jic062-dpo-v1-1-nemo

developer_uid: chace9580

submission_id: jic062-dpo-v1-1-nemo_v1

model_name: jic062-dpo-v1-1-nemo_v1

model_group: jic062/dpo-v1.1-Nemo

status: torndown

timestamp: 2024-09-10T04:09:51+00:00

num_battles: 10608

num_wins: 5413

celo_rating: 1243.75

family_friendly_score: 0.0

submission_type: basic

model_repo: jic062/dpo-v1.1-Nemo

model_architecture: MistralForCausalLM

model_num_parameters: 12772070400.0

best_of: 8

max_input_tokens: 512

max_output_tokens: 64

latencies: [{'batch_size': 1, 'throughput': 0.6932580399259243, 'latency_mean': 1.442374062538147, 'latency_p50': 1.4360888004302979, 'latency_p90': 1.6140645742416382}, {'batch_size': 3, 'throughput': 1.316740846900233, 'latency_mean': 2.2746323883533477, 'latency_p50': 2.3003424406051636, 'latency_p90': 2.516007423400879}, {'batch_size': 5, 'throughput': 1.5661280158644961, 'latency_mean': 3.180082701444626, 'latency_p50': 3.1782368421554565, 'latency_p90': 3.5487974882125854}, {'batch_size': 6, 'throughput': 1.6060576102229025, 'latency_mean': 3.716907887458801, 'latency_p50': 3.6666758060455322, 'latency_p90': 4.158643579483032}, {'batch_size': 8, 'throughput': 1.610182544688867, 'latency_mean': 4.946519954204559, 'latency_p50': 4.94711172580719, 'latency_p90': 5.635858702659607}, {'batch_size': 10, 'throughput': 1.5279601866624737, 'latency_mean': 6.4864281344413754, 'latency_p50': 6.439584136009216, 'latency_p90': 7.454421544075012}]

gpu_counts: {'NVIDIA RTX A5000': 1}

display_name: jic062-dpo-v1-1-nemo_v1

is_internal_developer: False

language_model: jic062/dpo-v1.1-Nemo

model_size: 13B

ranking_group: single

throughput_3p7s: 1.61

us_pacific_date: 2024-09-09

win_ratio: 0.5102752639517345

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', '[/INST]'], 'max_input_tokens': 512, 'best_of': 8, 'max_output_tokens': 64}

