sao10k-lyra-4-2-chai

submission_id: sao10k-lyra-4-2-chai_v4

developer_uid: sao10k

best_of: 8

celo_rating: 1271.1

display_name: lyra42x

family_friendly_score: 0.5595154903331003

family_friendly_standard_error: 0.009260153983625619

formatter: {'memory_template': '<|im_start|>system\n{memory}<|im_end|>\n', 'prompt_template': '<|im_start|>user\n{prompt}<|im_end|>\n', 'bot_template': '<|im_start|>assistant\n{bot_name}: {message}<|im_end|>\n', 'user_template': '<|im_start|>{user_name}\n{message}<|im_end|>\n', 'response_template': '<|im_start|>assistant\n{bot_name}:', 'truncate_by_message': False}

generation_params: {'temperature': 1.0, 'top_p': 1.0, 'min_p': 0.2, 'top_k': 50, 'presence_penalty': 0.0, 'frequency_penalty': 0.0, 'stopping_words': ['\n\n', '\nYou:', '[/INST]', '<|im_end|>', '</s>'], 'max_input_tokens': 512, 'best_of': 8, 'max_output_tokens': 64}

gpu_counts: {'NVIDIA RTX A5000': 1}

ineligible_reason: num_battles<5000

is_internal_developer: False

language_model: Sao10K/lyra-4-2-chai

latencies: [{'batch_size': 1, 'throughput': 0.683595261958705, 'latency_mean': 1.4627919149398805, 'latency_p50': 1.4670065641403198, 'latency_p90': 1.6162002325057983}, {'batch_size': 3, 'throughput': 1.311174944926516, 'latency_mean': 2.284343168735504, 'latency_p50': 2.3068336248397827, 'latency_p90': 2.5150757789611817}, {'batch_size': 5, 'throughput': 1.5523758962188237, 'latency_mean': 3.193103388547897, 'latency_p50': 3.17967426776886, 'latency_p90': 3.6056304931640626}, {'batch_size': 6, 'throughput': 1.6167526077935854, 'latency_mean': 3.6976825737953187, 'latency_p50': 3.701964020729065, 'latency_p90': 4.204604864120483}, {'batch_size': 8, 'throughput': 1.5889546455485224, 'latency_mean': 4.997425584793091, 'latency_p50': 4.998100280761719, 'latency_p90': 5.680343770980834}, {'batch_size': 10, 'throughput': 1.549748404167658, 'latency_mean': 6.411097681522369, 'latency_p50': 6.418901443481445, 'latency_p90': 7.353355193138122}]

