chaiml-gy-exp61-dpo-exp_38838

developer_uid: bogoconic1

submission_id: chaiml-gy-exp61-dpo-exp_38838_v1

model_name: chaiml-gy-exp61-dpo-exp_38838_v1

model_group: ChaiML/gy-exp61-dpo-exp3

status: torndown

timestamp: 2025-07-02T17:44:23+00:00

num_battles: 7712

num_wins: 4089

celo_rating: 1300.39

family_friendly_score: 0.5578000000000001

family_friendly_standard_error: 0.007023662292565041

submission_type: basic

model_repo: ChaiML/gy-exp61-dpo-exp32ep8s2-gy-datamix-v2.5-orig-safe-grok-rerank-agreement-ep2

model_architecture: MistralForCausalLM

model_num_parameters: 24096691200.0

best_of: 8

max_input_tokens: 1024

max_output_tokens: 64

latencies: [{'batch_size': 1, 'throughput': 0.5195416348514608, 'latency_mean': 1.924666873216629, 'latency_p50': 1.9252662658691406, 'latency_p90': 2.140011429786682}, {'batch_size': 3, 'throughput': 1.039746211337719, 'latency_mean': 2.871832983493805, 'latency_p50': 2.8679229021072388, 'latency_p90': 3.136714506149292}, {'batch_size': 5, 'throughput': 1.3173370395246888, 'latency_mean': 3.7664580941200256, 'latency_p50': 3.7925151586532593, 'latency_p90': 4.15570719242096}, {'batch_size': 6, 'throughput': 1.4160451004752048, 'latency_mean': 4.206128002405166, 'latency_p50': 4.224631071090698, 'latency_p90': 4.614780735969544}, {'batch_size': 8, 'throughput': 1.5473113778716472, 'latency_mean': 5.121824027299881, 'latency_p50': 5.0999451875686646, 'latency_p90': 5.742206192016601}, {'batch_size': 10, 'throughput': 1.6273139554618916, 'latency_mean': 6.0918235063552855, 'latency_p50': 6.0948874950408936, 'latency_p90': 6.838292479515076}]

gpu_counts: {'NVIDIA A100-SXM4-80GB': 1}

display_name: chaiml-gy-exp61-dpo-exp_38838_v1

is_internal_developer: True

language_model: ChaiML/gy-exp61-dpo-exp32ep8s2-gy-datamix-v2.5-orig-safe-grok-rerank-agreement-ep2

