chaiml-dpo-qwen15b-sft1_40097

developer_uid: NischayDnk

submission_id: chaiml-dpo-qwen15b-sft1_40097_v1

model_name: chaiml-dpo-qwen15b-sft1_40097_v1

model_group: ChaiML/dpo-qwen15b-sft12

status: torndown

timestamp: 2025-04-10T16:35:31+00:00

num_battles: 8380

num_wins: 4148

celo_rating: 1281.2

family_friendly_score: 0.5354

family_friendly_standard_error: 0.007053323188398501

submission_type: basic

model_repo: ChaiML/dpo-qwen15b-sft1290-fixbug-v2

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': 0.4734959072900481, 'latency_mean': 2.1118821501731873, 'latency_p50': 2.1213014125823975, 'latency_p90': 2.3309664011001585}, {'batch_size': 3, 'throughput': 0.880701670811976, 'latency_mean': 3.394742498397827, 'latency_p50': 3.38239061832428, 'latency_p90': 3.821647381782532}, {'batch_size': 5, 'throughput': 1.0716337431361647, 'latency_mean': 4.645122377872467, 'latency_p50': 4.621222496032715, 'latency_p90': 5.228978061676025}, {'batch_size': 6, 'throughput': 1.1305627546961274, 'latency_mean': 5.291116853952408, 'latency_p50': 5.244798183441162, 'latency_p90': 5.9639874458312985}, {'batch_size': 10, 'throughput': 1.1983114233085135, 'latency_mean': 8.280270256996154, 'latency_p50': 8.108829259872437, 'latency_p90': 9.453574776649475}]

gpu_counts: {'NVIDIA RTX A5000': 1}

display_name: chaiml-dpo-qwen15b-sft1_40097_v1

is_internal_developer: False

language_model: ChaiML/dpo-qwen15b-sft1290-fixbug-v2

model_size: 15B

ranking_group: single

throughput_3p7s: 0.94

us_pacific_date: 2025-04-10

win_ratio: 0.4949880668257757

generation_params: {'temperature': 0.76, 'top_p': 0.95, 'min_p': 0.025, 'top_k': 80, 'presence_penalty': 0.25, 'frequency_penalty': 0.25, 'stopping_words': ['You:', '\n'], 'max_input_tokens': 1024, 'best_of': 8, 'max_output_tokens': 64}

formatter: {'memory_template': '[INST]', 'prompt_template': '', 'bot_template': '{bot_name}:{message}\n', 'user_template': 'You:{message}\n', 'response_template': '{bot_name}####', 'truncate_by_message': True}

