rica40325-feedback-7000steps

submission_id: rica40325-feedback-7000steps_v1

developer_uid: rica40325

best_of: 16

celo_rating: 1247.87

display_name: rica40325-feedback-7000steps_v1

family_friendly_score: 0.0

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}

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': 512, 'best_of': 16, 'max_output_tokens': 64}

gpu_counts: {'NVIDIA RTX A5000': 1}

is_internal_developer: False

language_model: rica40325/feedback_7000steps

latencies: [{'batch_size': 1, 'throughput': 0.9144185440877848, 'latency_mean': 1.0934960639476776, 'latency_p50': 1.0970920324325562, 'latency_p90': 1.2310615539550782}, {'batch_size': 4, 'throughput': 1.810782739633487, 'latency_mean': 2.192737656831741, 'latency_p50': 2.1940892934799194, 'latency_p90': 2.4502484798431396}, {'batch_size': 5, 'throughput': 1.8845378592266568, 'latency_mean': 2.636907504796982, 'latency_p50': 2.6401166915893555, 'latency_p90': 2.909585475921631}, {'batch_size': 8, 'throughput': 2.0309053893775464, 'latency_mean': 3.906652754545212, 'latency_p50': 3.949413299560547, 'latency_p90': 4.381656908988953}, {'batch_size': 10, 'throughput': 2.078666692526697, 'latency_mean': 4.757255580425262, 'latency_p50': 4.757577061653137, 'latency_p90': 5.488437795639038}, {'batch_size': 12, 'throughput': 2.0661211485775373, 'latency_mean': 5.73168695807457, 'latency_p50': 5.767443299293518, 'latency_p90': 6.569678497314453}, {'batch_size': 15, 'throughput': 2.0572280843114568, 'latency_mean': 7.14754838347435, 'latency_p50': 7.215520262718201, 'latency_p90': 7.913970041275024}]

