jic062-dpo-v1-nemo

developer_uid: chace9580

submission_id: jic062-dpo-v1-nemo_v3

model_name: jic062-dpo-v1-nemo_v2

model_group: jic062/dpo-v1-Nemo

status: torndown

timestamp: 2024-09-09T17:09:42+00:00

num_battles: 11651

num_wins: 1007

celo_rating: 833.92

family_friendly_score: 0.0

submission_type: basic

model_repo: jic062/dpo-v1-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': 0.6962908812037444, 'latency_mean': 1.436084886789322, 'latency_p50': 1.438718557357788, 'latency_p90': 1.6030754566192627}, {'batch_size': 3, 'throughput': 1.3223977828168285, 'latency_mean': 2.2675793242454527, 'latency_p50': 2.286741852760315, 'latency_p90': 2.5000020265579224}, {'batch_size': 5, 'throughput': 1.547673507856371, 'latency_mean': 3.2131723058223725, 'latency_p50': 3.205398201942444, 'latency_p90': 3.6095923662185667}, {'batch_size': 6, 'throughput': 1.600662621715721, 'latency_mean': 3.727473603487015, 'latency_p50': 3.7468535900115967, 'latency_p90': 4.238950896263122}, {'batch_size': 8, 'throughput': 1.5701299376462328, 'latency_mean': 5.069521301984787, 'latency_p50': 5.019721627235413, 'latency_p90': 5.873053860664368}, {'batch_size': 10, 'throughput': 1.5310534777692537, 'latency_mean': 6.478893260955811, 'latency_p50': 6.577357649803162, 'latency_p90': 7.3410313606262205}]

gpu_counts: {'NVIDIA RTX A5000': 1}

display_name: jic062-dpo-v1-nemo_v2

is_internal_developer: False

language_model: jic062/dpo-v1-Nemo

model_size: 13B

ranking_group: single

throughput_3p7s: 1.61

us_pacific_date: 2024-09-09

win_ratio: 0.0864303493262381

generation_params: {'temperature': 1.0, 'top_p': 1.0, 'min_p': 0.1, 'top_k': 40, 'presence_penalty': 0.0, 'frequency_penalty': 0.0, 'stopping_words': ['\n', '\n\n', '\nYou:', '[/INST]', '<|im_end|>', '</s>'], 'max_input_tokens': 512, 'best_of': 8, 'max_output_tokens': 64}