formatter: {'memory_template': '[INST]system\n{memory}[/INST]\n', 'prompt_template': '[INST]user\n{prompt}[/INST]\n', 'bot_template': '[INST]assistant\n{bot_name}: {message}[/INST]\n', 'user_template': '[INST]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 jic062-dpo-v1-1-nemo-v1-mkmlizer
Waiting for job on jic062-dpo-v1-1-nemo-v1-mkmlizer to finish
jic062-dpo-v1-1-nemo-v1-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
jic062-dpo-v1-1-nemo-v1-mkmlizer: ║     _____            __           __                                ║
jic062-dpo-v1-1-nemo-v1-mkmlizer: ║    / _/ /_ ___    __/ /  ___ ___ / /                                ║
jic062-dpo-v1-1-nemo-v1-mkmlizer: ║   / _/ / // / |/|/ / _ \/ -_) -_) /                                 ║
jic062-dpo-v1-1-nemo-v1-mkmlizer: ║  /_//_/\_, /|__,__/_//_/\__/\__/_/                                  ║
jic062-dpo-v1-1-nemo-v1-mkmlizer: ║       /___/                                                         ║
jic062-dpo-v1-1-nemo-v1-mkmlizer: ║                                                                     ║
jic062-dpo-v1-1-nemo-v1-mkmlizer: ║  Version: 0.10.1                                                    ║
jic062-dpo-v1-1-nemo-v1-mkmlizer: ║  Copyright 2023 MK ONE TECHNOLOGIES Inc.                            ║
jic062-dpo-v1-1-nemo-v1-mkmlizer: ║  https://mk1.ai                                                     ║
jic062-dpo-v1-1-nemo-v1-mkmlizer: ║                                                                     ║
jic062-dpo-v1-1-nemo-v1-mkmlizer: ║  The license key for the current software has been verified as      ║
jic062-dpo-v1-1-nemo-v1-mkmlizer: ║  belonging to:                                                      ║
jic062-dpo-v1-1-nemo-v1-mkmlizer: ║                                                                     ║
jic062-dpo-v1-1-nemo-v1-mkmlizer: ║  Chai Research Corp.                                                ║
jic062-dpo-v1-1-nemo-v1-mkmlizer: ║  Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f                   ║
jic062-dpo-v1-1-nemo-v1-mkmlizer: ║  Expiration: 2024-10-15 23:59:59                                    ║
jic062-dpo-v1-1-nemo-v1-mkmlizer: ║                                                                     ║
jic062-dpo-v1-1-nemo-v1-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
jic062-dpo-v1-1-nemo-v1-mkmlizer: Downloaded to shared memory in 53.521s
jic062-dpo-v1-1-nemo-v1-mkmlizer: quantizing model to /dev/shm/model_cache, profile:s0, folder:/tmp/tmpqo_wzrcq, device:0
jic062-dpo-v1-1-nemo-v1-mkmlizer: Saving flywheel model at /dev/shm/model_cache
jic062-dpo-v1-1-nemo-v1-mkmlizer: quantized model in 35.293s
jic062-dpo-v1-1-nemo-v1-mkmlizer: Processed model jic062/dpo-v1.1-Nemo in 88.815s
jic062-dpo-v1-1-nemo-v1-mkmlizer: creating bucket guanaco-mkml-models
jic062-dpo-v1-1-nemo-v1-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
jic062-dpo-v1-1-nemo-v1-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/jic062-dpo-v1-1-nemo-v1
jic062-dpo-v1-1-nemo-v1-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/jic062-dpo-v1-1-nemo-v1/config.json
jic062-dpo-v1-1-nemo-v1-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/jic062-dpo-v1-1-nemo-v1/special_tokens_map.json
jic062-dpo-v1-1-nemo-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/jic062-dpo-v1-1-nemo-v1/tokenizer_config.json
jic062-dpo-v1-1-nemo-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/jic062-dpo-v1-1-nemo-v1/tokenizer.json
jic062-dpo-v1-1-nemo-v1-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/jic062-dpo-v1-1-nemo-v1/flywheel_model.0.safetensors
jic062-dpo-v1-1-nemo-v1-mkmlizer: 
Loading 0:   0%|          | 0/363 [00:00<?, ?it/s]
Loading 0:   1%|▏         | 5/363 [00:00<00:10, 35.18it/s]
Loading 0:   4%|▎         | 13/363 [00:00<00:06, 54.18it/s]
