developer_uid: NischayDnk
submission_id: chaiml-qwen-dpo-exp1-da_53786_v4
model_name: chaiml-qwen-dpo-exp1-da_53786_v4
model_group: ChaiML/qwen-dpo-exp1-dat
status: torndown
timestamp: 2025-04-14T17:32:35+00:00
num_battles: 6785
num_wins: 3461
celo_rating: 1291.83
family_friendly_score: 0.5566
family_friendly_standard_error: 0.007025616556573523
submission_type: basic
model_repo: ChaiML/qwen-dpo-exp1-datav1noffsftfull
model_architecture: Qwen2ForCausalLM
model_num_parameters: 32763417600.0
best_of: 5
max_input_tokens: 768
max_output_tokens: 64
reward_model: default
latencies: [{'batch_size': 1, 'throughput': 0.2771584138002738, 'latency_mean': 3.6079518675804136, 'latency_p50': 3.6178643703460693, 'latency_p90': 4.012407326698304}, {'batch_size': 2, 'throughput': 0.4544110776810392, 'latency_mean': 4.397809205055236, 'latency_p50': 4.407438397407532, 'latency_p90': 4.893822240829468}, {'batch_size': 3, 'throughput': 0.5958662530223521, 'latency_mean': 5.013467469215393, 'latency_p50': 5.0211403369903564, 'latency_p90': 5.48575656414032}, {'batch_size': 5, 'throughput': 0.790682160737079, 'latency_mean': 6.286132301092148, 'latency_p50': 6.296321392059326, 'latency_p90': 6.996623611450195}]
gpu_counts: {'NVIDIA RTX A6000': 1}
display_name: chaiml-qwen-dpo-exp1-da_53786_v4
is_internal_developer: False
language_model: ChaiML/qwen-dpo-exp1-datav1noffsftfull
model_size: 33B
ranking_group: single
throughput_3p7s: 0.3
us_pacific_date: 2025-04-14
win_ratio: 0.5100957995578482
generation_params: {'temperature': 0.45, 'top_p': 0.95, 'min_p': 0.025, 'top_k': 60, 'presence_penalty': 0.4, 'frequency_penalty': 0.4, 'stopping_words': ['\n', '<|im_end|>', '<|im_start|>'], 'max_input_tokens': 768, 'best_of': 5, 'max_output_tokens': 64}
formatter: {'memory_template': '<|system|>Family Friendly{memory}\n', 'prompt_template': '', 'bot_template': '<|im_start|>assistant\n{message}<|im_end|>\n', 'user_template': '<|im_start|>user\nYou:{message}<|im_end|>\n', 'response_template': '<|im_start|>assistant\n', '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 chaiml-qwen-dpo-exp1-da-53786-v4-mkmlizer
Waiting for job on chaiml-qwen-dpo-exp1-da-53786-v4-mkmlizer to finish
chaiml-qwen-dpo-exp1-da-53786-v4-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
chaiml-qwen-dpo-exp1-da-53786-v4-mkmlizer: ║ _____ __ __ ║
chaiml-qwen-dpo-exp1-da-53786-v4-mkmlizer: ║ / _/ /_ ___ __/ / ___ ___ / / ║
chaiml-qwen-dpo-exp1-da-53786-v4-mkmlizer: ║ / _/ / // / |/|/ / _ \/ -_) -_) / ║
chaiml-qwen-dpo-exp1-da-53786-v4-mkmlizer: ║ /_//_/\_, /|__,__/_//_/\__/\__/_/ ║
chaiml-qwen-dpo-exp1-da-53786-v4-mkmlizer: ║ /___/ ║
chaiml-qwen-dpo-exp1-da-53786-v4-mkmlizer: ║ ║
chaiml-qwen-dpo-exp1-da-53786-v4-mkmlizer: ║ Version: 0.12.8 ║
chaiml-qwen-dpo-exp1-da-53786-v4-mkmlizer: ║ Copyright 2023 MK ONE TECHNOLOGIES Inc. ║
