jic062-dpo-v1-4-nemo

submission_id: jic062-dpo-v1-4-nemo_v4

developer_uid: chace9580

best_of: 8

celo_rating: 1268.57

display_name: jic062-dpo-v1-4-nemo_v4

family_friendly_score: 0.0

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}

generation_params: {'temperature': 1.0, 'top_p': 0.9, 'min_p': 0.05, 'top_k': 80, 'presence_penalty': 0.0, 'frequency_penalty': 0.0, 'stopping_words': ['\n', '/s', '[/INST]'], 'max_input_tokens': 1024, 'best_of': 8, 'max_output_tokens': 64}

gpu_counts: {'NVIDIA RTX A5000': 1}

is_internal_developer: False

language_model: jic062/dpo-v1.4-Nemo

latencies: [{'batch_size': 1, 'throughput': 0.6076971172649224, 'latency_mean': 1.6454906630516053, 'latency_p50': 1.6537479162216187, 'latency_p90': 1.810513710975647}, {'batch_size': 3, 'throughput': 1.0765278242689391, 'latency_mean': 2.78095029592514, 'latency_p50': 2.7915364503860474, 'latency_p90': 3.089661717414856}, {'batch_size': 5, 'throughput': 1.2256795795835675, 'latency_mean': 4.065650686025619, 'latency_p50': 4.086111307144165, 'latency_p90': 4.584303140640259}, {'batch_size': 6, 'throughput': 1.2565417426907575, 'latency_mean': 4.758843611478806, 'latency_p50': 4.770857334136963, 'latency_p90': 5.35527925491333}, {'batch_size': 8, 'throughput': 1.2383655478772058, 'latency_mean': 6.4213353645801545, 'latency_p50': 6.421629428863525, 'latency_p90': 7.378744006156921}, {'batch_size': 10, 'throughput': 1.2065924844526084, 'latency_mean': 8.247651942968368, 'latency_p50': 8.278855800628662, 'latency_p90': 9.38875162601471}]

