developer_uid: chace9580
submission_id: jic062-dpo-v1-5-nemo-c1500_v2
model_name: jic062-dpo-v1-5-nemo-c1500_v2
model_group: jic062/dpo-v1.5-Nemo-c15
status: torndown
timestamp: 2024-09-15T00:32:57+00:00
num_battles: 10351
num_wins: 4990
celo_rating: 1240.31
family_friendly_score: 0.0
submission_type: basic
model_repo: jic062/dpo-v1.5-Nemo-c1500
model_architecture: MistralForCausalLM
model_num_parameters: 12772070400.0
best_of: 8
max_input_tokens: 512
max_output_tokens: 64
latencies: [{'batch_size': 1, 'throughput': 0.6992594218263102, 'latency_mean': 1.429983868598938, 'latency_p50': 1.438476324081421, 'latency_p90': 1.5873498678207398}, {'batch_size': 3, 'throughput': 1.3308816128928191, 'latency_mean': 2.2486436223983763, 'latency_p50': 2.2733291387557983, 'latency_p90': 2.4835954666137696}, {'batch_size': 5, 'throughput': 1.5912264328001946, 'latency_mean': 3.1294542598724364, 'latency_p50': 3.122525215148926, 'latency_p90': 3.528996086120605}, {'batch_size': 6, 'throughput': 1.621357286551791, 'latency_mean': 3.6680760645866393, 'latency_p50': 3.67388916015625, 'latency_p90': 4.143923735618591}, {'batch_size': 8, 'throughput': 1.5868483280082395, 'latency_mean': 5.01212232708931, 'latency_p50': 5.072692036628723, 'latency_p90': 5.655879426002502}, {'batch_size': 10, 'throughput': 1.5626448224406417, 'latency_mean': 6.3573137831687925, 'latency_p50': 6.434977054595947, 'latency_p90': 7.225063586235046}]
gpu_counts: {'NVIDIA RTX A5000': 1}
display_name: jic062-dpo-v1-5-nemo-c1500_v2
is_internal_developer: False
language_model: jic062/dpo-v1.5-Nemo-c1500
model_size: 13B
ranking_group: single
throughput_3p7s: 1.63
us_pacific_date: 2024-09-14
win_ratio: 0.4820790261810453
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', '\n\n', '\nYou:', '[/INST]', '<|im_end|>', '</s>'], '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-v1-5-nemo-c1500-v2-mkmlizer
Waiting for job on jic062-dpo-v1-5-nemo-c1500-v2-mkmlizer to finish
jic062-dpo-v1-5-nemo-c1500-v2-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
jic062-dpo-v1-5-nemo-c1500-v2-mkmlizer: ║ _____ __ __ ║
jic062-dpo-v1-5-nemo-c1500-v2-mkmlizer: ║ / _/ /_ ___ __/ / ___ ___ / / ║
jic062-dpo-v1-5-nemo-c1500-v2-mkmlizer: ║ / _/ / // / |/|/ / _ \/ -_) -_) / ║
jic062-dpo-v1-5-nemo-c1500-v2-mkmlizer: ║ /_//_/\_, /|__,__/_//_/\__/\__/_/ ║
jic062-dpo-v1-5-nemo-c1500-v2-mkmlizer: ║ /___/ ║
jic062-dpo-v1-5-nemo-c1500-v2-mkmlizer: ║ ║
jic062-dpo-v1-5-nemo-c1500-v2-mkmlizer: ║ Version: 0.10.1 ║
jic062-dpo-v1-5-nemo-c1500-v2-mkmlizer: ║ Copyright 2023 MK ONE TECHNOLOGIES Inc. ║
jic062-dpo-v1-5-nemo-c1500-v2-mkmlizer: ║ https://mk1.ai ║
jic062-dpo-v1-5-nemo-c1500-v2-mkmlizer: ║ ║
jic062-dpo-v1-5-nemo-c1500-v2-mkmlizer: ║ The license key for the current software has been verified as ║
jic062-dpo-v1-5-nemo-c1500-v2-mkmlizer: ║ belonging to: ║
jic062-dpo-v1-5-nemo-c1500-v2-mkmlizer: ║ ║
jic062-dpo-v1-5-nemo-c1500-v2-mkmlizer: ║ Chai Research Corp. ║