max_input_tokens: 512

max_output_tokens: 64

model_architecture: MistralForCausalLM

model_group: Sao10K/lyra-4-2-chai

model_name: lyra42x

model_num_parameters: 12772070400.0

model_repo: Sao10K/lyra-4-2-chai

model_size: 13B

num_battles: 2941

num_wins: 1537

ranking_group: single

status: torndown

submission_type: basic

throughput_3p7s: 1.62

timestamp: 2024-09-26T06:36:24+00:00

us_pacific_date: 2024-09-25

win_ratio: 0.5226113566814009

Download Preference Data

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-lyra-4-2-chai-v4-mkmlizer
Waiting for job on sao10k-lyra-4-2-chai-v4-mkmlizer to finish
sao10k-lyra-4-2-chai-v4-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
sao10k-lyra-4-2-chai-v4-mkmlizer: ║     _____            __           __                                ║
sao10k-lyra-4-2-chai-v4-mkmlizer: ║    / _/ /_ ___    __/ /  ___ ___ / /                                ║
sao10k-lyra-4-2-chai-v4-mkmlizer: ║   / _/ / // / |/|/ / _ \/ -_) -_) /                                 ║
sao10k-lyra-4-2-chai-v4-mkmlizer: ║  /_//_/\_, /|__,__/_//_/\__/\__/_/                                  ║
sao10k-lyra-4-2-chai-v4-mkmlizer: ║       /___/                                                         ║
sao10k-lyra-4-2-chai-v4-mkmlizer: ║                                                                     ║
sao10k-lyra-4-2-chai-v4-mkmlizer: ║  Version: 0.11.12                                                   ║
sao10k-lyra-4-2-chai-v4-mkmlizer: ║  Copyright 2023 MK ONE TECHNOLOGIES Inc.                            ║
sao10k-lyra-4-2-chai-v4-mkmlizer: ║  https://mk1.ai                                                     ║
sao10k-lyra-4-2-chai-v4-mkmlizer: ║                                                                     ║
sao10k-lyra-4-2-chai-v4-mkmlizer: ║  The license key for the current software has been verified as      ║
sao10k-lyra-4-2-chai-v4-mkmlizer: ║  belonging to:                                                      ║
sao10k-lyra-4-2-chai-v4-mkmlizer: ║                                                                     ║
sao10k-lyra-4-2-chai-v4-mkmlizer: ║  Chai Research Corp.                                                ║
sao10k-lyra-4-2-chai-v4-mkmlizer: ║  Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f                   ║
sao10k-lyra-4-2-chai-v4-mkmlizer: ║  Expiration: 2024-10-15 23:59:59                                    ║
sao10k-lyra-4-2-chai-v4-mkmlizer: ║                                                                     ║
sao10k-lyra-4-2-chai-v4-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
sao10k-lyra-4-2-chai-v4-mkmlizer: Downloaded to shared memory in 42.556s
sao10k-lyra-4-2-chai-v4-mkmlizer: quantizing model to /dev/shm/model_cache, profile:s0, folder:/tmp/tmp8zlqylyu, device:0
sao10k-lyra-4-2-chai-v4-mkmlizer: Saving flywheel model at /dev/shm/model_cache
sao10k-lyra-4-2-chai-v4-mkmlizer: quantized model in 37.545s
sao10k-lyra-4-2-chai-v4-mkmlizer: Processed model Sao10K/lyra-4-2-chai in 80.102s
sao10k-lyra-4-2-chai-v4-mkmlizer: creating bucket guanaco-mkml-models
sao10k-lyra-4-2-chai-v4-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
sao10k-lyra-4-2-chai-v4-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/sao10k-lyra-4-2-chai-v4
sao10k-lyra-4-2-chai-v4-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/sao10k-lyra-4-2-chai-v4/special_tokens_map.json
sao10k-lyra-4-2-chai-v4-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/sao10k-lyra-4-2-chai-v4/config.json
sao10k-lyra-4-2-chai-v4-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/sao10k-lyra-4-2-chai-v4/tokenizer_config.json
sao10k-lyra-4-2-chai-v4-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/sao10k-lyra-4-2-chai-v4/tokenizer.json
sao10k-lyra-4-2-chai-v4-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/sao10k-lyra-4-2-chai-v4/flywheel_model.0.safetensors
sao10k-lyra-4-2-chai-v4-mkmlizer: 
Loading 0:   0%|          | 0/363 [00:00<?, ?it/s]
Loading 0:   1%|▏         | 5/363 [00:00<00:11, 30.37it/s]
Loading 0:   4%|▍         | 14/363 [00:00<00:07, 45.60it/s]