chaiml-qwen-dpo-exp1-da-53786-v4-mkmlizer: ║ https://mk1.ai ║
chaiml-qwen-dpo-exp1-da-53786-v4-mkmlizer: ║ ║
chaiml-qwen-dpo-exp1-da-53786-v4-mkmlizer: ║ The license key for the current software has been verified as ║
chaiml-qwen-dpo-exp1-da-53786-v4-mkmlizer: ║ belonging to: ║
chaiml-qwen-dpo-exp1-da-53786-v4-mkmlizer: ║ ║
chaiml-qwen-dpo-exp1-da-53786-v4-mkmlizer: ║ Chai Research Corp. ║
chaiml-qwen-dpo-exp1-da-53786-v4-mkmlizer: ║ Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f ║
chaiml-qwen-dpo-exp1-da-53786-v4-mkmlizer: ║ Expiration: 2025-04-15 23:59:59 ║
chaiml-qwen-dpo-exp1-da-53786-v4-mkmlizer: ║ ║
chaiml-qwen-dpo-exp1-da-53786-v4-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
chaiml-qwen-dpo-exp1-da-53786-v4-mkmlizer: Downloaded to shared memory in 91.451s
chaiml-qwen-dpo-exp1-da-53786-v4-mkmlizer: quantizing model to /dev/shm/model_cache, profile:s0, folder:/tmp/tmp6izxkufh, device:0
chaiml-qwen-dpo-exp1-da-53786-v4-mkmlizer: Saving flywheel model at /dev/shm/model_cache
Unable to record family friendly update due to error: Invalid JSON input: JSON must contain 'User Safety' and 'Response Safety' fields
chaiml-qwen-dpo-exp1-da-53786-v4-mkmlizer: quantized model in 64.116s
chaiml-qwen-dpo-exp1-da-53786-v4-mkmlizer: Processed model ChaiML/qwen-dpo-exp1-datav1noffsftfull in 155.568s
chaiml-qwen-dpo-exp1-da-53786-v4-mkmlizer: creating bucket guanaco-mkml-models
chaiml-qwen-dpo-exp1-da-53786-v4-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
chaiml-qwen-dpo-exp1-da-53786-v4-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/chaiml-qwen-dpo-exp1-da-53786-v4
chaiml-qwen-dpo-exp1-da-53786-v4-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/chaiml-qwen-dpo-exp1-da-53786-v4/config.json
chaiml-qwen-dpo-exp1-da-53786-v4-mkmlizer: cp /dev/shm/model_cache/added_tokens.json s3://guanaco-mkml-models/chaiml-qwen-dpo-exp1-da-53786-v4/added_tokens.json
chaiml-qwen-dpo-exp1-da-53786-v4-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/chaiml-qwen-dpo-exp1-da-53786-v4/special_tokens_map.json
chaiml-qwen-dpo-exp1-da-53786-v4-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/chaiml-qwen-dpo-exp1-da-53786-v4/tokenizer_config.json
chaiml-qwen-dpo-exp1-da-53786-v4-mkmlizer: cp /dev/shm/model_cache/merges.txt s3://guanaco-mkml-models/chaiml-qwen-dpo-exp1-da-53786-v4/merges.txt
chaiml-qwen-dpo-exp1-da-53786-v4-mkmlizer: cp /dev/shm/model_cache/vocab.json s3://guanaco-mkml-models/chaiml-qwen-dpo-exp1-da-53786-v4/vocab.json
chaiml-qwen-dpo-exp1-da-53786-v4-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/chaiml-qwen-dpo-exp1-da-53786-v4/tokenizer.json
chaiml-qwen-dpo-exp1-da-53786-v4-mkmlizer: cp /dev/shm/model_cache/flywheel_model.2.safetensors s3://guanaco-mkml-models/chaiml-qwen-dpo-exp1-da-53786-v4/flywheel_model.2.safetensors