max_input_tokens: 512

max_output_tokens: 64

model_architecture: LlamaForCausalLM

model_group: rica40325/feedback_7000s

model_name: rica40325-feedback-7000steps_v1

model_num_parameters: 8030261248.0

model_repo: rica40325/feedback_7000steps

model_size: 8B

num_battles: 11020

num_wins: 5567

ranking_group: single

status: torndown

submission_type: basic

throughput_3p7s: 2.03

timestamp: 2024-09-09T11:55:10+00:00

us_pacific_date: 2024-09-09

win_ratio: 0.5051724137931034

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 rica40325-feedback-7000steps-v1-mkmlizer
Waiting for job on rica40325-feedback-7000steps-v1-mkmlizer to finish
rica40325-feedback-7000steps-v1-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
rica40325-feedback-7000steps-v1-mkmlizer: ║     _____            __           __                                ║
rica40325-feedback-7000steps-v1-mkmlizer: ║    / _/ /_ ___    __/ /  ___ ___ / /                                ║
rica40325-feedback-7000steps-v1-mkmlizer: ║   / _/ / // / |/|/ / _ \/ -_) -_) /                                 ║
rica40325-feedback-7000steps-v1-mkmlizer: ║  /_//_/\_, /|__,__/_//_/\__/\__/_/                                  ║
rica40325-feedback-7000steps-v1-mkmlizer: ║       /___/                                                         ║
rica40325-feedback-7000steps-v1-mkmlizer: ║                                                                     ║
rica40325-feedback-7000steps-v1-mkmlizer: ║  Version: 0.10.1                                                    ║
rica40325-feedback-7000steps-v1-mkmlizer: ║  Copyright 2023 MK ONE TECHNOLOGIES Inc.                            ║
rica40325-feedback-7000steps-v1-mkmlizer: ║  https://mk1.ai                                                     ║
rica40325-feedback-7000steps-v1-mkmlizer: ║                                                                     ║
rica40325-feedback-7000steps-v1-mkmlizer: ║  The license key for the current software has been verified as      ║
rica40325-feedback-7000steps-v1-mkmlizer: ║  belonging to:                                                      ║
rica40325-feedback-7000steps-v1-mkmlizer: ║                                                                     ║
rica40325-feedback-7000steps-v1-mkmlizer: ║  Chai Research Corp.                                                ║
rica40325-feedback-7000steps-v1-mkmlizer: ║  Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f                   ║
rica40325-feedback-7000steps-v1-mkmlizer: ║  Expiration: 2024-10-15 23:59:59                                    ║
rica40325-feedback-7000steps-v1-mkmlizer: ║                                                                     ║
rica40325-feedback-7000steps-v1-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
rica40325-feedback-7000steps-v1-mkmlizer: Downloaded to shared memory in 60.863s
rica40325-feedback-7000steps-v1-mkmlizer: quantizing model to /dev/shm/model_cache, profile:s0, folder:/tmp/tmpl8yulmj4, device:0
rica40325-feedback-7000steps-v1-mkmlizer: Saving flywheel model at /dev/shm/model_cache
rica40325-feedback-7000steps-v1-mkmlizer: quantized model in 29.194s
rica40325-feedback-7000steps-v1-mkmlizer: Processed model rica40325/feedback_7000steps in 90.057s
rica40325-feedback-7000steps-v1-mkmlizer: creating bucket guanaco-mkml-models
rica40325-feedback-7000steps-v1-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
rica40325-feedback-7000steps-v1-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/rica40325-feedback-7000steps-v1
rica40325-feedback-7000steps-v1-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/rica40325-feedback-7000steps-v1/special_tokens_map.json
rica40325-feedback-7000steps-v1-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/rica40325-feedback-7000steps-v1/config.json
rica40325-feedback-7000steps-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/rica40325-feedback-7000steps-v1/tokenizer_config.json
rica40325-feedback-7000steps-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/rica40325-feedback-7000steps-v1/tokenizer.json
