submission_id: zonemercy-vingt-deux-v1-1e5_v26
developer_uid: chai_backend_admin
best_of: 8
celo_rating: 1240.09
display_name: temp-5
family_friendly_score: 0.5887751500176491
family_friendly_standard_error: 0.009226053608853174
formatter: {'memory_template': '', 'prompt_template': '', 'bot_template': 'Bot: {message}\n', 'user_template': 'User: {message}\n', 'response_template': 'Bot:', 'truncate_by_message': False}
generation_params: {'temperature': 0.9, 'top_p': 1.0, 'min_p': 0.05, 'top_k': 80, 'presence_penalty': 0.0, 'frequency_penalty': 0.0, 'stopping_words': ['\n', '</s>', '####', 'Bot:', 'User:', 'You:', '<|im_end|>', '<|eot_id|>'], 'max_input_tokens': 1024, 'best_of': 8, 'max_output_tokens': 64}
gpu_counts: {'NVIDIA RTX A6000': 1}
ineligible_reason: num_battles<5000
is_internal_developer: True
language_model: zonemercy/Vingt-Deux-v1-1e5
latencies: [{'batch_size': 1, 'throughput': 0.38194169513100756, 'latency_mean': 2.6181399965286256, 'latency_p50': 2.6172034740448, 'latency_p90': 2.893639254570007}, {'batch_size': 2, 'throughput': 0.6060444513884212, 'latency_mean': 3.2889128303527833, 'latency_p50': 3.29646897315979, 'latency_p90': 3.601825165748596}, {'batch_size': 3, 'throughput': 0.7611179656867559, 'latency_mean': 3.918122501373291, 'latency_p50': 3.9465807676315308, 'latency_p90': 4.345513916015625}, {'batch_size': 4, 'throughput': 0.8895169046102677, 'latency_mean': 4.4795634806156155, 'latency_p50': 4.492785215377808, 'latency_p90': 5.072953295707703}, {'batch_size': 5, 'throughput': 0.9769346139297339, 'latency_mean': 5.090630036592484, 'latency_p50': 5.115854620933533, 'latency_p90': 5.762715315818786}]
max_input_tokens: 1024
max_output_tokens: 64
model_architecture: MistralForCausalLM
model_group: zonemercy/Vingt-Deux-v1-
model_name: temp-5
model_num_parameters: 22247282688.0
model_repo: zonemercy/Vingt-Deux-v1-1e5
model_size: 22B
num_battles: 2895
num_wins: 1406
ranking_group: single
status: torndown
submission_type: basic
throughput_3p7s: 0.72
timestamp: 2024-09-26T14:05:26+00:00
us_pacific_date: 2024-09-26
win_ratio: 0.4856649395509499
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 zonemercy-vingt-deux-v1-1e5-v26-mkmlizer
Waiting for job on zonemercy-vingt-deux-v1-1e5-v26-mkmlizer to finish
zonemercy-vingt-deux-v1-1e5-v26-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
zonemercy-vingt-deux-v1-1e5-v26-mkmlizer: ║ _____ __ __ ║
zonemercy-vingt-deux-v1-1e5-v26-mkmlizer: ║ / _/ /_ ___ __/ / ___ ___ / / ║
zonemercy-vingt-deux-v1-1e5-v26-mkmlizer: ║ / _/ / // / |/|/ / _ \/ -_) -_) / ║
zonemercy-vingt-deux-v1-1e5-v26-mkmlizer: ║ /_//_/\_, /|__,__/_//_/\__/\__/_/ ║
zonemercy-vingt-deux-v1-1e5-v26-mkmlizer: ║ /___/ ║
zonemercy-vingt-deux-v1-1e5-v26-mkmlizer: ║ ║
zonemercy-vingt-deux-v1-1e5-v26-mkmlizer: ║ Version: 0.11.12 ║
zonemercy-vingt-deux-v1-1e5-v26-mkmlizer: ║ Copyright 2023 MK ONE TECHNOLOGIES Inc. ║
zonemercy-vingt-deux-v1-1e5-v26-mkmlizer: ║ https://mk1.ai ║
zonemercy-vingt-deux-v1-1e5-v26-mkmlizer: ║ ║
zonemercy-vingt-deux-v1-1e5-v26-mkmlizer: ║ The license key for the current software has been verified as ║
