developer_uid: rinen0721
submission_id: rinen0721-llama0918_v1
model_name: rinen0721-llama0918_v1
model_group: rinen0721/llama0918
status: torndown
timestamp: 2024-09-18T02:48:31+00:00
num_battles: 10053
num_wins: 4479
celo_rating: 1214.16
family_friendly_score: 0.0
submission_type: basic
model_repo: rinen0721/llama0918
model_architecture: LlamaForCausalLM
model_num_parameters: 8030261248.0
best_of: 8
max_input_tokens: 1024
max_output_tokens: 64
latencies: [{'batch_size': 1, 'throughput': 0.8625134520574232, 'latency_mean': 1.1593343102931977, 'latency_p50': 1.1643022298812866, 'latency_p90': 1.2758588075637818}, {'batch_size': 4, 'throughput': 1.8682860649384445, 'latency_mean': 2.1317837011814116, 'latency_p50': 2.12291944026947, 'latency_p90': 2.3829901218414307}, {'batch_size': 5, 'throughput': 1.9941875974505936, 'latency_mean': 2.492738536596298, 'latency_p50': 2.49042010307312, 'latency_p90': 2.770179510116577}, {'batch_size': 8, 'throughput': 2.1961450556955087, 'latency_mean': 3.6165302288532257, 'latency_p50': 3.623639941215515, 'latency_p90': 4.003772640228271}, {'batch_size': 10, 'throughput': 2.267278420970496, 'latency_mean': 4.373199400901794, 'latency_p50': 4.397008776664734, 'latency_p90': 4.9007219791412355}, {'batch_size': 12, 'throughput': 2.329461192108001, 'latency_mean': 5.105537408590317, 'latency_p50': 5.100316047668457, 'latency_p90': 5.8313768148422245}, {'batch_size': 15, 'throughput': 2.312097011998787, 'latency_mean': 6.3958341038227084, 'latency_p50': 6.417673945426941, 'latency_p90': 7.0848267555236815}]
gpu_counts: {'NVIDIA RTX A5000': 1}
display_name: rinen0721-llama0918_v1
is_internal_developer: False
language_model: rinen0721/llama0918
model_size: 8B
ranking_group: single
throughput_3p7s: 2.22
us_pacific_date: 2024-09-17
win_ratio: 0.44553864518054315
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 rinen0721-llama0918-v1-mkmlizer
Waiting for job on rinen0721-llama0918-v1-mkmlizer to finish
rinen0721-llama0918-v1-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
rinen0721-llama0918-v1-mkmlizer: ║     _____            __           __                                ║
rinen0721-llama0918-v1-mkmlizer: ║    / _/ /_ ___    __/ /  ___ ___ / /                                ║
rinen0721-llama0918-v1-mkmlizer: ║   / _/ / // / |/|/ / _ \/ -_) -_) /                                 ║
rinen0721-llama0918-v1-mkmlizer: ║  /_//_/\_, /|__,__/_//_/\__/\__/_/                                  ║
rinen0721-llama0918-v1-mkmlizer: ║       /___/                                                         ║
rinen0721-llama0918-v1-mkmlizer: ║                                                                     ║
rinen0721-llama0918-v1-mkmlizer: ║  Version: 0.10.1                                                    ║
rinen0721-llama0918-v1-mkmlizer: ║  Copyright 2023 MK ONE TECHNOLOGIES Inc.                            ║
rinen0721-llama0918-v1-mkmlizer: ║  https://mk1.ai                                                     ║
rinen0721-llama0918-v1-mkmlizer: ║                                                                     ║
rinen0721-llama0918-v1-mkmlizer: ║  The license key for the current software has been verified as      ║
rinen0721-llama0918-v1-mkmlizer: ║  belonging to:                                                      ║
rinen0721-llama0918-v1-mkmlizer: ║                                                                     ║
rinen0721-llama0918-v1-mkmlizer: ║  Chai Research Corp.                                                ║
rinen0721-llama0918-v1-mkmlizer: ║  Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f                   ║
rinen0721-llama0918-v1-mkmlizer: ║  Expiration: 2024-10-15 23:59:59                                    ║
rinen0721-llama0918-v1-mkmlizer: ║                                                                     ║
rinen0721-llama0918-v1-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
rinen0721-llama0918-v1-mkmlizer: Downloaded to shared memory in 38.205s
rinen0721-llama0918-v1-mkmlizer: quantizing model to /dev/shm/model_cache, profile:s0, folder:/tmp/tmpsred8ovd, device:0
rinen0721-llama0918-v1-mkmlizer: Saving flywheel model at /dev/shm/model_cache
rinen0721-llama0918-v1-mkmlizer: quantized model in 25.641s
rinen0721-llama0918-v1-mkmlizer: Processed model rinen0721/llama0918 in 63.846s
rinen0721-llama0918-v1-mkmlizer: creating bucket guanaco-mkml-models
rinen0721-llama0918-v1-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
rinen0721-llama0918-v1-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/rinen0721-llama0918-v1
rinen0721-llama0918-v1-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/rinen0721-llama0918-v1/special_tokens_map.json
rinen0721-llama0918-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/rinen0721-llama0918-v1/tokenizer_config.json
rinen0721-llama0918-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/rinen0721-llama0918-v1/tokenizer.json