rica40325-feedback-7000steps-v1-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/rica40325-feedback-7000steps-v1/flywheel_model.0.safetensors
rica40325-feedback-7000steps-v1-mkmlizer: 
Loading 0:   0%|          | 0/291 [00:00<?, ?it/s]
Loading 0:   2%|▏         | 5/291 [00:00<00:12, 23.82it/s]
Loading 0:   4%|▍         | 12/291 [00:00<00:07, 37.92it/s]
Loading 0:   6%|▌         | 17/291 [00:00<00:07, 35.52it/s]
Loading 0:   7%|▋         | 21/291 [00:00<00:07, 35.51it/s]
Loading 0:   9%|▊         | 25/291 [00:00<00:07, 33.97it/s]
Loading 0:  11%|█         | 31/291 [00:00<00:06, 39.96it/s]
Loading 0:  12%|█▏        | 36/291 [00:01<00:11, 22.78it/s]
Loading 0:  14%|█▍        | 41/291 [00:01<00:10, 24.25it/s]
Loading 0:  16%|█▋        | 48/291 [00:01<00:07, 30.76it/s]
Loading 0:  18%|█▊        | 52/291 [00:01<00:07, 30.85it/s]
Loading 0:  20%|█▉        | 57/291 [00:01<00:06, 34.79it/s]
Loading 0:  21%|██        | 61/291 [00:01<00:07, 32.36it/s]
Loading 0:  23%|██▎       | 66/291 [00:02<00:06, 35.12it/s]
Loading 0:  24%|██▍       | 70/291 [00:02<00:06, 34.00it/s]
Loading 0:  25%|██▌       | 74/291 [00:02<00:06, 34.98it/s]
Loading 0:  27%|██▋       | 78/291 [00:02<00:06, 35.46it/s]
Loading 0:  28%|██▊       | 82/291 [00:02<00:08, 25.70it/s]
Loading 0:  30%|██▉       | 86/291 [00:02<00:07, 28.25it/s]
Loading 0:  31%|███▏      | 91/291 [00:02<00:06, 31.00it/s]
Loading 0:  33%|███▎      | 95/291 [00:03<00:05, 32.94it/s]
Loading 0:  34%|███▍      | 99/291 [00:03<00:05, 34.14it/s]
Loading 0:  35%|███▌      | 103/291 [00:03<00:05, 32.63it/s]
Loading 0:  37%|███▋      | 108/291 [00:03<00:05, 36.05it/s]
Loading 0:  38%|███▊      | 112/291 [00:03<00:05, 32.65it/s]
Loading 0:  40%|███▉      | 116/291 [00:03<00:05, 32.79it/s]
Loading 0:  42%|████▏     | 121/291 [00:03<00:04, 37.10it/s]
Loading 0:  43%|████▎     | 126/291 [00:03<00:04, 39.47it/s]
Loading 0:  45%|████▌     | 131/291 [00:03<00:03, 40.31it/s]
Loading 0:  47%|████▋     | 136/291 [00:04<00:06, 25.13it/s]
Loading 0:  48%|████▊     | 140/291 [00:04<00:06, 24.52it/s]
Loading 0:  51%|█████     | 147/291 [00:04<00:04, 31.32it/s]
Loading 0:  52%|█████▏    | 151/291 [00:04<00:04, 29.88it/s]
Loading 0:  54%|█████▎    | 156/291 [00:04<00:04, 32.30it/s]
Loading 0:  55%|█████▍    | 160/291 [00:05<00:04, 31.73it/s]
Loading 0:  57%|█████▋    | 165/291 [00:05<00:03, 34.41it/s]
Loading 0:  58%|█████▊    | 169/291 [00:05<00:03, 32.12it/s]
Loading 0:  60%|█████▉    | 174/291 [00:05<00:03, 33.92it/s]
Loading 0:  61%|██████    | 178/291 [00:05<00:03, 31.84it/s]
Loading 0:  63%|██████▎   | 183/291 [00:05<00:03, 34.97it/s]
Loading 0:  64%|██████▍   | 187/291 [00:05<00:04, 24.77it/s]
Loading 0:  65%|██████▌   | 190/291 [00:06<00:04, 23.90it/s]
Loading 0:  67%|██████▋   | 194/291 [00:06<00:04, 23.89it/s]
Loading 0:  69%|██████▉   | 201/291 [00:06<00:02, 31.19it/s]
Loading 0:  70%|███████   | 205/291 [00:06<00:02, 29.47it/s]
Loading 0:  72%|███████▏  | 210/291 [00:06<00:02, 31.76it/s]
Loading 0:  74%|███████▎  | 214/291 [00:06<00:02, 31.13it/s]
Loading 0:  75%|███████▌  | 219/291 [00:06<00:02, 33.11it/s]
Loading 0:  77%|███████▋  | 223/291 [00:07<00:02, 31.68it/s]
Loading 0:  78%|███████▊  | 227/291 [00:07<00:01, 32.71it/s]
Loading 0:  79%|███████▉  | 231/291 [00:07<00:01, 33.20it/s]
Loading 0:  81%|████████  | 235/291 [00:07<00:02, 25.68it/s]
Loading 0:  82%|████████▏ | 239/291 [00:07<00:02, 24.73it/s]
Loading 0:  85%|████████▍ | 246/291 [00:07<00:01, 31.99it/s]
Loading 0:  86%|████████▌ | 250/291 [00:08<00:01, 30.38it/s]
Loading 0:  88%|████████▊ | 255/291 [00:08<00:01, 32.42it/s]
Loading 0:  89%|████████▉ | 259/291 [00:08<00:01, 31.26it/s]
Loading 0:  91%|█████████ | 264/291 [00:08<00:00, 33.49it/s]
Loading 0:  92%|█████████▏| 268/291 [00:08<00:00, 32.80it/s]
Loading 0:  94%|█████████▍| 273/291 [00:08<00:00, 34.59it/s]
Loading 0:  95%|█████████▌| 277/291 [00:08<00:00, 32.66it/s]
Loading 0:  97%|█████████▋| 281/291 [00:08<00:00, 32.60it/s]
Loading 0:  98%|█████████▊| 286/291 [00:14<00:01,  2.60it/s]
Loading 0:  99%|█████████▉| 289/291 [00:14<00:00,  3.24it/s]
                                                            