formatter: {'memory_template': '', 'prompt_template': '', 'bot_template': '', 'user_template': '', 'response_template': '', '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-nemo-v3-mkmlizer
Waiting for job on jic062-dpo-v1-nemo-v3-mkmlizer to finish
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
jic062-dpo-v1-nemo-v3-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
jic062-dpo-v1-nemo-v3-mkmlizer: ║     _____            __           __                                ║
jic062-dpo-v1-nemo-v3-mkmlizer: ║    / _/ /_ ___    __/ /  ___ ___ / /                                ║
jic062-dpo-v1-nemo-v3-mkmlizer: ║   / _/ / // / |/|/ / _ \/ -_) -_) /                                 ║
jic062-dpo-v1-nemo-v3-mkmlizer: ║  /_//_/\_, /|__,__/_//_/\__/\__/_/                                  ║
jic062-dpo-v1-nemo-v3-mkmlizer: ║       /___/                                                         ║
jic062-dpo-v1-nemo-v3-mkmlizer: ║                                                                     ║
jic062-dpo-v1-nemo-v3-mkmlizer: ║  Version: 0.10.1                                                    ║
jic062-dpo-v1-nemo-v3-mkmlizer: ║  Copyright 2023 MK ONE TECHNOLOGIES Inc.                            ║
jic062-dpo-v1-nemo-v3-mkmlizer: ║  https://mk1.ai                                                     ║
jic062-dpo-v1-nemo-v3-mkmlizer: ║                                                                     ║
jic062-dpo-v1-nemo-v3-mkmlizer: ║  The license key for the current software has been verified as      ║
jic062-dpo-v1-nemo-v3-mkmlizer: ║  belonging to:                                                      ║
jic062-dpo-v1-nemo-v3-mkmlizer: ║                                                                     ║
jic062-dpo-v1-nemo-v3-mkmlizer: ║  Chai Research Corp.                                                ║
jic062-dpo-v1-nemo-v3-mkmlizer: ║  Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f                   ║
jic062-dpo-v1-nemo-v3-mkmlizer: ║  Expiration: 2024-10-15 23:59:59                                    ║
jic062-dpo-v1-nemo-v3-mkmlizer: ║                                                                     ║
jic062-dpo-v1-nemo-v3-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
Retrying (%r) after connection broken by '%r': %s
jic062-dpo-v1-nemo-v3-mkmlizer: Downloaded to shared memory in 28.398s
jic062-dpo-v1-nemo-v3-mkmlizer: quantizing model to /dev/shm/model_cache, profile:s0, folder:/tmp/tmp7yyg2b7l, device:0
jic062-dpo-v1-nemo-v3-mkmlizer: Saving flywheel model at /dev/shm/model_cache
jic062-dpo-v1-nemo-v3-mkmlizer: quantized model in 35.811s
jic062-dpo-v1-nemo-v3-mkmlizer: Processed model jic062/dpo-v1-Nemo in 64.210s
jic062-dpo-v1-nemo-v3-mkmlizer: creating bucket guanaco-mkml-models
jic062-dpo-v1-nemo-v3-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
jic062-dpo-v1-nemo-v3-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/jic062-dpo-v1-nemo-v3
jic062-dpo-v1-nemo-v3-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/jic062-dpo-v1-nemo-v3/config.json
jic062-dpo-v1-nemo-v3-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/jic062-dpo-v1-nemo-v3/special_tokens_map.json
jic062-dpo-v1-nemo-v3-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/jic062-dpo-v1-nemo-v3/tokenizer_config.json
jic062-dpo-v1-nemo-v3-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/jic062-dpo-v1-nemo-v3/tokenizer.json
jic062-dpo-v1-nemo-v3-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/jic062-dpo-v1-nemo-v3/flywheel_model.0.safetensors
jic062-dpo-v1-nemo-v3-mkmlizer: 
Loading 0:   0%|          | 0/363 [00:00<?, ?it/s]
Loading 0:   1%|▏         | 5/363 [00:00<00:10, 33.31it/s]
Loading 0:   4%|▎         | 13/363 [00:00<00:06, 54.04it/s]
Loading 0:   5%|▌         | 19/363 [00:00<00:06, 49.19it/s]
