developer_uid: chace9580
submission_id: jic062-dpo-v1-3-nemo-c500_v2
model_name: jic062-dpo-v1-3-nemo-c500_v2
model_group: jic062/dpo-v1.3-Nemo-c50
status: torndown
timestamp: 2024-09-12T04:40:32+00:00
num_battles: 11203
num_wins: 5546
celo_rating: 1240.86
family_friendly_score: 0.0
submission_type: basic
model_repo: jic062/dpo-v1.3-Nemo-c500
model_architecture: MistralForCausalLM
model_num_parameters: 12772070400.0
best_of: 8
max_input_tokens: 512
max_output_tokens: 64
latencies: [{'batch_size': 1, 'throughput': 0.6952263775753618, 'latency_mean': 1.4383211421966553, 'latency_p50': 1.4358549118041992, 'latency_p90': 1.604235291481018}, {'batch_size': 3, 'throughput': 1.3406045128151065, 'latency_mean': 2.2301754558086397, 'latency_p50': 2.2387841939926147, 'latency_p90': 2.4756609439849853}, {'batch_size': 5, 'throughput': 1.5807201753195377, 'latency_mean': 3.1491056847572327, 'latency_p50': 3.134657144546509, 'latency_p90': 3.54496054649353}, {'batch_size': 6, 'throughput': 1.6185394336924122, 'latency_mean': 3.6828977155685423, 'latency_p50': 3.6661629676818848, 'latency_p90': 4.1650789976119995}, {'batch_size': 8, 'throughput': 1.609824091162467, 'latency_mean': 4.926561148166656, 'latency_p50': 4.936550736427307, 'latency_p90': 5.621658849716186}, {'batch_size': 10, 'throughput': 1.5516397482971414, 'latency_mean': 6.408377851247788, 'latency_p50': 6.449912190437317, 'latency_p90': 7.305176830291748}]
gpu_counts: {'NVIDIA RTX A5000': 1}
display_name: jic062-dpo-v1-3-nemo-c500_v2
is_internal_developer: False
language_model: jic062/dpo-v1.3-Nemo-c500
model_size: 13B
ranking_group: single
throughput_3p7s: 1.63
us_pacific_date: 2024-09-11
win_ratio: 0.49504596982950994
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-v1-3-nemo-c500-v2-mkmlizer
Waiting for job on jic062-dpo-v1-3-nemo-c500-v2-mkmlizer to finish
jic062-dpo-v1-3-nemo-c500-v2-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
jic062-dpo-v1-3-nemo-c500-v2-mkmlizer: ║ _____ __ __ ║
jic062-dpo-v1-3-nemo-c500-v2-mkmlizer: ║ / _/ /_ ___ __/ / ___ ___ / / ║
jic062-dpo-v1-3-nemo-c500-v2-mkmlizer: ║ / _/ / // / |/|/ / _ \/ -_) -_) / ║
jic062-dpo-v1-3-nemo-c500-v2-mkmlizer: ║ /_//_/\_, /|__,__/_//_/\__/\__/_/ ║
jic062-dpo-v1-3-nemo-c500-v2-mkmlizer: ║ /___/ ║
jic062-dpo-v1-3-nemo-c500-v2-mkmlizer: ║ ║
jic062-dpo-v1-3-nemo-c500-v2-mkmlizer: ║ Version: 0.10.1 ║
jic062-dpo-v1-3-nemo-c500-v2-mkmlizer: ║ Copyright 2023 MK ONE TECHNOLOGIES Inc. ║
jic062-dpo-v1-3-nemo-c500-v2-mkmlizer: ║ https://mk1.ai ║
jic062-dpo-v1-3-nemo-c500-v2-mkmlizer: ║ ║
jic062-dpo-v1-3-nemo-c500-v2-mkmlizer: ║ The license key for the current software has been verified as ║
jic062-dpo-v1-3-nemo-c500-v2-mkmlizer: ║ belonging to: ║
jic062-dpo-v1-3-nemo-c500-v2-mkmlizer: ║ ║
jic062-dpo-v1-3-nemo-c500-v2-mkmlizer: ║ Chai Research Corp. ║
