rica40325-feedback-dpo-5

developer_uid: rica40325

submission_id: rica40325-feedback-dpo-5_v1

model_name: rica40325-feedback-dpo-5_v1

model_group: rica40325/feedback_dpo_5

status: torndown

timestamp: 2024-09-10T08:11:10+00:00

num_battles: 10731

num_wins: 1301

celo_rating: 855.66

family_friendly_score: 0.0

submission_type: basic

model_repo: rica40325/feedback_dpo_5

model_architecture: LlamaForCausalLM

model_num_parameters: 8030261248.0

best_of: 16

max_input_tokens: 512

max_output_tokens: 64

latencies: [{'batch_size': 1, 'throughput': 0.9127839132324801, 'latency_mean': 1.0954858672618866, 'latency_p50': 1.077151894569397, 'latency_p90': 1.2276214122772215}, {'batch_size': 4, 'throughput': 1.7892875813437075, 'latency_mean': 2.223102501630783, 'latency_p50': 2.2385947704315186, 'latency_p90': 2.4684191226959227}, {'batch_size': 5, 'throughput': 1.878947707680383, 'latency_mean': 2.6451274442672728, 'latency_p50': 2.6681915521621704, 'latency_p90': 2.9339876890182497}, {'batch_size': 8, 'throughput': 2.018782130735461, 'latency_mean': 3.932997866868973, 'latency_p50': 3.936123013496399, 'latency_p90': 4.4333083152771}, {'batch_size': 10, 'throughput': 2.0285419100298197, 'latency_mean': 4.887922847270966, 'latency_p50': 4.893164038658142, 'latency_p90': 5.559455895423889}, {'batch_size': 12, 'throughput': 2.0304084539112, 'latency_mean': 5.827128194570541, 'latency_p50': 5.835750937461853, 'latency_p90': 6.671876978874207}, {'batch_size': 15, 'throughput': 2.0155189477004973, 'latency_mean': 7.294881138801575, 'latency_p50': 7.363747954368591, 'latency_p90': 8.094710993766785}]

gpu_counts: {'NVIDIA RTX A5000': 1}

display_name: rica40325-feedback-dpo-5_v1

is_internal_developer: False

language_model: rica40325/feedback_dpo_5

model_size: 8B

ranking_group: single

throughput_3p7s: 2.02

us_pacific_date: 2024-09-10

win_ratio: 0.12123753611033454

generation_params: {'temperature': 1.0, 'top_p': 1.0, 'min_p': 0.0, 'top_k': 40, 'presence_penalty': 0.0, 'frequency_penalty': 0.0, 'stopping_words': ['\n'], 'max_input_tokens': 512, 'best_of': 16, 'max_output_tokens': 64}