chaiml-qwen-dpo-exp1-da-53786-v4-mkmlizer: cp /dev/shm/model_cache/flywheel_model.1.safetensors s3://guanaco-mkml-models/chaiml-qwen-dpo-exp1-da-53786-v4/flywheel_model.1.safetensors
chaiml-qwen-dpo-exp1-da-53786-v4-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/chaiml-qwen-dpo-exp1-da-53786-v4/flywheel_model.0.safetensors
chaiml-qwen-dpo-exp1-da-53786-v4-mkmlizer: Loading 0: 0%| | 0/771 [00:00<?, ?it/s] Loading 0: 1%| | 5/771 [00:00<00:19, 40.15it/s] Loading 0: 2%|▏ | 17/771 [00:00<00:11, 68.17it/s] Loading 0: 4%|▎ | 28/771 [00:00<00:08, 82.80it/s] Loading 0: 5%|▍ | 38/771 [00:00<00:08, 88.31it/s] Loading 0: 6%|▌ | 47/771 [00:00<00:12, 57.36it/s] Loading 0: 7%|▋ | 54/771 [00:00<00:13, 52.13it/s] Loading 0: 8%|▊ | 64/771 [00:01<00:11, 62.86it/s] Loading 0: 10%|▉ | 75/771 [00:01<00:09, 74.20it/s] Loading 0: 11%|█ | 84/771 [00:01<00:08, 77.96it/s] Loading 0: 12%|█▏ | 93/771 [00:01<00:09, 73.25it/s] Loading 0: 14%|█▎ | 106/771 [00:01<00:10, 63.40it/s] Loading 0: 15%|█▍ | 114/771 [00:01<00:10, 62.36it/s] Loading 0: 16%|█▌ | 121/771 [00:01<00:10, 59.40it/s] Loading 0: 17%|█▋ | 132/771 [00:01<00:09, 69.74it/s] Loading 0: 18%|█▊ | 142/771 [00:02<00:08, 76.42it/s] Loading 0: 20%|█▉ | 152/771 [00:02<00:07, 81.95it/s] Loading 0: 21%|██ | 161/771 [00:02<00:07, 76.87it/s] Loading 0: 22%|██▏ | 170/771 [00:02<00:11, 54.30it/s] Loading 0: 23%|██▎ | 177/771 [00:02<00:10, 54.31it/s] Loading 0: 24%|██▍ | 185/771 [00:02<00:10, 55.82it/s] Loading 0: 25%|██▌ | 196/771 [00:02<00:08, 67.64it/s] Loading 0: 27%|██▋ | 207/771 [00:03<00:07, 77.70it/s] Loading 0: 28%|██▊ | 216/771 [00:03<00:06, 80.57it/s] Loading 0: 29%|██▉ | 226/771 [00:03<00:08, 62.22it/s] Loading 0: 30%|███ | 234/771 [00:03<00:10, 53.47it/s] Loading 0: 32%|███▏ | 245/771 [00:03<00:08, 60.00it/s] Loading 0: 33%|███▎ | 256/771 [00:03<00:07, 70.36it/s] Loading 0: 35%|███▍ | 267/771 [00:03<00:06, 78.91it/s] Loading 0: 36%|███▌ | 276/771 [00:04<00:06, 81.52it/s] Loading 0: 37%|███▋ | 286/771 [00:04<00:07, 63.66it/s] Loading 0: 38%|███▊ | 294/771 [00:04<00:08, 53.87it/s] Loading 0: 39%|███▉ | 303/771 [00:18<03:42, 2.11it/s] Loading 0: 40%|████ | 311/771 [00:18<02:40, 2.86it/s] Loading 0: 41%|████ | 318/771 [00:18<02:00, 3.76it/s] Loading 0: 43%|████▎ | 329/771 [00:19<01:17, 5.73it/s] Loading 0: 45%|████▍ | 346/771 [00:19<00:45, 9.37it/s] Loading 0: 46%|████▌ | 353/771 [00:19<00:38, 10.72it/s] Loading 0: 47%|████▋ | 364/771 [00:19<00:27, 15.01it/s] Loading 0: 49%|████▊ | 375/771 [00:20<00:19, 20.49it/s] Loading 0: 50%|████▉ | 385/771 [00:20<00:14, 26.56it/s] Loading 0: 51%|█████ | 394/771 [00:20<00:11, 31.64it/s] Loading 0: 53%|█████▎ | 406/771 [00:20<00:10, 35.71it/s] Loading 0: 54%|█████▎ | 413/771 [00:20<00:10, 34.19it/s] Loading 0: 55%|█████▍ | 424/771 [00:20<00:07, 44.10it/s] Loading 0: 