jic062-dpo-v1-3-nemo-c500-v2-mkmlizer: ║ Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f ║
jic062-dpo-v1-3-nemo-c500-v2-mkmlizer: ║ Expiration: 2024-10-15 23:59:59 ║
jic062-dpo-v1-3-nemo-c500-v2-mkmlizer: ║ ║
jic062-dpo-v1-3-nemo-c500-v2-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
jic062-dpo-v1-3-nemo-c500-v2-mkmlizer: Downloaded to shared memory in 45.297s
jic062-dpo-v1-3-nemo-c500-v2-mkmlizer: quantizing model to /dev/shm/model_cache, profile:s0, folder:/tmp/tmpxfp92i60, device:0
jic062-dpo-v1-3-nemo-c500-v2-mkmlizer: Saving flywheel model at /dev/shm/model_cache
jic062-dpo-v1-3-nemo-c500-v2-mkmlizer: quantized model in 36.320s
jic062-dpo-v1-3-nemo-c500-v2-mkmlizer: Processed model jic062/dpo-v1.3-Nemo-c500 in 81.617s
jic062-dpo-v1-3-nemo-c500-v2-mkmlizer: creating bucket guanaco-mkml-models
jic062-dpo-v1-3-nemo-c500-v2-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
jic062-dpo-v1-3-nemo-c500-v2-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/jic062-dpo-v1-3-nemo-c500-v2
jic062-dpo-v1-3-nemo-c500-v2-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/jic062-dpo-v1-3-nemo-c500-v2/config.json
jic062-dpo-v1-3-nemo-c500-v2-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/jic062-dpo-v1-3-nemo-c500-v2/special_tokens_map.json
jic062-dpo-v1-3-nemo-c500-v2-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/jic062-dpo-v1-3-nemo-c500-v2/tokenizer_config.json
jic062-dpo-v1-3-nemo-c500-v2-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/jic062-dpo-v1-3-nemo-c500-v2/tokenizer.json
jic062-dpo-v1-3-nemo-c500-v2-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/jic062-dpo-v1-3-nemo-c500-v2/flywheel_model.0.safetensors
jic062-dpo-v1-3-nemo-c500-v2-mkmlizer: Loading 0: 0%| | 0/363 [00:00<?, ?it/s] Loading 0: 1%|▏ | 5/363 [00:00<00:12, 29.80it/s] Loading 0: 4%|▎ | 13/363 [00:00<00:07, 48.10it/s] Loading 0: 5%|▌ | 19/363 [00:00<00:07, 43.84it/s] Loading 0: 7%|▋ | 24/363 [00:00<00:08, 41.59it/s] Loading 0: 9%|▊ | 31/363 [00:00<00:06, 48.77it/s] Loading 0: 10%|█ | 37/363 [00:00<00:07, 43.92it/s] Loading 0: 12%|█▏ | 42/363 [00:00<00:07, 42.07it/s] Loading 0: 13%|█▎ | 49/363 [00:01<00:06, 47.62it/s] Loading 0: 15%|█▍ | 54/363 [00:01<00:06, 46.84it/s] Loading 0: 17%|█▋ | 60/363 [00:01<00:06, 44.40it/s] Loading 0: 18%|█▊ | 65/363 [00:01<00:09, 30.81it/s] Loading 0: 19%|█▉ | 69/363 [00:01<00:09, 30.62it/s] Loading 0: 20%|██ | 73/363 [00:01<00:09, 30.81it/s] Loading 0: 22%|██▏ | 80/363 [00:02<00:07, 38.42it/s] Loading 0: 23%|██▎ | 85/363 [00:02<00:07, 38.95it/s] Loading 0: 25%|██▍ | 90/363 [00:02<00:06, 41.16it/s] Loading 0: 26%|██▌ | 95/363 [00:02<00:06, 41.84it/s] Loading 0: 28%|██▊ | 100/363 [00:02<00:07, 35.00it/s] Loading 0: 29%|██▉ | 107/363 [00:02<00:05, 42.76it/s] Loading 0: 31%|███ | 112/363 [00:02<00:05, 43.69it/s] Loading 0: 32%|███▏ | 117/363 [00:02<00:05, 41.10it/s] Loading 0: 34%|███▎ | 122/363 [00:03<00:05, 42.00it/s] Loading 