06<04:47,  1.24it/s]
Loading 0:   4%|▎         | 13/363 [00:06<01:42,  3.40it/s]
Loading 0:   5%|▍         | 18/363 [00:06<01:03,  5.41it/s]
Loading 0:   6%|▋         | 23/363 [00:06<00:42,  8.00it/s]
Loading 0:   8%|▊         | 28/363 [00:06<00:29, 11.17it/s]
Loading 0:   9%|▉         | 33/363 [00:06<00:23, 13.84it/s]
Loading 0:  11%|█         | 40/363 [00:06<00:16, 20.14it/s]
Loading 0:  12%|█▏        | 45/363 [00:06<00:13, 23.92it/s]
Loading 0:  14%|█▍        | 50/363 [00:07<00:11, 27.78it/s]
Loading 0:  15%|█▌        | 56/363 [00:07<00:10, 30.69it/s]
Loading 0:  17%|█▋        | 61/363 [00:07<00:11, 25.45it/s]
Loading 0:  18%|█▊        | 67/363 [00:07<00:09, 31.16it/s]
Loading 0:  20%|█▉        | 72/363 [00:07<00:08, 34.10it/s]
Loading 0:  21%|██        | 77/363 [00:07<00:07, 37.13it/s]
Loading 0:  23%|██▎       | 82/363 [00:07<00:07, 39.67it/s]
Loading 0:  24%|██▍       | 87/363 [00:08<00:08, 34.47it/s]
Loading 0:  26%|██▌       | 94/363 [00:08<00:06, 41.96it/s]
Loading 0:  27%|██▋       | 99/363 [00:08<00:06, 43.07it/s]
Loading 0:  29%|██▊       | 104/363 [00:08<00:05, 44.19it/s]
Loading 0:  30%|███       | 110/363 [00:08<00:05, 42.45it/s]
Loading 0:  32%|███▏      | 115/363 [00:08<00:05, 41.51it/s]
Loading 0:  33%|███▎      | 121/363 [00:08<00:05, 45.18it/s]
Loading 0:  35%|███▍      | 126/363 [00:08<00:05, 43.39it/s]
Loading 0:  36%|███▌      | 131/363 [00:09<00:05, 42.10it/s]
Loading 0:  37%|███▋      | 136/363 [00:09<00:05, 43.79it/s]
Loading 0:  39%|███▉      | 141/363 [00:09<00:06, 36.69it/s]
Loading 0:  41%|████      | 148/363 [00:09<00:04, 44.08it/s]
Loading 0:  42%|████▏     | 153/363 [00:09<00:04, 42.83it/s]
Loading 0:  44%|████▎     | 158/363 [00:09<00:06, 30.60it/s]
Loading 0:  45%|████▍     | 163/363 [00:10<00:05, 34.28it/s]
Loading 0:  46%|████▋     | 168/363 [00:10<00:06, 30.78it/s]
Loading 0:  48%|████▊     | 175/363 [00:10<00:04, 37.71it/s]
Loading 0:  50%|████▉     | 180/363 [00:10<00:04, 38.87it/s]
Loading 0:  51%|█████     | 185/363 [00:10<00:04, 40.22it/s]
Loading 0:  53%|█████▎    | 191/363 [00:10<00:04, 39.27it/s]
Loading 0:  54%|█████▍    | 196/363 [00:10<00:04, 39.10it/s]
Loading 0:  56%|█████▌    | 202/363 [00:10<00:03, 43.80it/s]
Loading 0:  57%|█████▋    | 207/363 [00:11<00:03, 43.13it/s]
Loading 0:  58%|█████▊    | 212/363 [00:11<00:03, 42.23it/s]
Loading 0:  60%|█████▉    | 217/363 [00:11<00:03, 43.90it/s]
Loading 0:  61%|██████    | 222/363 [00:11<00:03, 36.47it/s]
Loading 0:  63%|██████▎   | 229/363 [00:11<00:03, 43.93it/s]
Loading 0:  64%|██████▍   | 234/363 [00:11<00:02, 44.46it/s]
Loading 0:  66%|██████▌   | 239/363 [00:11<00:02, 44.19it/s]
Loading 0:  67%|██████▋   | 245/363 [00:11<00:02, 42.30it/s]
Loading 0:  69%|██████▉   | 250/363 [00:12<00:02, 40.51it/s]
Loading 0:  71%|███████   | 256/363 [00:12<00:03, 31.51it/s]
Loading 0:  72%|███████▏  | 260/363 [00:12<00:03, 32.59it/s]
Loading 0:  73%|███████▎  | 265/363 [00:12<00:02, 35.95it/s]
Loading 0:  74%|███████▍  | 269/363 [00:12<00:02, 35.99it/s]
Loading 0:  76%|███████▌  | 275/363 [00:12<00:02, 40.32it/s]
Loading 0:  77%|███████▋  | 280/363 [00:12<00:01, 41.87it/s]
Loading 0:  79%|███████▊  | 285/363 [00:13<00:02, 35.21it/s]
Loading 0:  80%|████████  | 292/363 [00:13<00:01, 42.47it/s]
Loading 0:  82%|████████▏ | 297/363 [00:13<00:01, 41.95it/s]
Loading 0:  83%|████████▎ | 302/363 [00:13<00:01, 41.17it/s]
Loading 0:  85%|████████▍ | 308/363 [00:13<00:01, 40.41it/s]
Loading 0:  86%|████████▌ | 313/363 [00:13<00:01, 39.81it/s]
Loading 0:  88%|████████▊ | 319/363 [00:13<00:01, 42.78it/s]
Loading 0:  89%|████████▉ | 324/363 [00:14<00:00, 43.19it/s]
Loading 0:  91%|█████████ | 329/363 [00:14<00:00, 44.00it/s]
Loading 0:  92%|█████████▏| 334/363 [00:14<00:00, 45.05it/s]
Loading 0:  93%|█████████▎| 339/363 [00:14<00:00, 37.64it/s]
Loading 0:  95%|█████████▌| 346/363 [00:14<00:00, 44.07it/s]
Loading 0:  97%|█████████▋| 351/363 [00:14<00:00, 43.68it/s]
Loading 0:  98%|█████████▊| 356/363 [00:14<00:00, 31.28it/s]
Loading 0:  99%|█████████▉| 361/363 [00:15<00:00, 34.91it/s]
                                                            