zonemercy-vingt-deux-v1-1e5-v26-mkmlizer: ║ belonging to: ║
zonemercy-vingt-deux-v1-1e5-v26-mkmlizer: ║ ║
zonemercy-vingt-deux-v1-1e5-v26-mkmlizer: ║ Chai Research Corp. ║
zonemercy-vingt-deux-v1-1e5-v26-mkmlizer: ║ Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f ║
zonemercy-vingt-deux-v1-1e5-v26-mkmlizer: ║ Expiration: 2024-10-15 23:59:59 ║
zonemercy-vingt-deux-v1-1e5-v26-mkmlizer: ║ ║
zonemercy-vingt-deux-v1-1e5-v26-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
zonemercy-vingt-deux-v1-1e5-v26-mkmlizer: Downloaded to shared memory in 65.380s
zonemercy-vingt-deux-v1-1e5-v26-mkmlizer: quantizing model to /dev/shm/model_cache, profile:s0, folder:/tmp/tmpui2nxhvr, device:0
zonemercy-vingt-deux-v1-1e5-v26-mkmlizer: Saving flywheel model at /dev/shm/model_cache
zonemercy-vingt-deux-v1-1e5-v26-mkmlizer: quantized model in 48.052s
zonemercy-vingt-deux-v1-1e5-v26-mkmlizer: Processed model zonemercy/Vingt-Deux-v1-1e5 in 113.432s
zonemercy-vingt-deux-v1-1e5-v26-mkmlizer: creating bucket guanaco-mkml-models
zonemercy-vingt-deux-v1-1e5-v26-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
zonemercy-vingt-deux-v1-1e5-v26-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/zonemercy-vingt-deux-v1-1e5-v26
zonemercy-vingt-deux-v1-1e5-v26-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/zonemercy-vingt-deux-v1-1e5-v26/config.json
zonemercy-vingt-deux-v1-1e5-v26-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/zonemercy-vingt-deux-v1-1e5-v26/special_tokens_map.json
zonemercy-vingt-deux-v1-1e5-v26-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/zonemercy-vingt-deux-v1-1e5-v26/tokenizer_config.json
zonemercy-vingt-deux-v1-1e5-v26-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/zonemercy-vingt-deux-v1-1e5-v26/tokenizer.json
zonemercy-vingt-deux-v1-1e5-v26-mkmlizer: cp /dev/shm/model_cache/flywheel_model.1.safetensors s3://guanaco-mkml-models/zonemercy-vingt-deux-v1-1e5-v26/flywheel_model.1.safetensors
zonemercy-vingt-deux-v1-1e5-v26-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/zonemercy-vingt-deux-v1-1e5-v26/flywheel_model.0.safetensors
zonemercy-vingt-deux-v1-1e5-v26-mkmlizer: Loading 0: 0%| | 0/507 [00:00<?, ?it/s] Loading 0: 1%| | 5/507 [00:00<00:24, 20.31it/s] Loading 0: 2%|▏ | 10/507 [00:00<00:19, 25.82it/s] Loading 0: 3%|▎ | 14/507 [00:00<00:21, 22.85it/s] Loading 0: 4%|▎ | 19/507 [00:00<00:17, 28.25it/s] Loading 0: 5%|▍ | 23/507 [00:00<00:18, 25.79it/s] Loading 0: 6%|▌ | 28/507 [00:01<00:16, 29.60it/s] Loading 0: 6%|▋ | 32/507 [00:01<00:17, 26.45it/s] Loading 0: 7%|▋ | 37/507 [00:01<00:15, 29.65it/s] Loading 0: 8%|▊ | 41/507 [00:01<00:17, 26.26it/s] Loading 0: 9%|▉ | 46/507 [00:01<00:15, 29.93it/s] Loading 0: 10%|▉ | 50/507 [00:01<00:16, 27.52it/s] Loading 0: 10%|█ | 53/507 [00:02<00:24, 18.68it/s] Loading 0: 11%|█ | 56/507 [00:02<00:24, 18.28it/s] Loading 0: 12%|█▏ | 61/507 [00:02<00:19, 23.05it/s] Loading 0: 13%|█▎ | 65/507 [00:02<00:19, 22.89it/s] Loading 0: 14%|█▍ | 70/507 [00:02<00:15, 27.56it/s] Loading 0: 15%|█▍ | 