max_input_tokens: 1024

max_output_tokens: 64

model_architecture: MistralForCausalLM

model_group: jic062/dpo-v1.4-Nemo

model_name: jic062-dpo-v1-4-nemo_v4

model_num_parameters: 12772070400.0

model_repo: jic062/dpo-v1.4-Nemo

model_size: 13B

num_battles: 47185

num_wins: 25250

ranking_group: single

status: torndown

submission_type: basic

throughput_3p7s: 1.2

timestamp: 2024-09-23T23:37:56+00:00

us_pacific_date: 2024-09-23

win_ratio: 0.5351276888841793

Download Preference Data

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-v1-4-nemo-v4-mkmlizer
Waiting for job on jic062-dpo-v1-4-nemo-v4-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
jic062-dpo-v1-4-nemo-v4-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
jic062-dpo-v1-4-nemo-v4-mkmlizer: ║     _____            __           __                                ║
jic062-dpo-v1-4-nemo-v4-mkmlizer: ║    / _/ /_ ___    __/ /  ___ ___ / /                                ║
jic062-dpo-v1-4-nemo-v4-mkmlizer: ║   / _/ / // / |/|/ / _ \/ -_) -_) /                                 ║
jic062-dpo-v1-4-nemo-v4-mkmlizer: ║  /_//_/\_, /|__,__/_//_/\__/\__/_/                                  ║
jic062-dpo-v1-4-nemo-v4-mkmlizer: ║       /___/                                                         ║
jic062-dpo-v1-4-nemo-v4-mkmlizer: ║                                                                     ║
jic062-dpo-v1-4-nemo-v4-mkmlizer: ║  Version: 0.10.1                                                    ║
jic062-dpo-v1-4-nemo-v4-mkmlizer: ║  Copyright 2023 MK ONE TECHNOLOGIES Inc.                            ║
jic062-dpo-v1-4-nemo-v4-mkmlizer: ║  https://mk1.ai                                                     ║
jic062-dpo-v1-4-nemo-v4-mkmlizer: ║                                                                     ║
jic062-dpo-v1-4-nemo-v4-mkmlizer: ║  The license key for the current software has been verified as      ║
jic062-dpo-v1-4-nemo-v4-mkmlizer: ║  belonging to:                                                      ║
jic062-dpo-v1-4-nemo-v4-mkmlizer: ║                                                                     ║
jic062-dpo-v1-4-nemo-v4-mkmlizer: ║  Chai Research Corp.                                                ║
jic062-dpo-v1-4-nemo-v4-mkmlizer: ║  Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f                   ║
jic062-dpo-v1-4-nemo-v4-mkmlizer: ║  Expiration: 2024-10-15 23:59:59                                    ║
jic062-dpo-v1-4-nemo-v4-mkmlizer: ║                                                                     ║
jic062-dpo-v1-4-nemo-v4-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
jic062-dpo-v1-4-nemo-v4-mkmlizer: Downloaded to shared memory in 48.226s
jic062-dpo-v1-4-nemo-v4-mkmlizer: quantizing model to /dev/shm/model_cache, profile:s0, folder:/tmp/tmp5lz90pwm, device:0
jic062-dpo-v1-4-nemo-v4-mkmlizer: Saving flywheel model at /dev/shm/model_cache
jic062-dpo-v1-4-nemo-v4-mkmlizer: quantized model in 36.239s
jic062-dpo-v1-4-nemo-v4-mkmlizer: Processed model jic062/dpo-v1.4-Nemo in 84.465s
jic062-dpo-v1-4-nemo-v4-mkmlizer: creating bucket guanaco-mkml-models
jic062-dpo-v1-4-nemo-v4-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
jic062-dpo-v1-4-nemo-v4-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/jic062-dpo-v1-4-nemo-v4
jic062-dpo-v1-4-nemo-v4-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/jic062-dpo-v1-4-nemo-v4/special_tokens_map.json
jic062-dpo-v1-4-nemo-v4-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/jic062-dpo-v1-4-nemo-v4/config.json
jic062-dpo-v1-4-nemo-v4-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/jic062-dpo-v1-4-nemo-v4/tokenizer_config.json
jic062-dpo-v1-4-nemo-v4-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/jic062-dpo-v1-4-nemo-v4/tokenizer.json
jic062-dpo-v1-4-nemo-v4-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/jic062-dpo-v1-4-nemo-v4/flywheel_model.0.safetensors
jic062-dpo-v1-4-nemo-v4-mkmlizer: 
Loading 0:   0%|          | 0/363 [00:00<?, ?it/s]