56%|█████▋ | 435/771 [00:20<00:06, 54.30it/s] Loading 0: 58%|█████▊ | 444/771 [00:21<00:05, 59.90it/s] Loading 0: 59%|█████▉ | 453/771 [00:21<00:05, 60.97it/s] Loading 0: 60%|██████ | 466/771 [00:21<00:05, 53.86it/s] Loading 0: 61%|██████▏ | 473/771 [00:21<00:06, 47.05it/s] Loading 0: 63%|██████▎ | 484/771 [00:21<00:05, 57.21it/s] Loading 0: 64%|██████▍ | 495/771 [00:21<00:04, 67.10it/s] Loading 0: 65%|██████▌ | 504/771 [00:22<00:03, 71.19it/s] Loading 0: 67%|██████▋ | 513/771 [00:22<00:03, 69.44it/s] Loading 0: 68%|██████▊ | 526/771 [00:22<00:03, 61.37it/s] Loading 0: 69%|██████▉ | 533/771 [00:22<00:04, 51.92it/s] Loading 0: 71%|███████ | 544/771 [00:22<00:03, 62.45it/s] Loading 0: 72%|███████▏ | 555/771 [00:22<00:03, 71.94it/s] Loading 0: 73%|███████▎ | 564/771 [00:22<00:02, 75.58it/s] Loading 0: 74%|███████▍ | 573/771 [00:23<00:02, 72.69it/s] Loading 0: 76%|███████▌ | 586/771 [00:23<00:02, 62.98it/s] Loading 0: 77%|███████▋ | 594/771 [00:23<00:03, 53.22it/s] Loading 0: 78%|███████▊ | 605/771 [00:23<00:02, 59.59it/s] Loading 0: 80%|███████▉ | 616/771 [00:23<00:02, 69.32it/s] Loading 0: 81%|████████▏ | 627/771 [00:38<01:01, 2.34it/s] Loading 0: 83%|████████▎ | 637/771 [00:38<00:41, 3.25it/s] Loading 0: 84%|████████▍ | 646/771 [00:38<00:29, 4.31it/s] Loading 0: 85%|████████▍ | 653/771 [00:39<00:22, 5.27it/s] Loading 0: 86%|████████▌ | 664/771 [00:39<00:13, 7.81it/s] Loading 0: 88%|████████▊ | 675/771 [00:39<00:08, 11.22it/s] Loading 0: 89%|████████▊ | 684/771 [00:39<00:05, 14.76it/s] Loading 0: 90%|████████▉ | 693/771 [00:39<00:04, 18.86it/s] Loading 0: 92%|█████████▏| 706/771 [00:39<00:02, 24.37it/s] Loading 0: 92%|█████████▏| 713/771 [00:40<00:02, 25.85it/s] Loading 0: 94%|█████████▍| 724/771 [00:40<00:01, 34.00it/s] Loading 0: 95%|█████████▌| 735/771 [00:40<00:00, 42.74it/s] Loading 0: 96%|█████████▋| 743/771 [00:40<00:00, 47.05it/s] Loading 0: 97%|█████████▋| 751/771 [00:40<00:00, 49.30it/s] Loading 0: 99%|█████████▉| 766/771 [00:48<00:01, 4.28it/s]
Job chaiml-qwen-dpo-exp1-da-53786-v4-mkmlizer completed after 187.67s with status: succeeded
Stopping job with name chaiml-qwen-dpo-exp1-da-53786-v4-mkmlizer
Pipeline stage MKMLizer completed in 188.13s
run pipeline stage %s
Running pipeline stage MKMLTemplater
Pipeline stage MKMLTemplater completed in 0.14s
run pipeline stage %s
Running pipeline stage MKMLDeployer
Creating inference service chaiml-qwen-dpo-exp1-da-53786-v4
Waiting for inference service chaiml-qwen-dpo-exp1-da-53786-v4 to be ready
Inference service chaiml-qwen-dpo-exp1-da-53786-v4 ready after 120.42517566680908s
Pipeline stage MKMLDeployer completed in 120.91s
run pipeline stage %s
Running pipeline stage StressChecker
Received healthy response to inference request in 4.04881477355957s
Received healthy response to inference request in 3.6422648429870605s