jic062-dpo-v1-5-nemo-c1500-v2-mkmlizer: ║ Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f ║
jic062-dpo-v1-5-nemo-c1500-v2-mkmlizer: ║ Expiration: 2024-10-15 23:59:59 ║
jic062-dpo-v1-5-nemo-c1500-v2-mkmlizer: ║ ║
jic062-dpo-v1-5-nemo-c1500-v2-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
Failed to get response for submission blend_hokok_2024-09-09: ('http://neversleep-noromaid-v0-8068-v150-predictor.tenant-chaiml-guanaco.k.chaiverse.com/v1/models/GPT-J-6B-lit-v2:predict', '')
jic062-dpo-v1-5-nemo-c1500-v2-mkmlizer: Downloaded to shared memory in 29.780s
jic062-dpo-v1-5-nemo-c1500-v2-mkmlizer: quantizing model to /dev/shm/model_cache, profile:s0, folder:/tmp/tmp_hwgq7d9, device:0
jic062-dpo-v1-5-nemo-c1500-v2-mkmlizer: Saving flywheel model at /dev/shm/model_cache
jic062-dpo-v1-5-nemo-c1500-v2-mkmlizer: quantized model in 35.447s
jic062-dpo-v1-5-nemo-c1500-v2-mkmlizer: Processed model jic062/dpo-v1.5-Nemo-c1500 in 65.227s
jic062-dpo-v1-5-nemo-c1500-v2-mkmlizer: creating bucket guanaco-mkml-models
jic062-dpo-v1-5-nemo-c1500-v2-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
jic062-dpo-v1-5-nemo-c1500-v2-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/jic062-dpo-v1-5-nemo-c1500-v2
jic062-dpo-v1-5-nemo-c1500-v2-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/jic062-dpo-v1-5-nemo-c1500-v2/config.json
jic062-dpo-v1-5-nemo-c1500-v2-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/jic062-dpo-v1-5-nemo-c1500-v2/special_tokens_map.json
jic062-dpo-v1-5-nemo-c1500-v2-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/jic062-dpo-v1-5-nemo-c1500-v2/flywheel_model.0.safetensors
jic062-dpo-v1-5-nemo-c1500-v2-mkmlizer: Loading 0: 0%| | 0/363 [00:00<?, ?it/s] Loading 0: 1%|▏ | 5/363 [00:00<00:10, 34.47it/s] Loading 0: 4%|▎ | 13/363 [00:00<00:06, 54.22it/s] Loading 0: 5%|▌ | 19/363 [00:00<00:07, 47.24it/s] Loading 0: 7%|▋ | 24/363 [00:00<00:07, 43.26it/s] Loading 0: 8%|▊ | 30/363 [00:00<00:06, 47.82it/s] Loading 0: 10%|▉ | 35/363 [00:00<00:07, 46.80it/s] Loading 0: 11%|█ | 40/363 [00:00<00:07, 46.09it/s] Loading 0: 13%|█▎ | 46/363 [00:01<00:07, 43.23it/s] Loading 0: 14%|█▍ | 51/363 [00:01<00:07, 43.02it/s] Loading 0: 16%|█▋ | 59/363 [00:01<00:05, 52.52it/s] Loading 0: 18%|█▊ | 65/363 [00:01<00:09, 32.45it/s] Loading 0: 20%|█▉ | 72/363 [00:01<00:07, 38.81it/s] Loading 0: 21%|██ | 77/363 [00:01<00:07, 40.64it/s] Loading 0: 23%|██▎ | 82/363 [00:01<00:07, 35.51it/s] Loading 0: 25%|██▍ | 90/363 [00:02<00:06, 43.48it/s] Loading 0: 26%|██▋ | 96/363 [00:02<00:06, 42.29it/s] Loading 0: 28%|██▊ | 101/363 [00:02<00:06, 41.97it/s] Loading 0: 30%|██▉ | 108/363 [00:02<00:05, 48.19it/s] Loading 0: 31%|███▏ | 114/363 [00:02<00:05, 44.16it/s] Loading 0: 33%|███▎ | 119/363 [00:02<00:06, 40.02it/s] Loading 0: 34%|███▍ | 125/363 [00:02<00:05, 43.06it/s] Loading 0: 36%|███▌ | 130/363 [00:03<00:05, 41.95it/s] Loading 0: 37%|███▋ | 135/363 [00:03<00:05, 42.64it/s] Loading 0: 39%|███▉ | 141/363 [00:03<00:05, 41.08it/s] Loading 0: 40%|████ | 146/363 [00:03<00:07, 29.98it/s] Loading 0: 41%|████▏ | 150/363 [00:03<00:07, 30.28it/s] Loading 0: 43%|████▎ | 156/363 [00:03<00:05, 36.14it/s] Loading 0: 44%|████▍ | 161/363 [00:03<00:05, 38.52it/s] Loading 0: 46%|████▌ | 166/363 [00:04<00:04, 40.19it/s] Loading 0: 47%|████▋ | 171/363 [00:04<00:04, 42.34it/s] Loading 0: 48%|████▊ | 176/363 [00:04<00:05, 36.58it/s] Loading 0: 50%|█████ | 183/363 [00:04<00:04, 44.35it/s] Loading 0: 52%|█████▏ | 188/363 [00:04<00:03, 44.20it/s] Loading 0: 53%|█████▎ | 193/363 [00:04<00:03, 44.48it/s] Loading 0: 55%|█████▍ | 199/363 [00:04<00:03, 42.41it/s] Loading 0: 56%|█████▌ | 204/363 [00:04<00:03, 41.75it/s] Loading 0: 58%|█████▊ | 210/363 [00:05<00:03, 44.61it/s] Loading 0: 59%|█████▉ | 215/363 [00:05<00:03, 44.76it/s] Loading 0: 61%|██████ | 222/363 [00:05<00:03, 45.18it/s] Loading 0: 63%|██████▎ | 227/363 [00:05<00:04, 33.64it/s] Loading 0: 64%|██████▎ | 231/363 [00:05<00:03, 33.47it/s] Loading 0: 65%|██████▌ | 237/363 [00:05<00:03, 37.83it/s] Loading 0: 67%|██████▋ | 242/363 [00:05<00:03, 38.95it/s] Loading 0: 68%|██████▊ | 247/363 [00:06<00:02, 39.65it/s] Loading 0: 70%|██████▉ | 253/363 [00:06<00:02, 39.07it/s] Loading 0: 71%|███████ | 258/363 [00:06<00:02, 38.66it/s] Loading 0: 73%|███████▎ | 264/363 [00:06<00:02, 43.57it/s] Loading 0: 74%|███████▍ | 269/363 [00:06<00:02, 43.76it/s] Loading 0: 75%|███████▌ | 274/363 [00:06<00:02, 43.69it/s] Loading 0: 77%|███████▋ | 279/363 [00:06<00:01, 44.75it/s] Loading 0: 78%|███████▊ | 284/363 [00:06<00:02, 36.49it/s] Loading 0: 80%|████████ | 291/363 [00:07<00:01, 43.87it/s] Loading 0: 82%|████████▏ | 296/363 [00:07<00:01, 43.90it/s] Loading 0: 83%|████████▎ | 302/363 [00:07<00:01, 47.65it/s] Loading 0: 85%|████████▍ | 308/363 [00:14<00:20, 2.71it/s] Loading 0: 86%|████████▌ | 312/363 [00:14<00:14, 3.44it/s] Loading 0: 88%|████████▊ | 320/363 [00:14<00:07, 5.53it/s] Loading 0: 90%|████████▉ | 326/363 [00:14<00:04, 7.42it/s] Loading 0: 91%|█████████ | 331/363 [00:14<00:03, 9.49it/s] Loading 0: 93%|█████████▎| 338/363 [00:14<00:01, 13.45it/s] Loading 0: 95%|█████████▍| 344/363 [00:14<00:01, 16.90it/s] Loading 0: 96%|█████████▌| 349/363 [00:14<00:00, 20.12it/s] Loading 0: 98%|█████████▊| 356/363 [00:15<00:00, 26.31it/s] Loading 0: 100%|█████████▉| 362/363 [00:15<00:00, 29.41it/s]
Job jic062-dpo-v1-5-nemo-c1500-v2-mkmlizer completed after 83.96s with status: succeeded
Stopping job with name jic062-dpo-v1-5-nemo-c1500-v2-mkmlizer
Pipeline stage MKMLizer completed in 87.62s
run pipeline stage %s
Running pipeline stage MKMLTemplater
Pipeline stage MKMLTemplater completed in 0.52s
run pipeline stage %s
Running pipeline stage MKMLDeployer
Creating inference service jic062-dpo-v1-5-nemo-c1500-v2
Waiting for inference service jic062-dpo-v1-5-nemo-c1500-v2 to be ready
