chaiml-qwen14b-dpoexp2-sf_329

developer_uid: NischayDnk

submission_id: chaiml-qwen14b-dpoexp2-sf_329_v1

model_name: chaiml-qwen14b-dpoexp2-sf_329_v1

model_group: ChaiML/qwen14b-dpoexp2-s

status: torndown

timestamp: 2025-04-07T21:20:18+00:00

num_battles: 8523

num_wins: 4208

celo_rating: 1278.46

family_friendly_score: 0.5514

family_friendly_standard_error: 0.007033605618742069

submission_type: basic

model_repo: ChaiML/qwen14b-dpoexp2-sftdel12kexp4

model_architecture: Qwen2ForCausalLM

model_num_parameters: 14769689600.0

best_of: 8

max_input_tokens: 1024

max_output_tokens: 64

reward_model: default

latencies: [{'batch_size': 1, 'throughput': 0.47056799160620505, 'latency_mean': 2.1250301921367645, 'latency_p50': 2.136647939682007, 'latency_p90': 2.339197850227356}, {'batch_size': 3, 'throughput': 0.8680819973705975, 'latency_mean': 3.4422541999816896, 'latency_p50': 3.4679874181747437, 'latency_p90': 3.7838761806488037}, {'batch_size': 5, 'throughput': 1.0587784045882889, 'latency_mean': 4.688134444952011, 'latency_p50': 4.693992257118225, 'latency_p90': 5.2620515584945675}, {'batch_size': 6, 'throughput': 1.1052671761011608, 'latency_mean': 5.4010685086250305, 'latency_p50': 5.423127770423889, 'latency_p90': 6.118965458869934}, {'batch_size': 10, 'throughput': 1.2028399723545236, 'latency_mean': 8.248922028541564, 'latency_p50': 8.219088792800903, 'latency_p90': 9.298924422264099}]

gpu_counts: {'NVIDIA RTX A5000': 1}

display_name: chaiml-qwen14b-dpoexp2-sf_329_v1

is_internal_developer: False

language_model: ChaiML/qwen14b-dpoexp2-sftdel12kexp4

model_size: 15B

ranking_group: single

throughput_3p7s: 0.92

us_pacific_date: 2025-04-07

win_ratio: 0.4937228675349056

generation_params: {'temperature': 0.75, 'top_p': 0.96, 'min_p': 0.025, 'top_k': 60, 'presence_penalty': 0.2, 'frequency_penalty': 0.2, 'stopping_words': ['<|im_end|>', '\n'], 'max_input_tokens': 1024, 'best_of': 8, 'max_output_tokens': 64}