Loading 0:   5%|▌         | 19/363 [00:00<00:11, 30.36it/s]
Loading 0:   6%|▋         | 23/363 [00:00<00:11, 29.55it/s]
Loading 0:   8%|▊         | 30/363 [00:00<00:08, 38.03it/s]
Loading 0:  10%|▉         | 35/363 [00:00<00:08, 38.34it/s]
Loading 0:  11%|█         | 40/363 [00:01<00:09, 32.89it/s]
Loading 0:  12%|█▏        | 44/363 [00:01<00:11, 28.24it/s]
Loading 0:  14%|█▍        | 50/363 [00:01<00:10, 30.90it/s]
Loading 0:  16%|█▌        | 57/363 [00:01<00:08, 38.07it/s]
Loading 0:  17%|█▋        | 62/363 [00:01<00:07, 38.41it/s]
Loading 0:  18%|█▊        | 67/363 [00:01<00:07, 39.39it/s]
Loading 0:  20%|█▉        | 72/363 [00:02<00:07, 40.45it/s]
Loading 0:  21%|██        | 77/363 [00:02<00:07, 37.58it/s]
Loading 0:  22%|██▏       | 81/363 [00:02<00:10, 27.54it/s]
Loading 0:  24%|██▍       | 87/363 [00:02<00:08, 32.74it/s]
Loading 0:  25%|██▌       | 91/363 [00:02<00:08, 33.01it/s]
Loading 0:  26%|██▌       | 95/363 [00:02<00:08, 32.80it/s]
Loading 0:  27%|██▋       | 99/363 [00:02<00:08, 32.55it/s]
Loading 0:  28%|██▊       | 103/363 [00:03<00:08, 32.21it/s]
Loading 0:  29%|██▉       | 107/363 [00:03<00:09, 27.07it/s]
Loading 0:  31%|███       | 113/363 [00:03<00:08, 29.01it/s]
Loading 0:  33%|███▎      | 120/363 [00:03<00:06, 35.71it/s]
Loading 0:  34%|███▍      | 124/363 [00:03<00:06, 34.43it/s]
Loading 0:  35%|███▌      | 128/363 [00:03<00:06, 35.70it/s]
Loading 0:  36%|███▋      | 132/363 [00:03<00:07, 32.96it/s]
Loading 0:  38%|███▊      | 139/363 [00:04<00:05, 41.38it/s]
Loading 0:  40%|███▉      | 144/363 [00:04<00:07, 30.29it/s]
Loading 0:  41%|████      | 149/363 [00:04<00:06, 32.25it/s]
Loading 0:  43%|████▎     | 157/363 [00:04<00:05, 41.05it/s]
Loading 0:  45%|████▍     | 163/363 [00:04<00:04, 42.14it/s]
Loading 0:  46%|████▋     | 168/363 [00:04<00:04, 41.70it/s]
Loading 0:  48%|████▊     | 173/363 [00:04<00:05, 37.70it/s]
Loading 0:  49%|████▉     | 178/363 [00:05<00:04, 38.22it/s]
Loading 0:  51%|█████     | 184/363 [00:05<00:04, 41.74it/s]
Loading 0:  52%|█████▏    | 190/363 [00:05<00:04, 41.40it/s]
Loading 0:  54%|█████▎    | 195/363 [00:05<00:04, 41.43it/s]
Loading 0:  56%|█████▌    | 203/363 [00:05<00:03, 46.95it/s]
Loading 0:  57%|█████▋    | 208/363 [00:05<00:04, 36.49it/s]
Loading 0:  58%|█████▊    | 212/363 [00:05<00:04, 35.03it/s]
Loading 0:  60%|██████    | 219/363 [00:06<00:03, 41.92it/s]
Loading 0:  62%|██████▏   | 224/363 [00:06<00:03, 41.52it/s]
Loading 0:  63%|██████▎   | 229/363 [00:06<00:03, 36.39it/s]
Loading 0:  64%|██████▍   | 233/363 [00:06<00:03, 32.63it/s]
Loading 0:  66%|██████▌   | 239/363 [00:06<00:03, 34.84it/s]
Loading 0:  68%|██████▊   | 246/363 [00:06<00:02, 41.44it/s]
Loading 0:  69%|██████▉   | 251/363 [00:06<00:02, 41.57it/s]
Loading 0:  71%|███████   | 256/363 [00:07<00:02, 42.80it/s]
Loading 0:  72%|███████▏  | 262/363 [00:07<00:02, 41.88it/s]
Loading 0:  74%|███████▎  | 267/363 [00:07<00:02, 36.32it/s]
Loading 0:  75%|███████▍  | 271/363 [00:07<00:02, 34.78it/s]
Loading 0:  76%|███████▌  | 275/363 [00:07<00:02, 33.57it/s]
Loading 0:  78%|███████▊  | 282/363 [00:07<00:01, 41.00it/s]
Loading 0:  79%|███████▉  | 287/363 [00:07<00:01, 40.54it/s]
Loading 0:  80%|████████  | 292/363 [00:08<00:02, 34.33it/s]
Loading 0:  82%|████████▏ | 296/363 [00:08<00:02, 29.05it/s]
Loading 0:  83%|████████▎ | 302/363 [00:08<00:01, 31.27it/s]
Loading 0:  85%|████████▌ | 310/363 [00:08<00:01, 39.62it/s]
Loading 0:  87%|████████▋ | 315/363 [00:08<00:01, 41.84it/s]
Loading 0:  88%|████████▊ | 320/363 [00:08<00:01, 34.83it/s]
Loading 0:  91%|█████████ | 329/363 [00:09<00:00, 41.70it/s]
Loading 0:  92%|█████████▏| 334/363 [00:09<00:00, 34.03it/s]
Loading 0:  93%|█████████▎| 338/363 [00:09<00:00, 32.04it/s]
Loading 0:  95%|█████████▌| 345/363 [00:09<00:00, 38.71it/s]
Loading 0:  96%|█████████▋| 350/363 [00:09<00:00, 39.23it/s]
Loading 0:  98%|█████████▊| 355/363 [00:09<00:00, 33.29it/s]
Loading 0:  99%|█████████▉| 359/363 [00:16<00:01,  2.33it/s]
                                                            