75/507 [00:02<00:14, 30.75it/s] Loading 0: 16%|█▌ | 80/507 [00:03<00:13, 30.69it/s] Loading 0: 17%|█▋ | 85/507 [00:03<00:12, 33.41it/s] Loading 0: 18%|█▊ | 89/507 [00:03<00:14, 29.50it/s] Loading 0: 19%|█▊ | 94/507 [00:03<00:13, 31.32it/s] Loading 0: 19%|█▉ | 98/507 [00:03<00:14, 28.32it/s] Loading 0: 20%|██ | 103/507 [00:03<00:12, 31.76it/s] Loading 0: 21%|██ | 107/507 [00:03<00:13, 28.61it/s] Loading 0: 22%|██▏ | 112/507 [00:04<00:12, 32.26it/s] Loading 0: 23%|██▎ | 116/507 [00:04<00:18, 21.31it/s] Loading 0: 24%|██▍ | 122/507 [00:04<00:15, 25.02it/s] Loading 0: 25%|██▌ | 129/507 [00:04<00:11, 31.57it/s] Loading 0: 26%|██▌ | 133/507 [00:04<00:11, 31.45it/s] Loading 0: 27%|██▋ | 138/507 [00:04<00:11, 33.48it/s] Loading 0: 28%|██▊ | 142/507 [00:05<00:11, 32.57it/s] Loading 0: 29%|██▉ | 147/507 [00:05<00:10, 35.05it/s] Loading 0: 30%|██▉ | 151/507 [00:05<00:10, 32.75it/s] Loading 0: 31%|███ | 156/507 [00:05<00:10, 34.40it/s] Loading 0: 32%|███▏ | 160/507 [00:05<00:10, 33.13it/s] Loading 0: 32%|███▏ | 164/507 [00:05<00:10, 32.97it/s] Loading 0: 33%|███▎ | 168/507 [00:05<00:09, 34.32it/s] Loading 0: 34%|███▍ | 172/507 [00:06<00:15, 21.90it/s] Loading 0: 35%|███▍ | 176/507 [00:06<00:14, 22.58it/s] Loading 0: 36%|███▌ | 181/507 [00:06<00:11, 27.33it/s] Loading 0: 36%|███▋ | 185/507 [00:06<00:12, 26.43it/s] Loading 0: 37%|███▋ | 190/507 [00:06<00:10, 30.94it/s] Loading 0: 38%|███▊ | 194/507 [00:06<00:10, 28.89it/s] Loading 0: 39%|███▉ | 199/507 [00:07<00:09, 33.07it/s] Loading 0: 40%|████ | 203/507 [00:07<00:10, 30.27it/s] Loading 0: 41%|████ | 208/507 [00:07<00:08, 34.67it/s] Loading 0: 42%|████▏ | 212/507 [00:07<00:09, 31.38it/s] Loading 0: 43%|████▎ | 217/507 [00:07<00:08, 34.71it/s] Loading 0: 44%|████▎ | 221/507 [00:07<00:09, 31.41it/s] Loading 0: 44%|████▍ | 225/507 [00:08<00:12, 22.15it/s] Loading 0: 45%|████▌ | 230/507 [00:08<00:11, 24.27it/s] Loading 0: 46%|████▋ | 235/507 [00:08<00:09, 28.27it/s] Loading 0: 47%|████▋ | 239/507 [00:08<00:10, 26.53it/s] Loading 0: 48%|████▊ | 244/507 [00:08<00:08, 30.33it/s] Loading 0: 49%|████▉ | 248/507 [00:08<00:09, 27.95it/s] Loading 0: 50%|████▉ | 253/507 [00:08<00:07, 31.81it/s] Loading 0: 51%|█████ | 257/507 [00:09<00:08, 28.37it/s] Loading 0: 52%|█████▏ | 262/507 [00:09<00:07, 31.97it/s] Loading 0: 52%|█████▏ | 266/507 [00:09<00:08, 29.18it/s] Loading 0: 53%|█████▎ | 271/507 [00:09<00:07, 32.78it/s] Loading 0: 54%|█████▍ | 275/507 [00:09<00:07, 29.64it/s] Loading 0: 55%|█████▌ | 280/507 [00:09<00:06, 33.04it/s] Loading 0: 56%|█████▌ | 284/507 [00:10<00:09, 23.72it/s] Loading 0: 57%|█████▋ | 287/507 [00:10<00:09, 22.02it/s] Loading 0: 58%|█████▊ | 293/507 [00:10<00:08, 25.09it/s] Loading 0: 59%|█████▉ | 298/507 [00:10<00:07, 29.06it/s] Loading 0: 59%|█████▉ | 299/507 [00:25<00:07, 29.06it/s] Loading 0: 59%|█████▉ | 300/507 [00:25<04:03, 1.17s/it] Loading 0: 60%|█████▉ | 302/507 [00:25<03:20, 1.02it/s] Loading 0: 61%|██████ | 307/507 [00:25<02:01, 1.65it/s] Loading 0: 61%|██████ | 310/507 [00:25<01:32, 2.14it/s] Loading 0: 62%|██████▏ | 313/507 [00:26<01:08, 