developer_uid: chace9580
submission_id: jic062-dpo-v1-5-nemo_v1
model_name: jic062-dpo-v1-5-nemo_v1
model_group: jic062/dpo-v1.5-Nemo
status: torndown
timestamp: 2024-09-14T01:09:29+00:00
num_battles: 13468
num_wins: 6492
celo_rating: 1240.59
family_friendly_score: 0.0
submission_type: basic
model_repo: jic062/dpo-v1.5-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': 0.7021613784948677, 'latency_mean': 1.424095208644867, 'latency_p50': 1.422796368598938, 'latency_p90': 1.5962503910064698}, {'batch_size': 3, 'throughput': 1.3340855452826221, 'latency_mean': 2.2402485167980193, 'latency_p50': 2.232166886329651, 'latency_p90': 2.4988914489746095}, {'batch_size': 5, 'throughput': 1.593900492962745, 'latency_mean': 3.128035762310028, 'latency_p50': 3.103875756263733, 'latency_p90': 3.59266140460968}, {'batch_size': 6, 'throughput': 1.649176187481904, 'latency_mean': 3.623587077856064, 'latency_p50': 3.622355341911316, 'latency_p90': 4.108645820617676}, {'batch_size': 8, 'throughput': 1.6272299025810164, 'latency_mean': 4.876096196174622, 'latency_p50': 4.882611036300659, 'latency_p90': 5.573742389678955}, {'batch_size': 10, 'throughput': 1.5534349368121136, 'latency_mean': 6.384639489650726, 'latency_p50': 6.391626954078674, 'latency_p90': 7.2323837041854855}]
gpu_counts: {'NVIDIA RTX A5000': 1}
display_name: jic062-dpo-v1-5-nemo_v1
is_internal_developer: False
language_model: jic062/dpo-v1.5-Nemo
model_size: 13B
ranking_group: single
throughput_3p7s: 1.65
us_pacific_date: 2024-09-13
win_ratio: 0.48203148203148205
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-v1-mkmlizer
Waiting for job on jic062-dpo-v1-5-nemo-v1-mkmlizer to finish
jic062-dpo-v1-5-nemo-v1-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
jic062-dpo-v1-5-nemo-v1-mkmlizer: ║ _____ __ __ ║
jic062-dpo-v1-5-nemo-v1-mkmlizer: ║ / _/ /_ ___ __/ / ___ ___ / / ║
jic062-dpo-v1-5-nemo-v1-mkmlizer: ║ / _/ / // / |/|/ / _ \/ -_) -_) / ║
jic062-dpo-v1-5-nemo-v1-mkmlizer: ║ /_//_/\_, /|__,__/_//_/\__/\__/_/ ║
jic062-dpo-v1-5-nemo-v1-mkmlizer: ║ /___/ ║
jic062-dpo-v1-5-nemo-v1-mkmlizer: ║ ║
jic062-dpo-v1-5-nemo-v1-mkmlizer: ║ Version: 0.10.1 ║
jic062-dpo-v1-5-nemo-v1-mkmlizer: ║ Copyright 2023 MK ONE TECHNOLOGIES Inc. ║
jic062-dpo-v1-5-nemo-v1-mkmlizer: ║ https://mk1.ai ║
jic062-dpo-v1-5-nemo-v1-mkmlizer: ║ ║
jic062-dpo-v1-5-nemo-v1-mkmlizer: ║ The license key for the current software has been verified as ║
jic062-dpo-v1-5-nemo-v1-mkmlizer: ║ belonging to: ║
jic062-dpo-v1-5-nemo-v1-mkmlizer: ║ ║
jic062-dpo-v1-5-nemo-v1-mkmlizer: ║ Chai Research Corp. ║
jic062-dpo-v1-5-nemo-v1-mkmlizer: ║ Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f ║
jic062-dpo-v1-5-nemo-v1-mkmlizer: ║ Expiration: 2024-10-15 23:59:59 ║
jic062-dpo-v1-5-nemo-v1-mkmlizer: ║ ║
jic062-dpo-v1-5-nemo-v1-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', '')
Failed to get response for submission blend_tagim_2024-09-14: ('http://zonemercy-virgo-edit-v1-1e5-v3-predictor.tenant-chaiml-guanaco.k.chaiverse.com/v1/models/GPT-J-6B-lit-v2:predict', 'read tcp 127.0.0.1:59006->127.0.0.1:8080: read: connection reset by peer\n')
jic062-dpo-v1-5-nemo-v1-mkmlizer: Downloaded to shared memory in 53.750s
jic062-dpo-v1-5-nemo-v1-mkmlizer: quantizing model to /dev/shm/model_cache, profile:s0, folder:/tmp/tmp9oq3qb6_, device:0
jic062-dpo-v1-5-nemo-v1-mkmlizer: Saving flywheel model at /dev/shm/model_cache
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', '')
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-v1-mkmlizer: quantized model in 36.082s
jic062-dpo-v1-5-nemo-v1-mkmlizer: Processed model jic062/dpo-v1.5-Nemo in 89.832s