formatter: {'memory_template': "{bot_name}'s Persona: {memory}\n####\n", 'prompt_template': '{prompt}\n<START>\n', 'bot_template': '{bot_name}: {message}\n', 'user_template': '{user_name}: {message}\n', 'response_template': '{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 rica40325-feedback-dpo-5-v1-mkmlizer
Waiting for job on rica40325-feedback-dpo-5-v1-mkmlizer to finish
rica40325-feedback-dpo-5-v1-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
rica40325-feedback-dpo-5-v1-mkmlizer: ║     _____            __           __                                ║
rica40325-feedback-dpo-5-v1-mkmlizer: ║    / _/ /_ ___    __/ /  ___ ___ / /                                ║
rica40325-feedback-dpo-5-v1-mkmlizer: ║   / _/ / // / |/|/ / _ \/ -_) -_) /                                 ║
rica40325-feedback-dpo-5-v1-mkmlizer: ║  /_//_/\_, /|__,__/_//_/\__/\__/_/                                  ║
rica40325-feedback-dpo-5-v1-mkmlizer: ║       /___/                                                         ║
rica40325-feedback-dpo-5-v1-mkmlizer: ║                                                                     ║
rica40325-feedback-dpo-5-v1-mkmlizer: ║  Version: 0.10.1                                                    ║
rica40325-feedback-dpo-5-v1-mkmlizer: ║  Copyright 2023 MK ONE TECHNOLOGIES Inc.                            ║
rica40325-feedback-dpo-5-v1-mkmlizer: ║  https://mk1.ai                                                     ║
rica40325-feedback-dpo-5-v1-mkmlizer: ║                                                                     ║
rica40325-feedback-dpo-5-v1-mkmlizer: ║  The license key for the current software has been verified as      ║
rica40325-feedback-dpo-5-v1-mkmlizer: ║  belonging to:                                                      ║
rica40325-feedback-dpo-5-v1-mkmlizer: ║                                                                     ║
rica40325-feedback-dpo-5-v1-mkmlizer: ║  Chai Research Corp.                                                ║
rica40325-feedback-dpo-5-v1-mkmlizer: ║  Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f                   ║
rica40325-feedback-dpo-5-v1-mkmlizer: ║  Expiration: 2024-10-15 23:59:59                                    ║
rica40325-feedback-dpo-5-v1-mkmlizer: ║                                                                     ║
rica40325-feedback-dpo-5-v1-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
rica40325-feedback-dpo-5-v1-mkmlizer: Downloaded to shared memory in 62.775s
rica40325-feedback-dpo-5-v1-mkmlizer: quantizing model to /dev/shm/model_cache, profile:s0, folder:/tmp/tmpbqd80ftk, device:0
rica40325-feedback-dpo-5-v1-mkmlizer: Saving flywheel model at /dev/shm/model_cache
rica40325-feedback-dpo-5-v1-mkmlizer: quantized model in 29.185s
rica40325-feedback-dpo-5-v1-mkmlizer: Processed model rica40325/feedback_dpo_5 in 91.960s
rica40325-feedback-dpo-5-v1-mkmlizer: creating bucket guanaco-mkml-models
rica40325-feedback-dpo-5-v1-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
rica40325-feedback-dpo-5-v1-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/rica40325-feedback-dpo-5-v1
rica40325-feedback-dpo-5-v1-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/rica40325-feedback-dpo-5-v1/config.json
rica40325-feedback-dpo-5-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/rica40325-feedback-dpo-5-v1/tokenizer_config.json
rica40325-feedback-dpo-5-v1-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/rica40325-feedback-dpo-5-v1/special_tokens_map.json
rica40325-feedback-dpo-5-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/rica40325-feedback-dpo-5-v1/tokenizer.json
rica40325-feedback-dpo-5-v1-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/rica40325-feedback-dpo-5-v1/flywheel_model.0.safetensors
rica40325-feedback-dpo-5-v1-mkmlizer: 
Loading 0:   0%|          | 0/291 [00:00<?, ?it/s]