0: 35%|███▍ | 127/363 [00:03<00:06, 34.21it/s] Loading 0: 37%|███▋ | 134/363 [00:03<00:05, 41.56it/s] Loading 0: 38%|███▊ | 139/363 [00:03<00:05, 40.34it/s] Loading 0: 40%|███▉ | 144/363 [00:03<00:08, 26.92it/s] Loading 0: 41%|████ | 149/363 [00:03<00:07, 29.43it/s] Loading 0: 43%|████▎ | 156/363 [00:04<00:05, 36.64it/s] Loading 0: 44%|████▍ | 161/363 [00:04<00:05, 36.60it/s] Loading 0: 46%|████▌ | 166/363 [00:04<00:05, 37.53it/s] Loading 0: 47%|████▋ | 171/363 [00:04<00:04, 40.12it/s] Loading 0: 48%|████▊ | 176/363 [00:04<00:05, 34.35it/s] Loading 0: 50%|█████ | 183/363 [00:04<00:04, 42.08it/s] Loading 0: 52%|█████▏ | 188/363 [00:04<00:04, 41.43it/s] Loading 0: 53%|█████▎ | 193/363 [00:04<00:04, 41.29it/s] Loading 0: 55%|█████▍ | 198/363 [00:05<00:03, 42.53it/s] Loading 0: 56%|█████▌ | 203/363 [00:05<00:04, 36.15it/s] Loading 0: 58%|█████▊ | 210/363 [00:05<00:03, 43.33it/s] Loading 0: 59%|█████▉ | 215/363 [00:05<00:03, 43.17it/s] Loading 0: 61%|██████ | 221/363 [00:05<00:03, 47.09it/s] Loading 0: 62%|██████▏ | 226/363 [00:05<00:04, 29.32it/s] Loading 0: 63%|██████▎ | 230/363 [00:06<00:04, 28.90it/s] Loading 0: 65%|██████▌ | 237/363 [00:06<00:03, 35.78it/s] Loading 0: 67%|██████▋ | 242/363 [00:06<00:03, 36.93it/s] Loading 0: 68%|██████▊ | 247/363 [00:06<00:03, 38.55it/s] Loading 0: 69%|██████▉ | 252/363 [00:06<00:02, 40.63it/s] Loading 0: 71%|███████ | 257/363 [00:06<00:03, 35.27it/s] Loading 0: 73%|███████▎ | 264/363 [00:06<00:02, 42.63it/s] Loading 0: 74%|███████▍ | 269/363 [00:06<00:02, 42.60it/s] Loading 0: 75%|███████▌ | 274/363 [00:07<00:02, 41.30it/s] Loading 0: 77%|███████▋ | 279/363 [00:07<00:01, 42.79it/s] Loading 0: 78%|███████▊ | 284/363 [00:07<00:02, 34.88it/s] Loading 0: 80%|████████ | 291/363 [00:07<00:01, 40.57it/s] Loading 0: 82%|████████▏ | 296/363 [00:07<00:01, 40.86it/s] Loading 0: 83%|████████▎ | 301/363 [00:07<00:01, 41.79it/s] Loading 0: 84%|████████▍ | 306/363 [00:14<00:23, 2.45it/s] Loading 0: 85%|████████▌ | 310/363 [00:14<00:16, 3.18it/s] Loading 0: 87%|████████▋ | 314/363 [00:14<00:11, 4.16it/s] Loading 0: 88%|████████▊ | 320/363 [00:14<00:06, 6.21it/s] Loading 0: 90%|████████▉ | 326/363 [00:15<00:04, 8.63it/s] Loading 0: 91%|█████████ | 330/363 [00:15<00:03, 10.37it/s] Loading 0: 93%|█████████▎| 338/363 [00:15<00:01, 15.95it/s] Loading 0: 95%|█████████▍| 344/363 [00:15<00:00, 19.67it/s] Loading 0: 96%|█████████▌| 349/363 [00:15<00:00, 22.82it/s] Loading 0: 98%|█████████▊| 356/363 [00:15<00:00, 29.33it/s] Loading 0: 99%|█████████▉| 361/363 [00:15<00:00, 32.47it/s]
Job jic062-dpo-v1-3-nemo-c500-v2-mkmlizer completed after 106.13s with status: succeeded
Stopping job with name jic062-dpo-v1-3-nemo-c500-v2-mkmlizer
Pipeline stage MKMLizer completed in 106.95s
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 jic062-dpo-v1-3-nemo-c500-v2
Waiting for inference service jic062-dpo-v1-3-nemo-c500-v2 to be ready