2.83it/s] Loading 0: 63%|██████▎ | 318/507 [00:26<00:42, 4.42it/s] Loading 0: 64%|██████▎ | 322/507 [00:26<00:30, 5.98it/s] Loading 0: 64%|██████▍ | 327/507 [00:26<00:21, 8.56it/s] Loading 0: 65%|██████▌ | 331/507 [00:26<00:16, 10.73it/s] Loading 0: 66%|██████▌ | 335/507 [00:26<00:13, 13.21it/s] Loading 0: 67%|██████▋ | 339/507 [00:26<00:10, 16.42it/s] Loading 0: 68%|██████▊ | 343/507 [00:27<00:11, 14.87it/s] Loading 0: 68%|██████▊ | 347/507 [00:27<00:09, 16.65it/s] Loading 0: 69%|██████▉ | 352/507 [00:27<00:07, 21.45it/s] Loading 0: 70%|███████ | 356/507 [00:27<00:06, 22.11it/s] Loading 0: 71%|███████ | 361/507 [00:27<00:05, 26.84it/s] Loading 0: 72%|███████▏ | 365/507 [00:27<00:05, 26.07it/s] Loading 0: 73%|███████▎ | 370/507 [00:27<00:04, 30.47it/s] Loading 0: 74%|███████▍ | 374/507 [00:28<00:04, 28.66it/s] Loading 0: 75%|███████▌ | 381/507 [00:28<00:03, 35.49it/s] Loading 0: 76%|███████▌ | 385/507 [00:28<00:03, 33.79it/s] Loading 0: 77%|███████▋ | 389/507 [00:28<00:03, 33.11it/s] Loading 0: 78%|███████▊ | 393/507 [00:28<00:03, 32.04it/s] Loading 0: 78%|███████▊ | 397/507 [00:28<00:04, 22.79it/s] Loading 0: 79%|███████▉ | 401/507 [00:29<00:04, 23.08it/s] Loading 0: 80%|████████ | 406/507 [00:29<00:03, 27.74it/s] Loading 0: 81%|████████ | 410/507 [00:29<00:03, 26.62it/s] Loading 0: 82%|████████▏ | 415/507 [00:29<00:02, 30.95it/s] Loading 0: 83%|████████▎ | 419/507 [00:29<00:03, 28.89it/s] Loading 0: 84%|████████▎ | 424/507 [00:29<00:02, 33.31it/s] Loading 0: 84%|████████▍ | 428/507 [00:29<00:02, 30.86it/s] Loading 0: 85%|████████▌ | 433/507 [00:29<00:02, 34.80it/s] Loading 0: 86%|████████▌ | 437/507 [00:30<00:02, 31.61it/s] Loading 0: 88%|████████▊ | 444/507 [00:30<00:01, 38.27it/s] Loading 0: 89%|████████▊ | 449/507 [00:30<00:01, 37.45it/s] Loading 0: 90%|████████▉ | 454/507 [00:30<00:01, 34.13it/s] Loading 0: 90%|█████████ | 458/507 [00:32<00:08, 6.06it/s] Loading 0: 91%|█████████ | 461/507 [00:32<00:06, 7.20it/s] Loading 0: 92%|█████████▏| 465/507 [00:33<00:04, 8.75it/s] Loading 0: 93%|█████████▎| 470/507 [00:33<00:03, 12.05it/s] Loading 0: 93%|█████████▎| 474/507 [00:33<00:02, 13.94it/s] Loading 0: 94%|█████████▍| 479/507 [00:33<00:01, 17.98it/s] Loading 0: 95%|█████████▌| 483/507 [00:33<00:01, 19.40it/s] Loading 0: 96%|█████████▋| 488/507 [00:33<00:00, 23.75it/s] Loading 0: 97%|█████████▋| 492/507 [00:34<00:00, 22.57it/s] Loading 0: 98%|█████████▊| 497/507 [00:34<00:00, 26.71it/s] Loading 0: 99%|█████████▉| 501/507 [00:34<00:00, 25.25it/s] Loading 0: 100%|█████████▉| 506/507 [00:34<00:00, 29.39it/s]
Job zonemercy-vingt-deux-v1-1e5-v26-mkmlizer completed after 153.63s with status: succeeded
Stopping job with name zonemercy-vingt-deux-v1-1e5-v26-mkmlizer
Pipeline stage MKMLizer completed in 154.50s
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 zonemercy-vingt-deux-v1-1e5-v26
Waiting for inference service zonemercy-vingt-deux-v1-1e5-v26 to be ready
Inference service zonemercy-vingt-deux-v1-1e5-v26 ready after 220.5184895992279s
Pipeline stage MKMLDeployer completed in 220.88s