Loading 0:   7%|▋         | 25/363 [00:00<00:06, 50.19it/s]
Loading 0:   9%|▊         | 31/363 [00:00<00:06, 50.56it/s]
Loading 0:  10%|█         | 37/363 [00:00<00:07, 45.23it/s]
Loading 0:  12%|█▏        | 42/363 [00:00<00:07, 43.92it/s]
Loading 0:  13%|█▎        | 48/363 [00:01<00:06, 47.21it/s]
Loading 0:  15%|█▍        | 53/363 [00:01<00:06, 46.00it/s]
Loading 0:  17%|█▋        | 60/363 [00:01<00:06, 46.20it/s]
Loading 0:  18%|█▊        | 65/363 [00:01<00:09, 31.62it/s]
Loading 0:  20%|█▉        | 72/363 [00:01<00:07, 38.55it/s]
Loading 0:  21%|██        | 77/363 [00:01<00:07, 40.16it/s]
Loading 0:  23%|██▎       | 82/363 [00:01<00:07, 35.34it/s]
Loading 0:  25%|██▍       | 90/363 [00:02<00:06, 43.65it/s]
Loading 0:  26%|██▋       | 96/363 [00:02<00:06, 42.12it/s]
Loading 0:  28%|██▊       | 101/363 [00:02<00:06, 41.22it/s]
Loading 0:  30%|██▉       | 108/363 [00:02<00:05, 47.86it/s]
Loading 0:  31%|███▏      | 114/363 [00:02<00:05, 41.58it/s]
Loading 0:  33%|███▎      | 119/363 [00:02<00:06, 40.26it/s]
Loading 0:  34%|███▍      | 125/363 [00:02<00:05, 43.07it/s]
Loading 0:  36%|███▌      | 130/363 [00:03<00:05, 42.88it/s]
Loading 0:  37%|███▋      | 135/363 [00:03<00:05, 41.57it/s]
Loading 0:  39%|███▊      | 140/363 [00:03<00:05, 42.83it/s]
Loading 0:  40%|███▉      | 145/363 [00:03<00:07, 27.97it/s]
Loading 0:  41%|████      | 149/363 [00:03<00:07, 28.50it/s]
Loading 0:  43%|████▎     | 156/363 [00:03<00:05, 36.62it/s]
Loading 0:  44%|████▍     | 161/363 [00:03<00:05, 37.26it/s]
Loading 0:  46%|████▌     | 166/363 [00:04<00:05, 38.24it/s]
Loading 0:  47%|████▋     | 172/363 [00:04<00:04, 39.10it/s]
Loading 0:  49%|████▉     | 177/363 [00:04<00:04, 39.32it/s]
Loading 0:  50%|█████     | 183/363 [00:04<00:04, 44.28it/s]
Loading 0:  52%|█████▏    | 188/363 [00:04<00:03, 44.11it/s]
Loading 0:  53%|█████▎    | 193/363 [00:04<00:03, 44.31it/s]
Loading 0:  55%|█████▍    | 198/363 [00:04<00:03, 45.38it/s]
Loading 0:  56%|█████▌    | 203/363 [00:05<00:04, 36.08it/s]
Loading 0:  58%|█████▊    | 210/363 [00:05<00:03, 41.99it/s]
Loading 0:  59%|█████▉    | 215/363 [00:05<00:03, 41.87it/s]
Loading 0:  61%|██████    | 221/363 [00:05<00:03, 45.89it/s]
Loading 0:  62%|██████▏   | 226/363 [00:05<00:04, 28.15it/s]
Loading 0:  63%|██████▎   | 230/363 [00:05<00:04, 28.48it/s]
Loading 0:  65%|██████▌   | 237/363 [00:05<00:03, 36.29it/s]
Loading 0:  67%|██████▋   | 242/363 [00:06<00:03, 38.20it/s]
Loading 0:  68%|██████▊   | 247/363 [00:06<00:02, 39.88it/s]
Loading 0:  69%|██████▉   | 252/363 [00:06<00:02, 40.95it/s]
Loading 0:  71%|███████   | 257/363 [00:06<00:02, 35.40it/s]
Loading 0:  73%|███████▎  | 265/363 [00:06<00:02, 43.59it/s]
Loading 0:  75%|███████▍  | 271/363 [00:06<00:02, 42.52it/s]
Loading 0:  76%|███████▌  | 276/363 [00:06<00:02, 41.13it/s]
Loading 0:  78%|███████▊  | 283/363 [00:06<00:01, 45.70it/s]
Loading 0:  80%|███████▉  | 289/363 [00:07<00:01, 43.85it/s]
Loading 0:  81%|████████  | 294/363 [00:07<00:01, 37.19it/s]
Loading 0:  82%|████████▏ | 299/363 [00:07<00:01, 38.20it/s]
Loading 0:  84%|████████▎ | 304/363 [00:14<00:23,  2.53it/s]
Loading 0:  85%|████████▍ | 307/363 [00:14<00:18,  3.07it/s]
Loading 0:  86%|████████▌ | 312/363 [00:14<00:11,  4.33it/s]
Loading 0:  88%|████████▊ | 320/363 [00:14<00:06,  7.15it/s]
Loading 0:  90%|████████▉ | 326/363 [00:14<00:03,  9.63it/s]
Loading 0:  91%|█████████ | 331/363 [00:14<00:02, 12.13it/s]
Loading 0:  93%|█████████▎| 338/363 [00:14<00:01, 16.89it/s]
Loading 0:  95%|█████████▍| 344/363 [00:15<00:00, 20.58it/s]
Loading 0:  96%|█████████▌| 349/363 [00:15<00:00, 23.39it/s]
Loading 0:  98%|█████████▊| 356/363 [00:15<00:00, 29.75it/s]
Loading 0: 100%|█████████▉| 362/363 [00:15<00:00, 32.39it/s]
                                                            