Failed to get response for submission blend_hokok_2024-09-09: ('http://neversleep-noromaid-v0-8068-v150-predictor.tenant-chaiml-guanaco.k.chaiverse.com/v1/models/GPT-J-6B-lit-v2:predict', '')
Inference service jic062-dpo-v1-5-nemo-c1500-v2 ready after 170.88809490203857s
Pipeline stage MKMLDeployer completed in 171.45s
run pipeline stage %s
Running pipeline stage StressChecker
Received healthy response to inference request in 3.0665841102600098s
Received healthy response to inference request in 2.3170433044433594s
Received healthy response to inference request in 2.16507625579834s
Received healthy response to inference request in 1.815838098526001s
Received healthy response to inference request in 2.0386335849761963s
5 requests
0 failed requests
5th percentile: 1.8603971958160401
10th percentile: 1.904956293106079
20th percentile: 1.9940744876861571
30th percentile: 2.063922119140625
40th percentile: 2.1144991874694825
50th percentile: 2.16507625579834
60th percentile: 2.2258630752563477
70th percentile: 2.2866498947143556
80th percentile: 2.4669514656066895
90th percentile: 2.7667677879333494
95th percentile: 2.9166759490966796
99th percentile: 3.0366024780273437
mean time: 2.2806350708007814
Pipeline stage StressChecker completed in 12.26s
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 6.14s
Shutdown handler de-registered
jic062-dpo-v1-5-nemo-c1500_v2 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 jic062-dpo-v1-5-nemo-c1500-v2-profiler
Waiting for inference service jic062-dpo-v1-5-nemo-c1500-v2-profiler to be ready
Inference service jic062-dpo-v1-5-nemo-c1500-v2-profiler ready after 170.38193273544312s
Pipeline stage MKMLProfilerDeployer completed in 171.24s
run pipeline stage %s
Running pipeline stage MKMLProfilerRunner
kubectl cp /code/guanaco/guanaco_inference_services/src/inference_scripts tenant-chaiml-guanaco/jic062-dpo-v1-5-nemocecd638b800d143eb6ad5b173b62f666-deplorzhsf:/code/chaiverse_profiler_1726360872 --namespace tenant-chaiml-guanaco
kubectl exec -it jic062-dpo-v1-5-nemocecd638b800d143eb6ad5b173b62f666-deplorzhsf --namespace tenant-chaiml-guanaco -- sh -c 'cd /code/chaiverse_profiler_1726360872 && 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_1726360872/summary.json'
kubectl exec -it jic062-dpo-v1-5-nemocecd638b800d143eb6ad5b173b62f666-deplorzhsf --namespace tenant-chaiml-guanaco -- bash -c 'cat /code/chaiverse_profiler_1726360872/summary.json'
Pipeline stage MKMLProfilerRunner completed in 949.38s
run pipeline stage %s
Running pipeline stage MKMLProfilerDeleter
Checking if service jic062-dpo-v1-5-nemo-c1500-v2-profiler is running
Tearing down inference service jic062-dpo-v1-5-nemo-c1500-v2-profiler
Service jic062-dpo-v1-5-nemo-c1500-v2-profiler has been torndown
Pipeline stage MKMLProfilerDeleter completed in 2.89s
Shutdown handler de-registered
jic062-dpo-v1-5-nemo-c1500_v2 status is now inactive due to auto deactivation removed underperforming models
jic062-dpo-v1-5-nemo-c1500_v2 status is now torndown due to DeploymentManager action