Loading 0:   5%|▌         | 19/363 [00:00<00:07, 48.23it/s]
Loading 0:   7%|▋         | 24/363 [00:00<00:07, 46.54it/s]
Loading 0:   9%|▊         | 31/363 [00:00<00:06, 52.80it/s]
Loading 0:  10%|█         | 37/363 [00:00<00:06, 48.37it/s]
Loading 0:  12%|█▏        | 42/363 [00:00<00:06, 46.57it/s]
Loading 0:  13%|█▎        | 49/363 [00:00<00:05, 52.62it/s]
Loading 0:  15%|█▌        | 55/363 [00:01<00:06, 48.67it/s]
Loading 0:  17%|█▋        | 61/363 [00:01<00:07, 38.85it/s]
Loading 0:  18%|█▊        | 66/363 [00:01<00:07, 39.64it/s]
Loading 0:  20%|█▉        | 72/363 [00:01<00:06, 43.46it/s]
Loading 0:  21%|██▏       | 78/363 [00:01<00:06, 42.30it/s]
Loading 0:  23%|██▎       | 83/363 [00:01<00:06, 42.20it/s]
Loading 0:  25%|██▍       | 90/363 [00:01<00:05, 46.84it/s]
Loading 0:  26%|██▋       | 96/363 [00:02<00:06, 44.22it/s]
Loading 0:  28%|██▊       | 101/363 [00:02<00:06, 43.32it/s]
Loading 0:  29%|██▉       | 107/363 [00:02<00:05, 47.31it/s]
Loading 0:  31%|███       | 112/363 [00:02<00:05, 47.08it/s]
Loading 0:  32%|███▏      | 117/363 [00:02<00:05, 44.68it/s]
Loading 0:  34%|███▍      | 123/363 [00:02<00:05, 42.10it/s]
Loading 0:  35%|███▌      | 128/363 [00:02<00:05, 41.37it/s]
Loading 0:  37%|███▋      | 134/363 [00:02<00:05, 44.71it/s]
Loading 0:  38%|███▊      | 139/363 [00:03<00:05, 44.52it/s]
Loading 0:  40%|███▉      | 144/363 [00:03<00:07, 28.10it/s]
Loading 0:  41%|████      | 149/363 [00:03<00:06, 30.91it/s]
Loading 0:  43%|████▎     | 156/363 [00:03<00:05, 38.50it/s]
Loading 0:  44%|████▍     | 161/363 [00:03<00:05, 39.71it/s]
Loading 0:  46%|████▌     | 166/363 [00:03<00:04, 39.91it/s]
Loading 0:  47%|████▋     | 171/363 [00:04<00:04, 41.94it/s]
Loading 0:  48%|████▊     | 176/363 [00:04<00:05, 34.80it/s]
Loading 0:  50%|█████     | 183/363 [00:04<00:04, 41.71it/s]
Loading 0:  52%|█████▏    | 188/363 [00:04<00:04, 42.77it/s]
Loading 0:  53%|█████▎    | 193/363 [00:04<00:03, 43.35it/s]
Loading 0:  55%|█████▍    | 199/363 [00:04<00:03, 41.55it/s]
Loading 0:  56%|█████▌    | 204/363 [00:04<00:03, 41.13it/s]
Loading 0:  58%|█████▊    | 210/363 [00:04<00:03, 45.70it/s]
Loading 0:  59%|█████▉    | 215/363 [00:05<00:03, 45.84it/s]
Loading 0:  61%|██████    | 221/363 [00:05<00:02, 49.18it/s]
Loading 0:  63%|██████▎   | 227/363 [00:05<00:04, 32.28it/s]
Loading 0:  64%|██████▍   | 232/363 [00:05<00:03, 33.21it/s]
Loading 0:  65%|██████▌   | 237/363 [00:05<00:03, 36.35it/s]
Loading 0:  67%|██████▋   | 242/363 [00:05<00:03, 38.27it/s]
Loading 0:  68%|██████▊   | 247/363 [00:05<00:02, 39.58it/s]
Loading 0:  70%|██████▉   | 253/363 [00:06<00:02, 38.92it/s]
Loading 0:  71%|███████   | 258/363 [00:06<00:02, 38.68it/s]
Loading 0:  73%|███████▎  | 264/363 [00:06<00:02, 41.56it/s]
Loading 0:  74%|███████▍  | 269/363 [00:06<00:02, 42.32it/s]
Loading 0:  75%|███████▌  | 274/363 [00:06<00:02, 43.13it/s]
Loading 0:  77%|███████▋  | 280/363 [00:06<00:02, 41.29it/s]
Loading 0:  79%|███████▊  | 285/363 [00:06<00:01, 40.98it/s]
Loading 0:  80%|████████  | 292/363 [00:06<00:01, 46.02it/s]
Loading 0:  82%|████████▏ | 297/363 [00:07<00:01, 46.83it/s]
Loading 0:  83%|████████▎ | 302/363 [00:07<00:01, 46.44it/s]
Loading 0:  85%|████████▍ | 307/363 [00:13<00:22,  2.50it/s]
Loading 0:  86%|████████▌ | 312/363 [00:14<00:14,  3.43it/s]
Loading 0:  88%|████████▊ | 320/363 [00:14<00:07,  5.52it/s]
Loading 0:  90%|████████▉ | 326/363 [00:14<00:04,  7.41it/s]
Loading 0:  91%|█████████ | 331/363 [00:14<00:03,  9.45it/s]
Loading 0:  93%|█████████▎| 338/363 [00:14<00:01, 13.35it/s]
Loading 0:  95%|█████████▍| 344/363 [00:14<00:01, 16.57it/s]
Loading 0:  96%|█████████▌| 349/363 [00:14<00:00, 19.64it/s]
Loading 0:  98%|█████████▊| 355/363 [00:14<00:00, 24.80it/s]
Loading 0:  99%|█████████▉| 360/363 [00:15<00:00, 28.31it/s]
                                                            