Job jic062-dpo-v1-nemo-v3-mkmlizer completed after 85.98s with status: succeeded
Stopping job with name jic062-dpo-v1-nemo-v3-mkmlizer
Pipeline stage MKMLizer completed in 87.18s
run pipeline stage %s
Running pipeline stage MKMLTemplater
Pipeline stage MKMLTemplater completed in 0.09s
run pipeline stage %s
Running pipeline stage MKMLDeployer
Creating inference service jic062-dpo-v1-nemo-v3
Waiting for inference service jic062-dpo-v1-nemo-v3 to be ready
Failed to get response for submission epiculous-azure-dusk-v0-2_v1: ('http://chaiml-llama-8b-pairwis-8189-v19-predictor.tenant-chaiml-guanaco.k.chaiverse.com/v1/models/GPT-J-6B-lit-v2:predict', 'read tcp 127.0.0.1:59188->127.0.0.1:8080: read: connection reset by peer\n')
Failed to get response for submission blend_siken_2024-09-09: ('http://chaiml-llama-8b-pairwis-8189-v19-predictor.tenant-chaiml-guanaco.k.chaiverse.com/v1/models/GPT-J-6B-lit-v2:predict', 'readfrom tcp 127.0.0.1:37568->127.0.0.1:8080: write tcp 127.0.0.1:37568->127.0.0.1:8080: use of closed network connection\n')
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-nemo-v3 ready after 150.6018831729889s
Pipeline stage MKMLDeployer completed in 151.27s
run pipeline stage %s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.2581875324249268s
Received healthy response to inference request in 1.6919026374816895s
Received healthy response to inference request in 1.7259552478790283s
Received healthy response to inference request in 1.933830738067627s
Received healthy response to inference request in 1.6427502632141113s
5 requests
0 failed requests
5th percentile: 1.652580738067627
10th percentile: 1.6624112129211426
20th percentile: 1.6820721626281738
30th percentile: 1.6987131595611573
40th percentile: 1.7123342037200928
50th percentile: 1.7259552478790283
60th percentile: 1.8091054439544678
70th percentile: 1.892255640029907
80th percentile: 1.998702096939087
90th percentile: 2.1284448146820067
95th percentile: 2.1933161735534665
99th percentile: 2.2452132606506345
mean time: 1.8505252838134765
Pipeline stage StressChecker completed in 11.42s
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 7.38s
Shutdown handler de-registered
jic062-dpo-v1-nemo_v3 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 0.15s
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-nemo-v3-profiler
Waiting for inference service jic062-dpo-v1-nemo-v3-profiler to be ready
Inference service jic062-dpo-v1-nemo-v3-profiler ready after 150.373211145401s
Pipeline stage MKMLProfilerDeployer completed in 150.74s
run pipeline stage %s
Running pipeline stage MKMLProfilerRunner
kubectl cp /code/guanaco/guanaco_inference_services/src/inference_scripts tenant-chaiml-guanaco/jic062-dpo-v1-nemo-v3-profiler-predictor-00001-deployment-9pzgh:/code/chaiverse_profiler_1725902258 --namespace tenant-chaiml-guanaco
kubectl exec -it jic062-dpo-v1-nemo-v3-profiler-predictor-00001-deployment-9pzgh --namespace tenant-chaiml-guanaco -- sh -c 'cd /code/chaiverse_profiler_1725902258 && 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_1725902258/summary.json'
kubectl exec -it jic062-dpo-v1-nemo-v3-profiler-predictor-00001-deployment-9pzgh --namespace tenant-chaiml-guanaco -- bash -c 'cat /code/chaiverse_profiler_1725902258/summary.json'
Pipeline stage MKMLProfilerRunner completed in 958.80s
run pipeline stage %s
Running pipeline stage MKMLProfilerDeleter
Checking if service jic062-dpo-v1-nemo-v3-profiler is running
Tearing down inference service jic062-dpo-v1-nemo-v3-profiler
Service jic062-dpo-v1-nemo-v3-profiler has been torndown
Pipeline stage MKMLProfilerDeleter completed in 1.89s
Shutdown handler de-registered
jic062-dpo-v1-nemo_v3 status is now inactive due to auto deactivation removed underperforming models
jic062-dpo-v1-nemo_v3 status is now torndown due to DeploymentManager action