Received healthy response to inference request in 3.478243350982666s
Received healthy response to inference request in 3.755660057067871s
Received healthy response to inference request in 3.6961288452148438s
5 requests
0 failed requests
5th percentile: 3.5110476493835447
10th percentile: 3.543851947784424
20th percentile: 3.609460544586182
30th percentile: 3.6530376434326173
40th percentile: 3.6745832443237303
50th percentile: 3.6961288452148438
60th percentile: 3.719941329956055
70th percentile: 3.7437538146972655
80th percentile: 3.814291000366211
90th percentile: 3.9315528869628906
95th percentile: 3.9901838302612305
99th percentile: 4.0370885848999025
mean time: 3.7242223739624025
%s, retrying in %s seconds...
Received healthy response to inference request in 3.4102790355682373s
Received healthy response to inference request in 3.707852602005005s
Received healthy response to inference request in 3.2834858894348145s
Received healthy response to inference request in 3.43929123878479s
Received healthy response to inference request in 3.4389102458953857s
5 requests
0 failed requests
5th percentile: 3.308844518661499
10th percentile: 3.3342031478881835
20th percentile: 3.3849204063415526
30th percentile: 3.416005277633667
40th percentile: 3.4274577617645265
50th percentile: 3.4389102458953857
60th percentile: 3.4390626430511473
70th percentile: 3.4392150402069093
80th percentile: 3.493003511428833
90th percentile: 3.600428056716919
95th percentile: 3.654140329360962
99th percentile: 3.6971101474761965
mean time: 3.4559638023376467
Pipeline stage StressChecker completed in 38.56s
run pipeline stage %s
Running pipeline stage OfflineFamilyFriendlyTriggerPipeline
run_pipeline:run_in_cloud %s
starting trigger_guanaco_pipeline args=%s
triggered trigger_guanaco_pipeline args=%s
Pipeline stage OfflineFamilyFriendlyTriggerPipeline completed in 0.68s
run pipeline stage %s
Running pipeline stage TriggerMKMLProfilingPipeline
run_pipeline:run_in_cloud %s
starting trigger_guanaco_pipeline args=%s
triggered trigger_guanaco_pipeline args=%s
Pipeline stage TriggerMKMLProfilingPipeline completed in 0.69s
Shutdown handler de-registered
chaiml-qwen-dpo-exp1-da_53786_v4 status is now deployed due to DeploymentManager action
Shutdown handler registered
run pipeline %s
run pipeline stage %s
Running pipeline stage OfflineFamilyFriendlyScorer
Evaluating %s Family Friendly Score with %s threads
%s, retrying in %s seconds...
Evaluating %s Family Friendly Score with %s threads
Pipeline stage OfflineFamilyFriendlyScorer completed in 5905.94s
Shutdown handler de-registered
chaiml-qwen-dpo-exp1-da_53786_v4 status is now inactive due to auto deactivation removed underperforming models
chaiml-qwen-dpo-exp1-da_53786_v4 status is now torndown due to DeploymentManager action