Loading 0:   1%|▏         | 5/363 [00:00<00:11, 30.51it/s]
Loading 0:   3%|▎         | 12/363 [00:00<00:07, 46.90it/s]
Loading 0:   5%|▍         | 18/363 [00:00<00:07, 46.12it/s]
Loading 0:   6%|▋         | 23/363 [00:00<00:09, 34.98it/s]
Loading 0:   8%|▊         | 30/363 [00:00<00:07, 43.18it/s]
Loading 0:  10%|▉         | 35/363 [00:00<00:07, 42.66it/s]
Loading 0:  11%|█         | 40/363 [00:00<00:07, 44.32it/s]
Loading 0:  13%|█▎        | 46/363 [00:01<00:07, 42.65it/s]
Loading 0:  14%|█▍        | 51/363 [00:01<00:07, 41.72it/s]
Loading 0:  16%|█▋        | 59/363 [00:01<00:05, 51.04it/s]
Loading 0:  18%|█▊        | 65/363 [00:01<00:09, 31.83it/s]
Loading 0:  20%|█▉        | 72/363 [00:01<00:07, 37.92it/s]
Loading 0:  21%|██        | 77/363 [00:01<00:07, 39.78it/s]
Loading 0:  23%|██▎       | 82/363 [00:02<00:08, 34.24it/s]
Loading 0:  25%|██▍       | 89/363 [00:02<00:06, 40.57it/s]
Loading 0:  26%|██▌       | 94/363 [00:02<00:06, 40.25it/s]
Loading 0:  27%|██▋       | 99/363 [00:02<00:06, 40.12it/s]
Loading 0:  29%|██▊       | 104/363 [00:02<00:06, 41.23it/s]
Loading 0:  30%|███       | 109/363 [00:02<00:05, 43.22it/s]
Loading 0:  31%|███▏      | 114/363 [00:02<00:06, 37.25it/s]
Loading 0:  33%|███▎      | 119/363 [00:03<00:06, 36.88it/s]
Loading 0:  34%|███▍      | 125/363 [00:03<00:05, 40.23it/s]
Loading 0:  36%|███▌      | 130/363 [00:03<00:05, 39.80it/s]
Loading 0:  37%|███▋      | 135/363 [00:03<00:05, 39.38it/s]
Loading 0:  39%|███▊      | 140/363 [00:03<00:05, 41.13it/s]
Loading 0:  40%|███▉      | 145/363 [00:03<00:08, 26.43it/s]
Loading 0:  41%|████      | 149/363 [00:03<00:08, 26.64it/s]
Loading 0:  43%|████▎     | 156/363 [00:04<00:06, 33.67it/s]
Loading 0:  44%|████▍     | 160/363 [00:04<00:06, 33.58it/s]
Loading 0:  45%|████▌     | 165/363 [00:04<00:05, 35.80it/s]
Loading 0:  47%|████▋     | 169/363 [00:04<00:05, 35.76it/s]
Loading 0:  48%|████▊     | 174/363 [00:04<00:04, 38.99it/s]
Loading 0:  49%|████▉     | 179/363 [00:04<00:04, 39.69it/s]
Loading 0:  51%|█████     | 184/363 [00:04<00:04, 40.58it/s]
Loading 0:  52%|█████▏    | 190/363 [00:04<00:04, 40.48it/s]
Loading 0:  54%|█████▎    | 195/363 [00:05<00:04, 39.58it/s]
Loading 0:  55%|█████▌    | 201/363 [00:05<00:03, 43.09it/s]
Loading 0:  57%|█████▋    | 206/363 [00:05<00:03, 42.47it/s]
Loading 0:  58%|█████▊    | 211/363 [00:05<00:03, 42.77it/s]
Loading 0:  60%|█████▉    | 216/363 [00:05<00:03, 43.99it/s]
Loading 0:  61%|██████    | 221/363 [00:05<00:03, 44.65it/s]
Loading 0:  62%|██████▏   | 226/363 [00:06<00:05, 26.48it/s]
Loading 0:  63%|██████▎   | 230/363 [00:06<00:04, 27.92it/s]
Loading 0:  66%|██████▌   | 238/363 [00:06<00:03, 37.12it/s]
Loading 0:  67%|██████▋   | 244/363 [00:06<00:03, 39.42it/s]
Loading 0:  69%|██████▊   | 249/363 [00:06<00:02, 40.40it/s]
Loading 0:  71%|███████   | 256/363 [00:06<00:02, 45.66it/s]
Loading 0:  72%|███████▏  | 262/363 [00:06<00:02, 46.31it/s]
Loading 0:  74%|███████▎  | 267/363 [00:06<00:02, 46.05it/s]
Loading 0:  75%|███████▌  | 274/363 [00:06<00:01, 49.88it/s]
Loading 0:  77%|███████▋  | 280/363 [00:07<00:01, 46.80it/s]
Loading 0:  79%|███████▊  | 285/363 [00:07<00:01, 44.27it/s]
Loading 0:  80%|████████  | 291/363 [00:07<00:01, 47.75it/s]
Loading 0:  82%|████████▏ | 296/363 [00:07<00:01, 46.75it/s]
Loading 0:  83%|████████▎ | 303/363 [00:07<00:01, 47.06it/s]
Loading 0:  85%|████████▍ | 308/363 [00:14<00:20,  2.65it/s]
Loading 0:  86%|████████▌ | 312/363 [00:14<00:15,  3.36it/s]
Loading 0:  88%|████████▊ | 320/363 [00:14<00:07,  5.40it/s]
Loading 0:  90%|████████▉ | 325/363 [00:14<00:05,  7.04it/s]
Loading 0:  91%|█████████ | 330/363 [00:15<00:03,  8.84it/s]
Loading 0:  93%|█████████▎| 338/363 [00:15<00:01, 13.32it/s]
Loading 0:  94%|█████████▍| 343/363 [00:15<00:01, 16.37it/s]
Loading 0:  96%|█████████▌| 348/363 [00:15<00:00, 18.25it/s]
Loading 0:  98%|█████████▊| 355/363 [00:15<00:00, 24.55it/s]
Loading 0:  99%|█████████▉| 361/363 [00:15<00:00, 28.76it/s]
                                                            