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-dpo-qwen15b-sft1-40097-v1-mkmlizer
Waiting for job on chaiml-dpo-qwen15b-sft1-40097-v1-mkmlizer to finish
chaiml-dpo-qwen15b-sft1-40097-v1-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
chaiml-dpo-qwen15b-sft1-40097-v1-mkmlizer: ║     _____            __           __                                ║
chaiml-dpo-qwen15b-sft1-40097-v1-mkmlizer: ║    / _/ /_ ___    __/ /  ___ ___ / /                                ║
chaiml-dpo-qwen15b-sft1-40097-v1-mkmlizer: ║   / _/ / // / |/|/ / _ \/ -_) -_) /                                 ║
chaiml-dpo-qwen15b-sft1-40097-v1-mkmlizer: ║  /_//_/\_, /|__,__/_//_/\__/\__/_/                                  ║
chaiml-dpo-qwen15b-sft1-40097-v1-mkmlizer: ║       /___/                                                         ║
chaiml-dpo-qwen15b-sft1-40097-v1-mkmlizer: ║                                                                     ║
chaiml-dpo-qwen15b-sft1-40097-v1-mkmlizer: ║  Version: 0.12.8                                                    ║
chaiml-dpo-qwen15b-sft1-40097-v1-mkmlizer: ║  Copyright 2023 MK ONE TECHNOLOGIES Inc.                            ║
chaiml-dpo-qwen15b-sft1-40097-v1-mkmlizer: ║  https://mk1.ai                                                     ║
chaiml-dpo-qwen15b-sft1-40097-v1-mkmlizer: ║                                                                     ║
chaiml-dpo-qwen15b-sft1-40097-v1-mkmlizer: ║  The license key for the current software has been verified as      ║
chaiml-dpo-qwen15b-sft1-40097-v1-mkmlizer: ║  belonging to:                                                      ║
chaiml-dpo-qwen15b-sft1-40097-v1-mkmlizer: ║                                                                     ║
chaiml-dpo-qwen15b-sft1-40097-v1-mkmlizer: ║  Chai Research Corp.                                                ║
chaiml-dpo-qwen15b-sft1-40097-v1-mkmlizer: ║  Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f                   ║
chaiml-dpo-qwen15b-sft1-40097-v1-mkmlizer: ║  Expiration: 2025-04-15 23:59:59                                    ║
chaiml-dpo-qwen15b-sft1-40097-v1-mkmlizer: ║                                                                     ║
chaiml-dpo-qwen15b-sft1-40097-v1-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
chaiml-dpo-qwen15b-sft1-40097-v1-mkmlizer: Downloaded to shared memory in 54.634s
chaiml-dpo-qwen15b-sft1-40097-v1-mkmlizer: quantizing model to /dev/shm/model_cache, profile:s0, folder:/tmp/tmp1902eixs, device:0
chaiml-dpo-qwen15b-sft1-40097-v1-mkmlizer: Saving flywheel model at /dev/shm/model_cache
chaiml-dpo-qwen15b-sft1-40097-v1-mkmlizer: quantized model in 36.269s
chaiml-dpo-qwen15b-sft1-40097-v1-mkmlizer: Processed model ChaiML/dpo-qwen15b-sft1290-fixbug-v2 in 90.904s
chaiml-dpo-qwen15b-sft1-40097-v1-mkmlizer: creating bucket guanaco-mkml-models
chaiml-dpo-qwen15b-sft1-40097-v1-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
chaiml-dpo-qwen15b-sft1-40097-v1-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/chaiml-dpo-qwen15b-sft1-40097-v1
chaiml-dpo-qwen15b-sft1-40097-v1-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/chaiml-dpo-qwen15b-sft1-40097-v1/config.json
chaiml-dpo-qwen15b-sft1-40097-v1-mkmlizer: cp /dev/shm/model_cache/added_tokens.json s3://guanaco-mkml-models/chaiml-dpo-qwen15b-sft1-40097-v1/added_tokens.json
chaiml-dpo-qwen15b-sft1-40097-v1-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/chaiml-dpo-qwen15b-sft1-40097-v1/special_tokens_map.json
chaiml-dpo-qwen15b-sft1-40097-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/chaiml-dpo-qwen15b-sft1-40097-v1/tokenizer_config.json
chaiml-dpo-qwen15b-sft1-40097-v1-mkmlizer: cp /dev/shm/model_cache/merges.txt s3://guanaco-mkml-models/chaiml-dpo-qwen15b-sft1-40097-v1/merges.txt
chaiml-dpo-qwen15b-sft1-40097-v1-mkmlizer: cp /dev/shm/model_cache/vocab.json s3://guanaco-mkml-models/chaiml-dpo-qwen15b-sft1-40097-v1/vocab.json
chaiml-dpo-qwen15b-sft1-40097-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/chaiml-dpo-qwen15b-sft1-40097-v1/tokenizer.json
chaiml-dpo-qwen15b-sft1-40097-v1-mkmlizer: cp /dev/shm/model_cache/flywheel_model.1.safetensors s3://guanaco-mkml-models/chaiml-dpo-qwen15b-sft1-40097-v1/flywheel_model.1.safetensors
chaiml-dpo-qwen15b-sft1-40097-v1-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/chaiml-dpo-qwen15b-sft1-40097-v1/flywheel_model.0.safetensors
chaiml-dpo-qwen15b-sft1-40097-v1-mkmlizer: 
Loading 0:   0%|          | 0/579 [00:00<?, ?it/s]
Loading 0:   3%|▎         | 15/579 [00:00<00:03, 142.71it/s]
Loading 0:   5%|▌         | 30/579 [00:00<00:04, 128.66it/s]
Loading 0:   8%|▊         | 49/579 [00:00<00:03, 153.27it/s]
Loading 0:  11%|█         | 65/579 [00:00<00:03, 134.99it/s]
Loading 0:  14%|█▍        | 81/579 [00:00<00:04, 100.65it/s]
Loading 0:  16%|█▌        | 94/579 [00:00<00:04, 106.47it/s]
Loading 0:  18%|█▊        | 107/579 [00:00<00:04, 112.09it/s]
Loading 0:  21%|██        | 120/579 [00:01<00:03, 115.74it/s]
Loading 0:  24%|██▍       | 140/579 [00:01<00:03, 138.61it/s]
Loading 0:  27%|██▋       | 155/579 [00:01<00:03, 132.30it/s]
Loading 0:  29%|██▉       | 169/579 [00:01<00:03, 133.40it/s]
Loading 0:  32%|███▏      | 187/579 [00:01<00:02, 146.32it/s]
Loading 0:  35%|███▌      | 203/579 [00:01<00:04, 90.78it/s] 
Loading 0:  38%|███▊      | 219/579 [00:01<00:03, 103.91it/s]
Loading 0:  40%|████      | 233/579 [00:01<00:03, 106.70it/s]
Loading 0:  44%|████▎     | 252/579 [00:02<00:02, 125.84it/s]
Loading 0:  46%|████▌     | 267/579 [00:02<00:02, 129.84it/s]
Loading 0:  49%|████▊     | 282/579 [00:02<00:02, 129.14it/s]
Loading 0:  51%|█████▏    | 297/579 [00:02<00:02, 105.26it/s]
Loading 0:  53%|█████▎    | 309/579 [00:02<00:02, 98.54it/s] 
Loading 0:  56%|█████▋    | 327/579 [00:02<00:02, 113.14it/s]
Loading 0:  59%|█████▊    | 340/579 [00:02<00:02, 116.34it/s]
Loading 0:  61%|██████    | 353/579 [00:02<00:01, 118.55it/s]
Loading 0:  64%|██████▍   | 371/579 [00:03<00:01, 133.87it/s]
Loading 0:  67%|██████▋   | 387/579 [00:03<00:01, 138.72it/s]
Loading 0:  70%|██████▉   | 405/579 [00:03<00:01, 103.72it/s]
Loading 0:  72%|███████▏  | 418/579 [00:03<00:01, 100.66it/s]
Loading 0:  75%|███████▌  | 435/579 [00:03<00:01, 112.62it/s]
Loading 0:  77%|███████▋  | 448/579 [00:03<00:01, 116.18it/s]
Loading 0:  80%|███████▉  | 461/579 [00:03<00:00, 119.59it/s]
Loading 0:  83%|████████▎ | 483/579 [00:04<00:00, 140.25it/s]
Loading 0:  86%|████████▌ | 498/579 [00:04<00:00, 136.83it/s]
Loading 0:  88%|████████▊ | 512/579 [00:21<00:00, 136.83it/s]
Loading 0:  89%|████████▊ | 513/579 [00:25<00:27,  2.41it/s] 
Loading 0:  90%|████████▉ | 521/579 [00:25<00:19,  2.92it/s]
Loading 0:  92%|█████████▏| 534/579 [00:25<00:11,  4.08it/s]
Loading 0:  96%|█████████▌| 554/579 [00:26<00:03,  6.62it/s]
Loading 0:  98%|█████████▊| 569/579 [00:26<00:01,  9.19it/s]
                                                            