Loading 0:   2%|▏         | 5/291 [00:00<00:11, 25.03it/s]
Loading 0:   3%|▎         | 10/291 [00:00<00:07, 35.32it/s]
Loading 0:   5%|▍         | 14/291 [00:00<00:10, 26.20it/s]
Loading 0:   7%|▋         | 21/291 [00:00<00:07, 36.42it/s]
Loading 0:   9%|▉         | 26/291 [00:00<00:07, 35.31it/s]
Loading 0:  11%|█         | 32/291 [00:00<00:06, 38.20it/s]
Loading 0:  13%|█▎        | 37/291 [00:01<00:09, 26.49it/s]
Loading 0:  14%|█▍        | 41/291 [00:01<00:09, 25.60it/s]
Loading 0:  16%|█▋        | 48/291 [00:01<00:07, 32.41it/s]
Loading 0:  18%|█▊        | 52/291 [00:01<00:07, 31.66it/s]
Loading 0:  20%|█▉        | 57/291 [00:01<00:06, 34.16it/s]
Loading 0:  21%|██        | 61/291 [00:01<00:07, 32.68it/s]
Loading 0:  23%|██▎       | 66/291 [00:02<00:06, 34.23it/s]
Loading 0:  24%|██▍       | 70/291 [00:02<00:06, 32.77it/s]
Loading 0:  25%|██▌       | 74/291 [00:02<00:06, 32.35it/s]
Loading 0:  27%|██▋       | 78/291 [00:02<00:06, 32.34it/s]
Loading 0:  28%|██▊       | 82/291 [00:02<00:09, 22.62it/s]
Loading 0:  29%|██▉       | 85/291 [00:02<00:08, 23.45it/s]
Loading 0:  31%|███       | 90/291 [00:03<00:07, 27.04it/s]
Loading 0:  32%|███▏      | 93/291 [00:03<00:07, 26.16it/s]
Loading 0:  34%|███▍      | 99/291 [00:03<00:06, 31.82it/s]
Loading 0:  35%|███▌      | 103/291 [00:03<00:06, 30.76it/s]
Loading 0:  37%|███▋      | 108/291 [00:03<00:05, 33.55it/s]
Loading 0:  38%|███▊      | 112/291 [00:03<00:05, 32.24it/s]
Loading 0:  40%|███▉      | 116/291 [00:03<00:05, 32.75it/s]
Loading 0:  42%|████▏     | 122/291 [00:03<00:04, 37.04it/s]
Loading 0:  44%|████▎     | 127/291 [00:04<00:04, 34.34it/s]
Loading 0:  45%|████▌     | 132/291 [00:04<00:04, 37.83it/s]
Loading 0:  47%|████▋     | 136/291 [00:04<00:06, 24.77it/s]
Loading 0:  48%|████▊     | 140/291 [00:04<00:06, 24.40it/s]
Loading 0:  51%|█████     | 147/291 [00:04<00:04, 30.87it/s]
Loading 0:  52%|█████▏    | 151/291 [00:04<00:04, 30.18it/s]
Loading 0:  54%|█████▎    | 156/291 [00:05<00:04, 33.21it/s]
Loading 0:  55%|█████▍    | 160/291 [00:05<00:04, 31.69it/s]
Loading 0:  57%|█████▋    | 165/291 [00:05<00:03, 34.21it/s]
Loading 0:  58%|█████▊    | 169/291 [00:05<00:03, 32.18it/s]
Loading 0:  60%|█████▉    | 174/291 [00:05<00:03, 34.56it/s]
Loading 0:  61%|██████    | 178/291 [00:05<00:03, 33.21it/s]
Loading 0:  63%|██████▎   | 184/291 [00:05<00:02, 39.25it/s]
Loading 0:  65%|██████▍   | 189/291 [00:06<00:04, 24.72it/s]
Loading 0:  67%|██████▋   | 194/291 [00:06<00:03, 25.08it/s]
Loading 0:  69%|██████▉   | 201/291 [00:06<00:02, 31.08it/s]
Loading 0:  70%|███████   | 205/291 [00:06<00:02, 29.60it/s]
Loading 0:  72%|███████▏  | 210/291 [00:06<00:02, 31.35it/s]
Loading 0:  74%|███████▎  | 214/291 [00:06<00:02, 31.09it/s]
Loading 0:  75%|███████▌  | 219/291 [00:07<00:02, 34.28it/s]
Loading 0:  77%|███████▋  | 223/291 [00:07<00:02, 32.38it/s]
Loading 0:  78%|███████▊  | 227/291 [00:07<00:02, 31.67it/s]
Loading 0:  79%|███████▉  | 231/291 [00:07<00:01, 31.60it/s]
Loading 0:  81%|████████  | 235/291 [00:07<00:02, 24.80it/s]
Loading 0:  82%|████████▏ | 239/291 [00:07<00:02, 24.85it/s]
Loading 0:  85%|████████▍ | 246/291 [00:07<00:01, 32.96it/s]
Loading 0:  86%|████████▌ | 250/291 [00:08<00:01, 32.73it/s]
Loading 0:  88%|████████▊ | 255/291 [00:08<00:01, 34.20it/s]
Loading 0:  89%|████████▉ | 259/291 [00:08<00:00, 33.10it/s]
Loading 0:  91%|█████████ | 264/291 [00:08<00:00, 35.67it/s]
Loading 0:  92%|█████████▏| 268/291 [00:08<00:00, 32.89it/s]
Loading 0:  94%|█████████▍| 273/291 [00:08<00:00, 35.16it/s]
Loading 0:  95%|█████████▌| 277/291 [00:08<00:00, 33.09it/s]
Loading 0:  97%|█████████▋| 281/291 [00:09<00:00, 32.67it/s]
Loading 0:  98%|█████████▊| 285/291 [00:09<00:00, 34.15it/s]
Loading 0:  99%|█████████▉| 289/291 [00:14<00:00,  2.36it/s]
                                                            