formatter: {'memory_template': '[INST]<|im_start|>system\n{memory}<|im_end|>\n', 'prompt_template': '<|im_start|>user\n{prompt}<|im_end|>\n', 'bot_template': '<|im_start|>assistant\n{bot_name}: {message}<|im_end|>\n', 'user_template': '<|im_start|>user\n{user_name}: {message}<|im_end|>\n', 'response_template': '<|im_start|>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 chaiml-qwen14b-dpoexp2-sf-329-v1-mkmlizer
Waiting for job on chaiml-qwen14b-dpoexp2-sf-329-v1-mkmlizer to finish
chaiml-qwen14b-dpoexp2-sf-329-v1-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
chaiml-qwen14b-dpoexp2-sf-329-v1-mkmlizer: ║     _____            __           __                                ║
chaiml-qwen14b-dpoexp2-sf-329-v1-mkmlizer: ║    / _/ /_ ___    __/ /  ___ ___ / /                                ║
chaiml-qwen14b-dpoexp2-sf-329-v1-mkmlizer: ║   / _/ / // / |/|/ / _ \/ -_) -_) /                                 ║
chaiml-qwen14b-dpoexp2-sf-329-v1-mkmlizer: ║  /_//_/\_, /|__,__/_//_/\__/\__/_/                                  ║
chaiml-qwen14b-dpoexp2-sf-329-v1-mkmlizer: ║       /___/                                                         ║
chaiml-qwen14b-dpoexp2-sf-329-v1-mkmlizer: ║                                                                     ║
chaiml-qwen14b-dpoexp2-sf-329-v1-mkmlizer: ║  Version: 0.12.8                                                    ║
chaiml-qwen14b-dpoexp2-sf-329-v1-mkmlizer: ║  Copyright 2023 MK ONE TECHNOLOGIES Inc.                            ║
chaiml-qwen14b-dpoexp2-sf-329-v1-mkmlizer: ║  https://mk1.ai                                                     ║
chaiml-qwen14b-dpoexp2-sf-329-v1-mkmlizer: ║                                                                     ║
chaiml-qwen14b-dpoexp2-sf-329-v1-mkmlizer: ║  The license key for the current software has been verified as      ║
chaiml-qwen14b-dpoexp2-sf-329-v1-mkmlizer: ║  belonging to:                                                      ║
chaiml-qwen14b-dpoexp2-sf-329-v1-mkmlizer: ║                                                                     ║
chaiml-qwen14b-dpoexp2-sf-329-v1-mkmlizer: ║  Chai Research Corp.                                                ║
chaiml-qwen14b-dpoexp2-sf-329-v1-mkmlizer: ║  Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f                   ║
chaiml-qwen14b-dpoexp2-sf-329-v1-mkmlizer: ║  Expiration: 2025-04-15 23:59:59                                    ║
chaiml-qwen14b-dpoexp2-sf-329-v1-mkmlizer: ║                                                                     ║
chaiml-qwen14b-dpoexp2-sf-329-v1-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
chaiml-qwen14b-dpoexp2-sf-329-v1-mkmlizer: Downloaded to shared memory in 53.171s
chaiml-qwen14b-dpoexp2-sf-329-v1-mkmlizer: quantizing model to /dev/shm/model_cache, profile:s0, folder:/tmp/tmpt3ai1gm_, device:0
chaiml-qwen14b-dpoexp2-sf-329-v1-mkmlizer: Saving flywheel model at /dev/shm/model_cache
chaiml-qwen14b-dpoexp2-sf-329-v1-mkmlizer: quantized model in 36.041s
chaiml-qwen14b-dpoexp2-sf-329-v1-mkmlizer: Processed model ChaiML/qwen14b-dpoexp2-sftdel12kexp4 in 89.213s
chaiml-qwen14b-dpoexp2-sf-329-v1-mkmlizer: creating bucket guanaco-mkml-models
chaiml-qwen14b-dpoexp2-sf-329-v1-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
chaiml-qwen14b-dpoexp2-sf-329-v1-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/chaiml-qwen14b-dpoexp2-sf-329-v1
chaiml-qwen14b-dpoexp2-sf-329-v1-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/chaiml-qwen14b-dpoexp2-sf-329-v1/config.json
chaiml-qwen14b-dpoexp2-sf-329-v1-mkmlizer: cp /dev/shm/model_cache/added_tokens.json s3://guanaco-mkml-models/chaiml-qwen14b-dpoexp2-sf-329-v1/added_tokens.json
chaiml-qwen14b-dpoexp2-sf-329-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/chaiml-qwen14b-dpoexp2-sf-329-v1/tokenizer_config.json
chaiml-qwen14b-dpoexp2-sf-329-v1-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/chaiml-qwen14b-dpoexp2-sf-329-v1/special_tokens_map.json
chaiml-qwen14b-dpoexp2-sf-329-v1-mkmlizer: cp /dev/shm/model_cache/merges.txt s3://guanaco-mkml-models/chaiml-qwen14b-dpoexp2-sf-329-v1/merges.txt
chaiml-qwen14b-dpoexp2-sf-329-v1-mkmlizer: cp /dev/shm/model_cache/vocab.json s3://guanaco-mkml-models/chaiml-qwen14b-dpoexp2-sf-329-v1/vocab.json
chaiml-qwen14b-dpoexp2-sf-329-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/chaiml-qwen14b-dpoexp2-sf-329-v1/tokenizer.json
chaiml-qwen14b-dpoexp2-sf-329-v1-mkmlizer: cp /dev/shm/model_cache/flywheel_model.1.safetensors s3://guanaco-mkml-models/chaiml-qwen14b-dpoexp2-sf-329-v1/flywheel_model.1.safetensors
chaiml-qwen14b-dpoexp2-sf-329-v1-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/chaiml-qwen14b-dpoexp2-sf-329-v1/flywheel_model.0.safetensors
chaiml-qwen14b-dpoexp2-sf-329-v1-mkmlizer: 
Loading 0:   0%|          | 0/579 [00:00<?, ?it/s]
Loading 0:   3%|▎         | 15/579 [00:00<00:04, 139.99it/s]
Loading 0:   5%|▌         | 29/579 [00:00<00:04, 123.59it/s]
Loading 0:   8%|▊         | 48/579 [00:00<00:03, 151.12it/s]
Loading 0:  11%|█         | 64/579 [00:00<00:03, 142.37it/s]
Loading 0:  14%|█▍        | 81/579 [00:00<00:05, 95.53it/s] 
Loading 0:  16%|█▌        | 94/579 [00:00<00:04, 102.12it/s]
Loading 0:  18%|█▊        | 107/579 [00:00<00:04, 108.17it/s]
Loading 0:  21%|██        | 120/579 [00:01<00:04, 113.21it/s]
Loading 0:  24%|██▍       | 138/579 [00:01<00:03, 130.83it/s]
Loading 0:  27%|██▋       | 154/579 [00:01<00:03, 136.78it/s]
Loading 0:  29%|██▉       | 169/579 [00:01<00:03, 131.63it/s]
Loading 0:  33%|███▎      | 189/579 [00:01<00:03, 100.18it/s]
Loading 0:  35%|███▍      | 201/579 [00:01<00:03, 96.69it/s] 
Loading 0:  38%|███▊      | 219/579 [00:01<00:03, 111.98it/s]
Loading 0:  40%|████      | 232/579 [00:02<00:02, 115.85it/s]
Loading 0:  42%|████▏     | 245/579 [00:02<00:02, 118.33it/s]
Loading 0:  45%|████▌     | 263/579 [00:02<00:02, 133.17it/s]
Loading 0:  48%|████▊     | 279/579 [00:02<00:02, 138.80it/s]
Loading 0:  51%|█████▏    | 297/579 [00:02<00:02, 102.64it/s]
Loading 0:  54%|█████▎    | 310/579 [00:02<00:02, 99.96it/s] 
Loading 0:  56%|█████▋    | 327/579 [00:02<00:02, 112.91it/s]
Loading 0:  59%|█████▊    | 340/579 [00:02<00:02, 116.54it/s]
Loading 0:  61%|██████    | 353/579 [00:03<00:01, 119.12it/s]
Loading 0:  64%|██████▍   | 371/579 [00:03<00:01, 134.95it/s]
Loading 0:  67%|██████▋   | 387/579 [00:03<00:01, 139.19it/s]
Loading 0:  70%|██████▉   | 405/579 [00:03<00:01, 104.07it/s]
Loading 0:  72%|███████▏  | 418/579 [00:03<00:01, 100.80it/s]
Loading 0:  75%|███████▌  | 435/579 [00:03<00:01, 113.30it/s]
Loading 0:  77%|███████▋  | 448/579 [00:03<00:01, 116.73it/s]
Loading 0:  80%|███████▉  | 461/579 [00:03<00:00, 118.93it/s]
Loading 0:  83%|████████▎ | 480/579 [00:04<00:00, 137.48it/s]
Loading 0:  85%|████████▌ | 495/579 [00:04<00:00, 140.21it/s]
Loading 0:  88%|████████▊ | 512/579 [00:25<00:00, 140.21it/s]
Loading 0:  89%|████████▊ | 513/579 [00:25<00:25,  2.55it/s] 
Loading 0:  90%|████████▉ | 521/579 [00:25<00:18,  3.06it/s]
Loading 0:  92%|█████████▏| 534/579 [00:25<00:10,  4.26it/s]
Loading 0:  96%|█████████▌| 554/579 [00:25<00:03,  6.86it/s]
Loading 0:  98%|█████████▊| 569/579 [00:25<00:01,  9.48it/s]
                                                            