Job chaiml-dpo-qwen15b-sft1-40097-v1-mkmlizer completed after 115.79s with status: succeeded
Stopping job with name chaiml-dpo-qwen15b-sft1-40097-v1-mkmlizer
Pipeline stage MKMLizer completed in 116.33s
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-dpo-qwen15b-sft1-40097-v1
Waiting for inference service chaiml-dpo-qwen15b-sft1-40097-v1 to be ready
Inference service chaiml-dpo-qwen15b-sft1-40097-v1 ready after 110.6891188621521s
Pipeline stage MKMLDeployer completed in 111.23s
run pipeline stage %s
Running pipeline stage StressChecker
Received healthy response to inference request in 3.070786476135254s
Received healthy response to inference request in 2.1840600967407227s
Received healthy response to inference request in 2.102388620376587s
Received healthy response to inference request in 7.122741460800171s
Received healthy response to inference request in 1.991154432296753s
5 requests
0 failed requests
5th percentile: 2.0134012699127197
10th percentile: 2.0356481075286865
20th percentile: 2.08014178276062
30th percentile: 2.118722915649414
40th percentile: 2.1513915061950684
50th percentile: 2.1840600967407227
60th percentile: 2.538750648498535
70th percentile: 2.8934412002563477
80th percentile: 3.881177473068238
90th percentile: 5.5019594669342045
95th percentile: 6.312350463867187
99th percentile: 6.960663261413574
mean time: 3.2942262172698973
Pipeline stage StressChecker completed in 17.71s
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 0.75s
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 0.60s
Shutdown handler de-registered
chaiml-dpo-qwen15b-sft1_40097_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
Pipeline stage OfflineFamilyFriendlyScorer completed in 2678.63s
Shutdown handler de-registered
chaiml-dpo-qwen15b-sft1_40097_v1 status is now inactive due to auto deactivation removed underperforming models
chaiml-dpo-qwen15b-sft1_40097_v1 status is now torndown due to DeploymentManager action