Job rica40325-feedback-dpo-5-v1-mkmlizer completed after 124.57s with status: succeeded
Stopping job with name rica40325-feedback-dpo-5-v1-mkmlizer
Pipeline stage MKMLizer completed in 127.46s
run pipeline stage %s
Running pipeline stage MKMLTemplater
Pipeline stage MKMLTemplater completed in 0.23s
run pipeline stage %s
Running pipeline stage MKMLDeployer
Creating inference service rica40325-feedback-dpo-5-v1
Waiting for inference service rica40325-feedback-dpo-5-v1 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
Inference service rica40325-feedback-dpo-5-v1 ready after 151.23896598815918s
Pipeline stage MKMLDeployer completed in 152.91s
run pipeline stage %s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.0042707920074463s
Received healthy response to inference request in 1.7374019622802734s
Received healthy response to inference request in 1.8954074382781982s
Received healthy response to inference request in 1.9128336906433105s
Received healthy response to inference request in 1.0353453159332275s
5 requests
0 failed requests
5th percentile: 1.1757566452026367
10th percentile: 1.3161679744720458
20th percentile: 1.5969906330108643
30th percentile: 1.7690030574798583
40th percentile: 1.8322052478790283
50th percentile: 1.8954074382781982
60th percentile: 1.9023779392242433
70th percentile: 1.909348440170288
80th percentile: 1.9311211109161377
90th percentile: 1.967695951461792
95th percentile: 1.9859833717346191
99th percentile: 2.000613307952881
mean time: 1.7170518398284913
Pipeline stage StressChecker completed in 9.61s
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 6.59s
Shutdown handler de-registered
rica40325-feedback-dpo-5_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 0.10s
run pipeline stage %s
Running pipeline stage MKMLProfilerTemplater
Pipeline stage MKMLProfilerTemplater completed in 0.12s
run pipeline stage %s
Running pipeline stage MKMLProfilerDeployer
Creating inference service rica40325-feedback-dpo-5-v1-profiler
Waiting for inference service rica40325-feedback-dpo-5-v1-profiler to be ready
Inference service rica40325-feedback-dpo-5-v1-profiler ready after 245.14028358459473s
Pipeline stage MKMLProfilerDeployer completed in 245.51s
run pipeline stage %s
Running pipeline stage MKMLProfilerRunner
kubectl cp /code/guanaco/guanaco_inference_services/src/inference_scripts tenant-chaiml-guanaco/rica40325-feedback-dpo-5-v1-profiler-predictor-00001-deplobrdd4:/code/chaiverse_profiler_1725956450 --namespace tenant-chaiml-guanaco
kubectl exec -it rica40325-feedback-dpo-5-v1-profiler-predictor-00001-deplobrdd4 --namespace tenant-chaiml-guanaco -- sh -c 'cd /code/chaiverse_profiler_1725956450 && python profiles.py profile --best_of_n 16 --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_1725956450/summary.json'
kubectl exec -it rica40325-feedback-dpo-5-v1-profiler-predictor-00001-deplobrdd4 --namespace tenant-chaiml-guanaco -- bash -c 'cat /code/chaiverse_profiler_1725956450/summary.json'
Pipeline stage MKMLProfilerRunner completed in 838.67s
run pipeline stage %s
Running pipeline stage MKMLProfilerDeleter
Checking if service rica40325-feedback-dpo-5-v1-profiler is running
Tearing down inference service rica40325-feedback-dpo-5-v1-profiler
Service rica40325-feedback-dpo-5-v1-profiler has been torndown
Pipeline stage MKMLProfilerDeleter completed in 1.70s
Shutdown handler de-registered
rica40325-feedback-dpo-5_v1 status is now inactive due to auto deactivation removed underperforming models
rica40325-feedback-dpo-5_v1 status is now torndown due to DeploymentManager action