Job sao10k-mn-12b-lyra-v4a2-v1-mkmlizer completed after 107.06s with status: succeeded
Stopping job with name sao10k-mn-12b-lyra-v4a2-v1-mkmlizer
Pipeline stage MKMLizer completed in 107.84s
run pipeline stage %s
Running pipeline stage MKMLTemplater
Pipeline stage MKMLTemplater completed in 0.09s
run pipeline stage %s
Running pipeline stage MKMLDeployer
Creating inference service sao10k-mn-12b-lyra-v4a2-v1
Waiting for inference service sao10k-mn-12b-lyra-v4a2-v1 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
Inference service sao10k-mn-12b-lyra-v4a2-v1 ready after 140.46243238449097s
Pipeline stage MKMLDeployer completed in 141.03s
run pipeline stage %s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.242931842803955s
Received healthy response to inference request in 1.9962458610534668s
Received healthy response to inference request in 1.820894718170166s
Received healthy response to inference request in 2.1069960594177246s
Received healthy response to inference request in 2.1953511238098145s
5 requests
0 failed requests
5th percentile: 1.8559649467468262
10th percentile: 1.8910351753234864
20th percentile: 1.9611756324768066
30th percentile: 2.0183959007263184
40th percentile: 2.0626959800720215
50th percentile: 2.1069960594177246
60th percentile: 2.1423380851745604
70th percentile: 2.1776801109313966
80th percentile: 2.2048672676086425
90th percentile: 2.223899555206299
95th percentile: 2.233415699005127
99th percentile: 2.2410286140441893
mean time: 2.0724839210510253
Pipeline stage StressChecker completed in 11.85s
run pipeline stage %s
Running pipeline stage TriggerMKMLProfilingPipeline
run_pipeline:run_in_cloud %s
starting trigger_guanaco_pipeline args=%s
Pipeline stage TriggerMKMLProfilingPipeline completed in 4.85s
Shutdown handler de-registered
sao10k-mn-12b-lyra-v4a2_v1 status is now deployed due to DeploymentManager action
Shutdown handler registered
run pipeline %s
run pipeline stage %s
Running pipeline stage MKMLProfilerDeleter
Skipping teardown as no inference service was successfully deployed
Pipeline stage MKMLProfilerDeleter completed in 0.12s
run pipeline stage %s
Running pipeline stage MKMLProfilerTemplater
Pipeline stage MKMLProfilerTemplater completed in 0.12s
run pipeline stage %s
Running pipeline stage MKMLProfilerDeployer
Creating inference service sao10k-mn-12b-lyra-v4a2-v1-profiler
Waiting for inference service sao10k-mn-12b-lyra-v4a2-v1-profiler to be ready
Inference service sao10k-mn-12b-lyra-v4a2-v1-profiler ready after 140.33778977394104s
Pipeline stage MKMLProfilerDeployer completed in 140.69s
run pipeline stage %s
Running pipeline stage MKMLProfilerRunner
kubectl cp /code/guanaco/guanaco_inference_services/src/inference_scripts tenant-chaiml-guanaco/sao10k-mn-12b-lyra-v4a2-v1-profiler-predictor-00001-deployk26js:/code/chaiverse_profiler_1725551643 --namespace tenant-chaiml-guanaco
kubectl exec -it sao10k-mn-12b-lyra-v4a2-v1-profiler-predictor-00001-deployk26js --namespace tenant-chaiml-guanaco -- sh -c 'cd /code/chaiverse_profiler_1725551643 && python profiles.py profile --best_of_n 8 --auto_batch 5 --batches 1,5,10,15,20,25,30,35,40,45,50,55,60,65,70,75,80,85,90,95,100,105,110,115,120,125,130,135,140,145,150,155,160,165,170,175,180,185,190,195 --samples 200 --input_tokens 512 --output_tokens 64 --summary /code/chaiverse_profiler_1725551643/summary.json'
kubectl exec -it sao10k-mn-12b-lyra-v4a2-v1-profiler-predictor-00001-deployk26js --namespace tenant-chaiml-guanaco -- bash -c 'cat /code/chaiverse_profiler_1725551643/summary.json'
Pipeline stage MKMLProfilerRunner completed in 960.61s
run pipeline stage %s
Running pipeline stage MKMLProfilerDeleter
Checking if service sao10k-mn-12b-lyra-v4a2-v1-profiler is running
Tearing down inference service sao10k-mn-12b-lyra-v4a2-v1-profiler
Service sao10k-mn-12b-lyra-v4a2-v1-profiler has been torndown
Pipeline stage MKMLProfilerDeleter completed in 2.34s
Shutdown handler de-registered
sao10k-mn-12b-lyra-v4a2_v1 status is now inactive due to auto deactivation removed underperforming models
sao10k-mn-12b-lyra-v4a2_v1 status is now torndown due to DeploymentManager action