Job rica40325-feedback-7000steps-v1-mkmlizer completed after 115.92s with status: succeeded
Stopping job with name rica40325-feedback-7000steps-v1-mkmlizer
Pipeline stage MKMLizer completed in 117.43s
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 rica40325-feedback-7000steps-v1
Waiting for inference service rica40325-feedback-7000steps-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
Connection pool is full, discarding connection: %s. Connection pool size: %s
Inference service rica40325-feedback-7000steps-v1 ready after 151.24953818321228s
Pipeline stage MKMLDeployer completed in 151.59s
run pipeline stage %s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.3970870971679688s
Received healthy response to inference request in 3.3521833419799805s
Received healthy response to inference request in 1.6512925624847412s
Received healthy response to inference request in 1.816476583480835s
Connection pool is full, discarding connection: %s. Connection pool size: %s
Received healthy response to inference request in 2.541536569595337s
5 requests
0 failed requests
5th percentile: 1.68432936668396
10th percentile: 1.7173661708831787
20th percentile: 1.7834397792816161
30th percentile: 1.9325986862182618
40th percentile: 2.164842891693115
50th percentile: 2.3970870971679688
60th percentile: 2.454866886138916
70th percentile: 2.5126466751098633
80th percentile: 2.703665924072266
90th percentile: 3.027924633026123
95th percentile: 3.1900539875030516
99th percentile: 3.3197574710845945
mean time: 2.3517152309417724
Pipeline stage StressChecker completed in 12.60s
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 5.05s
Shutdown handler de-registered
rica40325-feedback-7000steps_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 0.11s
run pipeline stage %s
Running pipeline stage MKMLProfilerTemplater
Pipeline stage MKMLProfilerTemplater completed in 0.10s
run pipeline stage %s
Running pipeline stage MKMLProfilerDeployer
Creating inference service rica40325-feedback-7000steps-v1-profiler
Waiting for inference service rica40325-feedback-7000steps-v1-profiler to be ready
Inference service rica40325-feedback-7000steps-v1-profiler ready after 150.36715698242188s
Pipeline stage MKMLProfilerDeployer completed in 150.76s
run pipeline stage %s
Running pipeline stage MKMLProfilerRunner
kubectl cp /code/guanaco/guanaco_inference_services/src/inference_scripts tenant-chaiml-guanaco/rica40325-feedback-791bf0ec141b57a31ce98fce3583adca1-deplo72qpz:/code/chaiverse_profiler_1725883400 --namespace tenant-chaiml-guanaco
kubectl exec -it rica40325-feedback-791bf0ec141b57a31ce98fce3583adca1-deplo72qpz --namespace tenant-chaiml-guanaco -- sh -c 'cd /code/chaiverse_profiler_1725883400 && python profiles.py profile --best_of_n 16 --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_1725883400/summary.json'
kubectl exec -it rica40325-feedback-791bf0ec141b57a31ce98fce3583adca1-deplo72qpz --namespace tenant-chaiml-guanaco -- bash -c 'cat /code/chaiverse_profiler_1725883400/summary.json'
Pipeline stage MKMLProfilerRunner completed in 829.28s
run pipeline stage %s
Running pipeline stage MKMLProfilerDeleter
Checking if service rica40325-feedback-7000steps-v1-profiler is running
Tearing down inference service rica40325-feedback-7000steps-v1-profiler
Service rica40325-feedback-7000steps-v1-profiler has been torndown
Pipeline stage MKMLProfilerDeleter completed in 1.71s
Shutdown handler de-registered
rica40325-feedback-7000steps_v1 status is now inactive due to auto deactivation removed underperforming models
rica40325-feedback-7000steps_v1 status is now torndown due to DeploymentManager action