Job sao10k-lyra-4-2-chai-v4-mkmlizer completed after 113.19s with status: succeeded
Stopping job with name sao10k-lyra-4-2-chai-v4-mkmlizer
Pipeline stage MKMLizer completed in 113.92s
run pipeline stage %s
Running pipeline stage MKMLTemplater
Pipeline stage MKMLTemplater completed in 0.08s
run pipeline stage %s
Running pipeline stage MKMLDeployer
Creating inference service sao10k-lyra-4-2-chai-v4
Waiting for inference service sao10k-lyra-4-2-chai-v4 to be ready
Inference service sao10k-lyra-4-2-chai-v4 ready after 220.4857075214386s
Pipeline stage MKMLDeployer completed in 220.78s
run pipeline stage %s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.2762374877929688s
Received healthy response to inference request in 1.6228973865509033s
Received healthy response to inference request in 1.6330046653747559s
Received healthy response to inference request in 2.0069963932037354s
Received healthy response to inference request in 1.5273878574371338s
5 requests
0 failed requests
5th percentile: 1.5464897632598877
10th percentile: 1.5655916690826417
20th percentile: 1.6037954807281494
30th percentile: 1.6249188423156737
40th percentile: 1.6289617538452148
50th percentile: 1.6330046653747559
60th percentile: 1.7826013565063477
70th percentile: 1.9321980476379395
80th percentile: 2.060844612121582
90th percentile: 2.1685410499572755
95th percentile: 2.222389268875122
99th percentile: 2.2654678440093994
mean time: 1.8133047580718995
Pipeline stage StressChecker completed in 9.70s
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 3.62s
Shutdown handler de-registered
sao10k-lyra-4-2-chai_v4 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 0.11s
run pipeline stage %s
Running pipeline stage MKMLProfilerDeployer
Creating inference service sao10k-lyra-4-2-chai-v4-profiler
Waiting for inference service sao10k-lyra-4-2-chai-v4-profiler to be ready
Inference service sao10k-lyra-4-2-chai-v4-profiler ready after 220.49238276481628s
Pipeline stage MKMLProfilerDeployer completed in 220.85s
run pipeline stage %s
Running pipeline stage MKMLProfilerRunner
kubectl cp /code/guanaco/guanaco_inference_services/src/inference_scripts tenant-chaiml-guanaco/sao10k-lyra-4-2-chai-v4-profiler-predictor-00001-deploymen6j9jj:/code/chaiverse_profiler_1727333203 --namespace tenant-chaiml-guanaco
kubectl exec -it sao10k-lyra-4-2-chai-v4-profiler-predictor-00001-deploymen6j9jj --namespace tenant-chaiml-guanaco -- sh -c 'cd /code/chaiverse_profiler_1727333203 && 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_1727333203/summary.json'
kubectl exec -it sao10k-lyra-4-2-chai-v4-profiler-predictor-00001-deploymen6j9jj --namespace tenant-chaiml-guanaco -- bash -c 'cat /code/chaiverse_profiler_1727333203/summary.json'
Pipeline stage MKMLProfilerRunner completed in 958.85s
run pipeline stage %s
Running pipeline stage MKMLProfilerDeleter
Checking if service sao10k-lyra-4-2-chai-v4-profiler is running
Tearing down inference service sao10k-lyra-4-2-chai-v4-profiler
Service sao10k-lyra-4-2-chai-v4-profiler has been torndown
Pipeline stage MKMLProfilerDeleter completed in 1.97s
Shutdown handler de-registered
sao10k-lyra-4-2-chai_v4 status is now inactive due to auto deactivation removed underperforming models
Shutdown handler not registered because Python interpreter is not running in the main thread
Waiting for inference service blend-rofur-2024-10-03 to be ready
run pipeline %s
Shutdown handler not registered because Python interpreter is not running in the main thread
Deleting key rica40325-mistral-3678-v4/config.json from bucket guanaco-mkml-models
admin requested tearing down of sao10k-lyra-4-2-chai_v4
run pipeline %s
run pipeline stage %s
run pipeline %s
Deleting key rica40325-mistral-3678-v4/flywheel_model.0.safetensors from bucket guanaco-mkml-models
Shutdown handler not registered because Python interpreter is not running in the main thread
run pipeline stage %s
Running pipeline stage MKMLDeleter
admin requested tearing down of blend_rofur_2024-10-03
sao10k-lyra-4-2-chai_v4 status is now torndown due to DeploymentManager action