Inference service jic062-dpo-v1-3-nemo-c500-v2 ready after 171.20384287834167s
Pipeline stage MKMLDeployer completed in 171.69s
run pipeline stage %s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.1533138751983643s
Received healthy response to inference request in 2.189089059829712s
Received healthy response to inference request in 12.591163873672485s
Received healthy response to inference request in 2.2070231437683105s
Received healthy response to inference request in 3.075990676879883s
5 requests
0 failed requests
5th percentile: 2.160468912124634
10th percentile: 2.1676239490509035
20th percentile: 2.1819340229034423
30th percentile: 2.1926758766174315
40th percentile: 2.199849510192871
50th percentile: 2.2070231437683105
60th percentile: 2.5546101570129394
70th percentile: 2.902197170257568
80th percentile: 4.979025316238405
90th percentile: 8.785094594955446
95th percentile: 10.688129234313964
99th percentile: 12.21055694580078
mean time: 4.443316125869751
Pipeline stage StressChecker completed in 23.68s
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 3.90s
Shutdown handler de-registered
jic062-dpo-v1-3-nemo-c500_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 0.11s
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-v1-3-nemo-c500-v2-profiler
Waiting for inference service jic062-dpo-v1-3-nemo-c500-v2-profiler to be ready
Inference service jic062-dpo-v1-3-nemo-c500-v2-profiler ready after 170.42005515098572s
Pipeline stage MKMLProfilerDeployer completed in 170.78s
run pipeline stage %s
Running pipeline stage MKMLProfilerRunner
kubectl cp /code/guanaco/guanaco_inference_services/src/inference_scripts tenant-chaiml-guanaco/jic062-dpo-v1-3-nemo151305a29c77044faed772f6e5efee83-deplov8bm4:/code/chaiverse_profiler_1726116552 --namespace tenant-chaiml-guanaco
kubectl exec -it jic062-dpo-v1-3-nemo151305a29c77044faed772f6e5efee83-deplov8bm4 --namespace tenant-chaiml-guanaco -- sh -c 'cd /code/chaiverse_profiler_1726116552 && 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_1726116552/summary.json'
kubectl exec -it jic062-dpo-v1-3-nemo151305a29c77044faed772f6e5efee83-deplov8bm4 --namespace tenant-chaiml-guanaco -- bash -c 'cat /code/chaiverse_profiler_1726116552/summary.json'
Pipeline stage MKMLProfilerRunner completed in 945.64s
run pipeline stage %s
Running pipeline stage MKMLProfilerDeleter
Checking if service jic062-dpo-v1-3-nemo-c500-v2-profiler is running
Tearing down inference service jic062-dpo-v1-3-nemo-c500-v2-profiler
Service jic062-dpo-v1-3-nemo-c500-v2-profiler has been torndown
Pipeline stage MKMLProfilerDeleter completed in 1.95s
Shutdown handler de-registered
jic062-dpo-v1-3-nemo-c500_v2 status is now inactive due to auto deactivation removed underperforming models
jic062-dpo-v1-3-nemo-c500_v2 status is now torndown due to DeploymentManager action