Job jic062-dpo-v1-4-nemo-v4-mkmlizer completed after 107.01s with status: succeeded
Stopping job with name jic062-dpo-v1-4-nemo-v4-mkmlizer
Pipeline stage MKMLizer completed in 109.26s
run pipeline stage %s
Running pipeline stage MKMLTemplater
Pipeline stage MKMLTemplater completed in 0.11s
run pipeline stage %s
Running pipeline stage MKMLDeployer
Creating inference service jic062-dpo-v1-4-nemo-v4
Waiting for inference service jic062-dpo-v1-4-nemo-v4 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
Inference service jic062-dpo-v1-4-nemo-v4 ready after 201.86853456497192s
Pipeline stage MKMLDeployer completed in 202.23s
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
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 2.689032554626465s
Received healthy response to inference request in 2.367047071456909s
Received healthy response to inference request in 1.9370012283325195s
Received healthy response to inference request in 2.270519495010376s
Received healthy response to inference request in 2.5655884742736816s
5 requests
0 failed requests
5th percentile: 2.003704881668091
10th percentile: 2.0704085350036623
20th percentile: 2.2038158416748046
30th percentile: 2.2898250102996824
40th percentile: 2.328436040878296
50th percentile: 2.367047071456909
60th percentile: 2.446463632583618
70th percentile: 2.5258801937103272
80th percentile: 2.5902772903442384
90th percentile: 2.6396549224853514
95th percentile: 2.664343738555908
99th percentile: 2.6840947914123534
mean time: 2.3658377647399904
Pipeline stage StressChecker completed in 13.12s
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 6.30s
Shutdown handler de-registered
jic062-dpo-v1-4-nemo_v4 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-v1-4-nemo-v4-profiler
Waiting for inference service jic062-dpo-v1-4-nemo-v4-profiler to be ready
Inference service jic062-dpo-v1-4-nemo-v4-profiler ready after 190.42248034477234s
Pipeline stage MKMLProfilerDeployer completed in 190.81s
run pipeline stage %s
Running pipeline stage MKMLProfilerRunner
kubectl cp /code/guanaco/guanaco_inference_services/src/inference_scripts tenant-chaiml-guanaco/jic062-dpo-v1-4-nemo-v4-profiler-predictor-00001-deploymenmr78j:/code/chaiverse_profiler_1727135245 --namespace tenant-chaiml-guanaco
kubectl exec -it jic062-dpo-v1-4-nemo-v4-profiler-predictor-00001-deploymenmr78j --namespace tenant-chaiml-guanaco -- sh -c 'cd /code/chaiverse_profiler_1727135245 && 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 1024 --output_tokens 64 --summary /code/chaiverse_profiler_1727135245/summary.json'
kubectl exec -it jic062-dpo-v1-4-nemo-v4-profiler-predictor-00001-deploymenmr78j --namespace tenant-chaiml-guanaco -- bash -c 'cat /code/chaiverse_profiler_1727135245/summary.json'
Pipeline stage MKMLProfilerRunner completed in 1170.00s
run pipeline stage %s
Running pipeline stage MKMLProfilerDeleter
Checking if service jic062-dpo-v1-4-nemo-v4-profiler is running
Tearing down inference service jic062-dpo-v1-4-nemo-v4-profiler
Service jic062-dpo-v1-4-nemo-v4-profiler has been torndown
Pipeline stage MKMLProfilerDeleter completed in 2.06s
Shutdown handler de-registered
jic062-dpo-v1-4-nemo_v4 status is now inactive due to auto deactivation removed underperforming models
jic062-nemo-v1-6_v3 status is now torndown due to DeploymentManager action
jic062-dpo-v1-4-nemo_v4 status is now torndown due to DeploymentManager action