run pipeline stage %s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.80950665473938s
Received healthy response to inference request in 4.425158739089966s
Received healthy response to inference request in 2.407651901245117s
Received healthy response to inference request in 2.2951037883758545s
Received healthy response to inference request in 2.272883415222168s
5 requests
0 failed requests
5th percentile: 2.2773274898529055
10th percentile: 2.2817715644836425
20th percentile: 2.290659713745117
30th percentile: 2.317613410949707
40th percentile: 2.362632656097412
50th percentile: 2.407651901245117
60th percentile: 2.5683938026428224
70th percentile: 2.729135704040527
80th percentile: 3.1326370716094973
90th percentile: 3.778897905349732
95th percentile: 4.102028322219848
99th percentile: 4.360532655715942
mean time: 2.842060899734497
Pipeline stage StressChecker completed in 14.89s
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 7.91s
Shutdown handler de-registered
zonemercy-vingt-deux-v1-1e5_v26 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 0.11s
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 zonemercy-vingt-deux-v1-1e5-v26-profiler
Waiting for inference service zonemercy-vingt-deux-v1-1e5-v26-profiler to be ready
Inference service zonemercy-vingt-deux-v1-1e5-v26-profiler ready after 230.52907013893127s
Pipeline stage MKMLProfilerDeployer completed in 230.90s
run pipeline stage %s
Running pipeline stage MKMLProfilerRunner
kubectl cp /code/guanaco/guanaco_inference_services/src/inference_scripts tenant-chaiml-guanaco/zonemercy-vingt-deux20e41579c2ea6012c5cdde221fa4f2ef-deplodr4gs:/code/chaiverse_profiler_1727360208 --namespace tenant-chaiml-guanaco
kubectl exec -it zonemercy-vingt-deux20e41579c2ea6012c5cdde221fa4f2ef-deplodr4gs --namespace tenant-chaiml-guanaco -- sh -c 'cd /code/chaiverse_profiler_1727360208 && 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 1024 --output_tokens 64 --summary /code/chaiverse_profiler_1727360208/summary.json'
kubectl exec -it zonemercy-vingt-deux20e41579c2ea6012c5cdde221fa4f2ef-deplodr4gs --namespace tenant-chaiml-guanaco -- bash -c 'cat /code/chaiverse_profiler_1727360208/summary.json'
Pipeline stage MKMLProfilerRunner completed in 1552.33s
run pipeline stage %s
Running pipeline stage MKMLProfilerDeleter
Checking if service zonemercy-vingt-deux-v1-1e5-v26-profiler is running
Tearing down inference service zonemercy-vingt-deux-v1-1e5-v26-profiler
Service zonemercy-vingt-deux-v1-1e5-v26-profiler has been torndown
Pipeline stage MKMLProfilerDeleter completed in 2.22s
Shutdown handler de-registered
zonemercy-vingt-deux-v1-1e5_v26 status is now inactive due to auto deactivation removed underperforming models
Deleting key zonemercy-vingt-deux-v2-1e5-v26/tokenizer_config.json from bucket guanaco-mkml-models
Deleting key zonemercy-vingt-deux-v2-1e5-v27/tokenizer.json from bucket guanaco-mkml-models
zonemercy-vingt-deux-v1-1e5_v26 status is now torndown due to DeploymentManager action
Shutdown handler de-registered
Pipeline stage MKMLModelDeleter completed in 96.77s