Job chaiml-qwen14b-dpoexp2-sf-329-v1-mkmlizer completed after 116.15s with status: succeeded
Stopping job with name chaiml-qwen14b-dpoexp2-sf-329-v1-mkmlizer
Pipeline stage MKMLizer completed in 116.64s
run pipeline stage %s
Running pipeline stage MKMLTemplater
Pipeline stage MKMLTemplater completed in 0.15s
run pipeline stage %s
Running pipeline stage MKMLDeployer
Creating inference service chaiml-qwen14b-dpoexp2-sf-329-v1
Waiting for inference service chaiml-qwen14b-dpoexp2-sf-329-v1 to be ready
Inference service chaiml-qwen14b-dpoexp2-sf-329-v1 ready after 100.35221672058105s
Pipeline stage MKMLDeployer completed in 100.83s
run pipeline stage %s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.707059383392334s
Received healthy response to inference request in 1.9321942329406738s
Received healthy response to inference request in 1.9667060375213623s
Received healthy response to inference request in 1.9217529296875s
Received healthy response to inference request in 2.075956344604492s
5 requests
0 failed requests
5th percentile: 1.9238411903381347
10th percentile: 1.9259294509887694
20th percentile: 1.930105972290039
30th percentile: 1.9390965938568114
40th percentile: 1.9529013156890869
50th percentile: 1.9667060375213623
60th percentile: 2.010406160354614
70th percentile: 2.054106283187866
80th percentile: 2.2021769523620605
90th percentile: 2.4546181678771974
95th percentile: 2.5808387756347657
99th percentile: 2.6818152618408204
mean time: 2.1207337856292723
Pipeline stage StressChecker completed in 11.90s
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 0.65s
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 0.64s
Shutdown handler de-registered
chaiml-qwen14b-dpoexp2-sf_329_v1 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 5004.55s
Shutdown handler de-registered
chaiml-qwen14b-dpoexp2-sf_329_v1 status is now inactive due to auto deactivation removed underperforming models
chaiml-qwen14b-dpoexp2-sf_329_v1 status is now torndown due to DeploymentManager action