Job jic062-dpo-v1-1-nemo-v1-mkmlizer completed after 114.15s with status: succeeded
Stopping job with name jic062-dpo-v1-1-nemo-v1-mkmlizer
Pipeline stage MKMLizer completed in 114.83s
run pipeline stage %s
Running pipeline stage MKMLTemplater
Pipeline stage MKMLTemplater completed in 0.07s
run pipeline stage %s
Running pipeline stage MKMLDeployer
Creating inference service jic062-dpo-v1-1-nemo-v1
Waiting for inference service jic062-dpo-v1-1-nemo-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
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
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
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
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
Connection pool is full, discarding connection: %s. Connection pool size: %s
Inference service jic062-dpo-v1-1-nemo-v1 ready after 160.59972143173218s
Pipeline stage MKMLDeployer completed in 161.12s
run pipeline stage %s
Running pipeline stage StressChecker
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
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
Received healthy response to inference request in 2.0912251472473145s
Received healthy response to inference request in 1.9181475639343262s
Received healthy response to inference request in 2.0403871536254883s
Received healthy response to inference request in 2.1676225662231445s
Received healthy response to inference request in 2.6576366424560547s
5 requests
0 failed requests
5th percentile: 1.9425954818725586
10th percentile: 1.967043399810791
20th percentile: 2.015939235687256
30th percentile: 2.0505547523498535
40th percentile: 2.070889949798584
50th percentile: 2.0912251472473145
60th percentile: 2.1217841148376464
70th percentile: 2.1523430824279783
80th percentile: 2.2656253814697265
90th percentile: 2.461631011962891
95th percentile: 2.5596338272094727
99th percentile: 2.638036079406738
mean time: 2.1750038146972654
Pipeline stage StressChecker completed in 11.64s
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 7.30s
Shutdown handler de-registered
jic062-dpo-v1-1-nemo_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 0.13s
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 jic062-dpo-v1-1-nemo-v1-profiler
Waiting for inference service jic062-dpo-v1-1-nemo-v1-profiler to be ready
Inference service jic062-dpo-v1-1-nemo-v1-profiler ready after 160.37012457847595s
Pipeline stage MKMLProfilerDeployer completed in 160.76s
run pipeline stage %s
Running pipeline stage MKMLProfilerRunner
kubectl cp /code/guanaco/guanaco_inference_services/src/inference_scripts tenant-chaiml-guanaco/jic062-dpo-v1-1-nemo-v1-profiler-predictor-00001-deploymen62mbv:/code/chaiverse_profiler_1725941885 --namespace tenant-chaiml-guanaco
kubectl exec -it jic062-dpo-v1-1-nemo-v1-profiler-predictor-00001-deploymen62mbv --namespace tenant-chaiml-guanaco -- sh -c 'cd /code/chaiverse_profiler_1725941885 && 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_1725941885/summary.json'
kubectl exec -it jic062-dpo-v1-1-nemo-v1-profiler-predictor-00001-deploymen62mbv --namespace tenant-chaiml-guanaco -- bash -c 'cat /code/chaiverse_profiler_1725941885/summary.json'
Pipeline stage MKMLProfilerRunner completed in 952.32s
run pipeline stage %s
Running pipeline stage MKMLProfilerDeleter
Checking if service jic062-dpo-v1-1-nemo-v1-profiler is running
Tearing down inference service jic062-dpo-v1-1-nemo-v1-profiler
Service jic062-dpo-v1-1-nemo-v1-profiler has been torndown
Pipeline stage MKMLProfilerDeleter completed in 2.06s
Shutdown handler de-registered
jic062-dpo-v1-1-nemo_v1 status is now inactive due to auto deactivation removed underperforming models
jic062-dpo-v1-1-nemo_v1 status is now torndown due to DeploymentManager action