jic062-dpo-v1-5-nemo-v1-mkmlizer: creating bucket guanaco-mkml-models
jic062-dpo-v1-5-nemo-v1-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/jic062-dpo-v1-5-nemo-v1/flywheel_model.0.safetensors
jic062-dpo-v1-5-nemo-v1-mkmlizer: Loading 0: 0%| | 0/363 [00:00<?, ?it/s] Loading 0: 1%|▏ | 5/363 [00:00<00:11, 31.96it/s] Loading 0: 4%|▎ | 13/363 [00:00<00:06, 50.18it/s] Loading 0: 5%|▌ | 19/363 [00:00<00:07, 44.27it/s] Loading 0: 7%|▋ | 24/363 [00:00<00:08, 42.14it/s] Loading 0: 9%|▊ | 31/363 [00:00<00:06, 49.00it/s] Loading 0: 10%|█ | 37/363 [00:00<00:07, 45.42it/s] Loading 0: 12%|█▏ | 42/363 [00:00<00:07, 44.76it/s] Loading 0: 13%|█▎ | 49/363 [00:01<00:06, 50.74it/s] Loading 0: 15%|█▌ | 55/363 [00:01<00:06, 48.40it/s] Loading 0: 17%|█▋ | 60/363 [00:01<00:06, 47.88it/s] Loading 0: 18%|█▊ | 65/363 [00:01<00:10, 29.13it/s] Loading 0: 20%|█▉ | 72/363 [00:01<00:08, 35.96it/s] Loading 0: 21%|██▏ | 78/363 [00:01<00:07, 36.74it/s] Loading 0: 23%|██▎ | 83/363 [00:02<00:07, 36.24it/s] Loading 0: 25%|██▍ | 89/363 [00:02<00:06, 41.25it/s] Loading 0: 26%|██▌ | 94/363 [00:02<00:06, 41.58it/s] Loading 0: 27%|██▋ | 99/363 [00:02<00:06, 42.47it/s] Loading 0: 29%|██▊ | 104/363 [00:02<00:05, 43.52it/s] Loading 0: 30%|███ | 109/363 [00:02<00:05, 45.21it/s] Loading 0: 31%|███▏ | 114/363 [00:02<00:06, 39.63it/s] Loading 0: 33%|███▎ | 119/363 [00:02<00:06, 38.28it/s] Loading 0: 34%|███▍ | 125/363 [00:03<00:05, 41.91it/s] Loading 0: 36%|███▌ | 130/363 [00:03<00:05, 42.41it/s] Loading 0: 37%|███▋ | 135/363 [00:03<00:05, 42.54it/s] Loading 0: 39%|███▉ | 141/363 [00:03<00:05, 42.44it/s] Loading 0: 40%|████ | 146/363 [00:03<00:07, 30.50it/s] Loading 0: 41%|████▏ | 150/363 [00:03<00:06, 30.60it/s] Loading 0: 43%|████▎ | 157/363 [00:03<00:05, 35.16it/s] Loading 0: 45%|████▍ | 163/363 [00:04<00:05, 36.97it/s] Loading 0: 46%|████▌ | 167/363 [00:04<00:05, 35.40it/s] Loading 0: 48%|████▊ | 175/363 [00:04<00:04, 43.78it/s] Loading 0: 50%|████▉ | 181/363 [00:04<00:04, 42.63it/s] Loading 0: 51%|█████ | 186/363 [00:04<00:04, 41.08it/s] Loading 0: 53%|█████▎ | 192/363 [00:04<00:03, 44.05it/s] Loading 0: 54%|█████▍ | 197/363 [00:04<00:03, 43.62it/s] Loading 0: 56%|█████▌ | 202/363 [00:04<00:03, 44.42it/s] Loading 0: 57%|█████▋ | 207/363 [00:05<00:03, 44.48it/s] Loading 0: 58%|█████▊ | 212/363 [00:05<00:04, 37.50it/s] Loading 0: 61%|██████ | 220/363 [00:05<00:03, 47.22it/s] Loading 0: 62%|██████▏ | 226/363 [00:05<00:04, 31.94it/s] Loading 0: 64%|██████▎ | 231/363 [00:05<00:03, 33.38it/s] Loading 0: 65%|██████▌ | 237/363 [00:05<00:03, 37.35it/s] Loading 0: 67%|██████▋ | 242/363 [00:06<00:03, 39.05it/s] Loading 0: 68%|██████▊ | 247/363 [00:06<00:02, 41.02it/s] Loading 0: 69%|██████▉ | 252/363 [00:06<00:02, 43.15it/s] Loading 0: 71%|███████ | 257/363 [00:06<00:02, 37.31it/s] Loading 0: 73%|███████▎ | 265/363 [00:06<00:02, 45.44it/s] Loading 0: 74%|███████▍ | 270/363 [00:06<00:02, 45.73it/s] Loading 0: 76%|███████▌ | 275/363 [00:06<00:02, 36.86it/s] Loading 0: 78%|███████▊ | 283/363 [00:06<00:01, 44.70it/s] Loading 0: 79%|███████▉ | 288/363 [00:07<00:01, 45.14it/s] Loading 0: 81%|████████ | 293/363 [00:07<00:01, 37.53it/s] Loading 0: 82%|████████▏ | 299/363 [00:07<00:01, 40.48it/s] Loading 0: 84%|████████▎ | 304/363 [00:14<00:22, 2.57it/s] Loading 0: 85%|████████▍ | 308/363 [00:14<00:16, 3.30it/s] Loading 0: 86%|████████▌ | 312/363 [00:14<00:11, 4.29it/s] Loading 0: 88%|████████▊ | 320/363 [00:14<00:06, 7.11it/s] Loading 0: 90%|████████▉ | 326/363 [00:14<00:03, 9.63it/s] Loading 0: 91%|█████████ | 331/363 [00:14<00:02, 12.07it/s] Loading 0: 93%|█████████▎| 337/363 [00:14<00:01, 16.08it/s] Loading 0: 94%|█████████▍| 342/363 [00:15<00:01, 19.47it/s] Loading 0: 96%|█████████▌| 348/363 [00:15<00:00, 22.83it/s] Loading 0: 98%|█████████▊| 356/363 [00:15<00:00, 31.01it/s] Loading 0: 100%|█████████▉| 362/363 [00:15<00:00, 33.45it/s]