rinen0721-llama0918-v1-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/rinen0721-llama0918-v1/flywheel_model.0.safetensors
rinen0721-llama0918-v1-mkmlizer: 
Loading 0:   0%|          | 0/291 [00:00<?, ?it/s]
Loading 0:   2%|▏         | 7/291 [00:00<00:05, 55.91it/s]
Loading 0:   8%|▊         | 22/291 [00:00<00:02, 93.69it/s]
Loading 0:  11%|█         | 32/291 [00:00<00:02, 95.98it/s]
Loading 0:  14%|█▍        | 42/291 [00:00<00:02, 93.30it/s]
Loading 0:  18%|█▊        | 52/291 [00:00<00:02, 84.45it/s]
Loading 0:  23%|██▎       | 67/291 [00:00<00:02, 94.50it/s]
Loading 0:  27%|██▋       | 79/291 [00:00<00:02, 91.49it/s]
Loading 0:  31%|███       | 89/291 [00:02<00:08, 25.02it/s]
Loading 0:  35%|███▌      | 103/291 [00:02<00:05, 34.37it/s]
Loading 0:  39%|███▉      | 114/291 [00:02<00:04, 42.70it/s]
Loading 0:  42%|████▏     | 123/291 [00:02<00:03, 48.94it/s]
Loading 0:  46%|████▌     | 133/291 [00:02<00:02, 54.13it/s]
Loading 0:  51%|█████     | 148/291 [00:02<00:02, 66.25it/s]
Loading 0:  54%|█████▍    | 158/291 [00:02<00:01, 72.61it/s]
Loading 0:  58%|█████▊    | 169/291 [00:02<00:01, 72.44it/s]
Loading 0:  63%|██████▎   | 184/291 [00:03<00:01, 82.21it/s]
Loading 0:  67%|██████▋   | 194/291 [00:04<00:03, 25.72it/s]
Loading 0:  70%|███████   | 205/291 [00:04<00:02, 31.96it/s]
Loading 0:  76%|███████▌  | 220/291 [00:04<00:01, 43.11it/s]
Loading 0:  79%|███████▉  | 231/291 [00:04<00:01, 51.49it/s]
Loading 0:  83%|████████▎ | 241/291 [00:04<00:00, 54.52it/s]
Loading 0:  88%|████████▊ | 256/291 [00:04<00:00, 67.04it/s]
Loading 0:  92%|█████████▏| 268/291 [00:04<00:00, 72.02it/s]
Loading 0:  97%|█████████▋| 283/291 [00:05<00:00, 80.74it/s]
                                                            

Job rinen0721-llama0918-v1-mkmlizer completed after 82.9s with status: succeeded
Stopping job with name rinen0721-llama0918-v1-mkmlizer
Pipeline stage MKMLizer completed in 83.67s
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 rinen0721-llama0918-v1
Waiting for inference service rinen0721-llama0918-v1 to be ready
Inference service rinen0721-llama0918-v1 ready after 250.5518457889557s
Pipeline stage MKMLDeployer completed in 250.90s
run pipeline stage %s
Running pipeline stage StressChecker
Received healthy response to inference request in 1.8570480346679688s
Received healthy response to inference request in 1.5883910655975342s
Received healthy response to inference request in 2.018411874771118s
Received healthy response to inference request in 1.2524340152740479s
Received healthy response to inference request in 1.4622538089752197s
5 requests
0 failed requests
5th percentile: 1.2943979740142821
10th percentile: 1.3363619327545166
20th percentile: 1.4202898502349854
30th percentile: 1.4874812602996825
40th percentile: 1.5379361629486084
50th percentile: 1.5883910655975342
60th percentile: 1.695853853225708
70th percentile: 1.8033166408538819
80th percentile: 1.8893208026885986
90th percentile: 1.9538663387298585
95th percentile: 1.9861391067504883
99th percentile: 2.0119573211669923
mean time: 1.6357077598571776
Pipeline stage StressChecker completed in 8.77s
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 7.95s
Shutdown handler de-registered
rinen0721-llama0918_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 0.11s
run pipeline stage %s
Running pipeline stage MKMLProfilerDeployer
Creating inference service rinen0721-llama0918-v1-profiler
Waiting for inference service rinen0721-llama0918-v1-profiler to be ready
Inference service rinen0721-llama0918-v1-profiler ready after 180.54304265975952s
Pipeline stage MKMLProfilerDeployer completed in 180.94s
run pipeline stage %s
Running pipeline stage MKMLProfilerRunner
kubectl cp /code/guanaco/guanaco_inference_services/src/inference_scripts tenant-chaiml-guanaco/rinen0721-llama0918-v1-profiler-predictor-00001-deploymentmtfz7:/code/chaiverse_profiler_1726628285 --namespace tenant-chaiml-guanaco
kubectl exec -it rinen0721-llama0918-v1-profiler-predictor-00001-deploymentmtfz7 --namespace tenant-chaiml-guanaco -- sh -c 'cd /code/chaiverse_profiler_1726628285 && 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_1726628285/summary.json'
kubectl exec -it rinen0721-llama0918-v1-profiler-predictor-00001-deploymentmtfz7 --namespace tenant-chaiml-guanaco -- bash -c 'cat /code/chaiverse_profiler_1726628285/summary.json'
Pipeline stage MKMLProfilerRunner completed in 797.85s
run pipeline stage %s
Running pipeline stage MKMLProfilerDeleter
Checking if service rinen0721-llama0918-v1-profiler is running
Tearing down inference service rinen0721-llama0918-v1-profiler
Service rinen0721-llama0918-v1-profiler has been torndown
Pipeline stage MKMLProfilerDeleter completed in 1.90s
Shutdown handler de-registered
rinen0721-llama0918_v1 status is now inactive due to auto deactivation removed underperforming models
rinen0721-llama0918_v1 status is now torndown due to DeploymentManager action