Job jic062-dpo-v1-5-nemo-v1-mkmlizer completed after 122.5s with status: succeeded
Stopping job with name jic062-dpo-v1-5-nemo-v1-mkmlizer
Pipeline stage MKMLizer completed in 123.41s
run pipeline stage %s
Running pipeline stage MKMLTemplater
Failed to get response for submission blend_siken_2024-09-09: ('http://zonemercy-virgo-edit-v1-1e5-v3-predictor.tenant-chaiml-guanaco.k.chaiverse.com/v1/models/GPT-J-6B-lit-v2:predict', 'pending request queue full\n')
Pipeline stage MKMLTemplater completed in 0.08s
run pipeline stage %s
Running pipeline stage MKMLDeployer
Creating inference service jic062-dpo-v1-5-nemo-v1
Waiting for inference service jic062-dpo-v1-5-nemo-v1 to be ready
Connection pool is full, discarding connection: %s. Connection pool size: %s
Inference service jic062-dpo-v1-5-nemo-v1 ready after 170.46356773376465s
Pipeline stage MKMLDeployer completed in 170.87s
run pipeline stage %s
Running pipeline stage StressChecker
Received healthy response to inference request in 3.580021858215332s
Received healthy response to inference request in 2.3386101722717285s
Received healthy response to inference request in 1.836899757385254s
Received healthy response to inference request in 1.930572748184204s
Received healthy response to inference request in 2.1017284393310547s
5 requests
0 failed requests
5th percentile: 1.8556343555450439
10th percentile: 1.874368953704834
20th percentile: 1.9118381500244142
30th percentile: 1.9648038864135742
40th percentile: 2.0332661628723145
50th percentile: 2.1017284393310547
60th percentile: 2.196481132507324
70th percentile: 2.2912338256835936
80th percentile: 2.5868925094604496
90th percentile: 3.083457183837891
95th percentile: 3.331739521026611
99th percentile: 3.530365390777588
mean time: 2.3575665950775146
Pipeline stage StressChecker completed in 13.52s
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 6.61s
Shutdown handler de-registered
jic062-dpo-v1-5-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 0.11s
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 jic062-dpo-v1-5-nemo-v1-profiler
Waiting for inference service jic062-dpo-v1-5-nemo-v1-profiler to be ready
Inference service jic062-dpo-v1-5-nemo-v1-profiler ready after 170.408784866333s
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-v1-5-nemo-v1-profiler-predictor-00001-deploymenxwlv2:/code/chaiverse_profiler_1726276703 --namespace tenant-chaiml-guanaco
kubectl exec -it jic062-dpo-v1-5-nemo-v1-profiler-predictor-00001-deploymenxwlv2 --namespace tenant-chaiml-guanaco -- sh -c 'cd /code/chaiverse_profiler_1726276703 && 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_1726276703/summary.json'
kubectl exec -it jic062-dpo-v1-5-nemo-v1-profiler-predictor-00001-deploymenxwlv2 --namespace tenant-chaiml-guanaco -- bash -c 'cat /code/chaiverse_profiler_1726276703/summary.json'
Pipeline stage MKMLProfilerRunner completed in 939.30s
run pipeline stage %s
Running pipeline stage MKMLProfilerDeleter
Checking if service jic062-dpo-v1-5-nemo-v1-profiler is running
Tearing down inference service jic062-dpo-v1-5-nemo-v1-profiler
Service jic062-dpo-v1-5-nemo-v1-profiler has been torndown
Pipeline stage MKMLProfilerDeleter completed in 2.17s
Shutdown handler de-registered
jic062-dpo-v1-5-nemo_v1 status is now inactive due to auto deactivation removed underperforming models
jic062-dpo-v1-5-nemo_v1 status is now torndown due to DeploymentManager action