model_size: 24B

ranking_group: single

throughput_3p7s: 1.31

us_pacific_date: 2025-07-02

win_ratio: 0.5302126556016598

generation_params: {'temperature': 0.6, 'top_p': 0.95, 'min_p': 0.025, 'top_k': 60, 'presence_penalty': 0.4, 'frequency_penalty': 0.4, 'stopping_words': ['<|im_start|>', '\n', '<|im_end|>'], 'max_input_tokens': 1024, 'best_of': 8, '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': 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-gy-exp61-dpo-exp-38838-v1-mkmlizer
Waiting for job on chaiml-gy-exp61-dpo-exp-38838-v1-mkmlizer to finish
chaiml-gy-exp61-dpo-exp-38838-v1-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
chaiml-gy-exp61-dpo-exp-38838-v1-mkmlizer: ║                                                                     ║
chaiml-gy-exp61-dpo-exp-38838-v1-mkmlizer: ║   ██████   ██████  █████   ████  ████                               ║
chaiml-gy-exp61-dpo-exp-38838-v1-mkmlizer: ║  ░░██████ ██████  ░░███   ███░  ░░███                               ║
chaiml-gy-exp61-dpo-exp-38838-v1-mkmlizer: ║   ░███░█████░███   ░███  ███     ░███                               ║
chaiml-gy-exp61-dpo-exp-38838-v1-mkmlizer: ║   ░███░░███ ░███   ░███████      ░███                               ║
chaiml-gy-exp61-dpo-exp-38838-v1-mkmlizer: ║   ░███ ░░░  ░███   ░███░░███     ░███                               ║
chaiml-gy-exp61-dpo-exp-38838-v1-mkmlizer: ║   ░███      ░███   ░███ ░░███    ░███                               ║
chaiml-gy-exp61-dpo-exp-38838-v1-mkmlizer: ║   █████     █████  █████ ░░████  █████                              ║
chaiml-gy-exp61-dpo-exp-38838-v1-mkmlizer: ║  ░░░░░     ░░░░░  ░░░░░   ░░░░  ░░░░░                               ║
chaiml-gy-exp61-dpo-exp-38838-v1-mkmlizer: ║                                                                     ║
chaiml-gy-exp61-dpo-exp-38838-v1-mkmlizer: ║  Version: 0.29.8                                                    ║
chaiml-gy-exp61-dpo-exp-38838-v1-mkmlizer: ║  Features: FLYWHEEL, CUDA                                           ║
chaiml-gy-exp61-dpo-exp-38838-v1-mkmlizer: ║  Copyright 2023-2025 MK ONE TECHNOLOGIES Inc.                       ║
chaiml-gy-exp61-dpo-exp-38838-v1-mkmlizer: ║  https://mk1.ai                                                     ║
chaiml-gy-exp61-dpo-exp-38838-v1-mkmlizer: ║                                                                     ║
chaiml-gy-exp61-dpo-exp-38838-v1-mkmlizer: ║  The license key for the current software has been verified as      ║
chaiml-gy-exp61-dpo-exp-38838-v1-mkmlizer: ║  belonging to:                                                      ║
chaiml-gy-exp61-dpo-exp-38838-v1-mkmlizer: ║                                                                     ║
chaiml-gy-exp61-dpo-exp-38838-v1-mkmlizer: ║  Chai Research Corp.                                                ║
chaiml-gy-exp61-dpo-exp-38838-v1-mkmlizer: ║  Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f                   ║
chaiml-gy-exp61-dpo-exp-38838-v1-mkmlizer: ║  Expiration: 2028-03-31 23:59:59                                    ║
chaiml-gy-exp61-dpo-exp-38838-v1-mkmlizer: ║                                                                     ║
chaiml-gy-exp61-dpo-exp-38838-v1-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
chaiml-gy-exp61-dpo-exp-38838-v1-mkmlizer: Downloaded to shared memory in 84.129s
chaiml-gy-exp61-dpo-exp-38838-v1-mkmlizer: Checking if ChaiML/gy-exp61-dpo-exp32ep8s2-gy-datamix-v2.5-orig-safe-grok-rerank-agreement-ep2 already exists in ChaiML
chaiml-gy-exp61-dpo-exp-38838-v1-mkmlizer: quantizing model to /dev/shm/model_cache, profile:s0, folder:/tmp/tmpjgp2de56, device:0
chaiml-gy-exp61-dpo-exp-38838-v1-mkmlizer: Saving flywheel model at /dev/shm/model_cache
chaiml-gy-exp61-dpo-exp-38838-v1-mkmlizer: creating bucket guanaco-mkml-models
chaiml-gy-exp61-dpo-exp-38838-v1-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
chaiml-gy-exp61-dpo-exp-38838-v1-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/chaiml-gy-exp61-dpo-exp-38838-v1/nvidia
chaiml-gy-exp61-dpo-exp-38838-v1-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/chaiml-gy-exp61-dpo-exp-38838-v1/nvidia/config.json
chaiml-gy-exp61-dpo-exp-38838-v1-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/chaiml-gy-exp61-dpo-exp-38838-v1/nvidia/special_tokens_map.json
chaiml-gy-exp61-dpo-exp-38838-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/chaiml-gy-exp61-dpo-exp-38838-v1/nvidia/tokenizer_config.json
chaiml-gy-exp61-dpo-exp-38838-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/chaiml-gy-exp61-dpo-exp-38838-v1/nvidia/tokenizer.json
chaiml-gy-exp61-dpo-exp-38838-v1-mkmlizer: cp /dev/shm/model_cache/flywheel_model.1.safetensors s3://guanaco-mkml-models/chaiml-gy-exp61-dpo-exp-38838-v1/nvidia/flywheel_model.1.safetensors
chaiml-gy-exp61-dpo-exp-38838-v1-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/chaiml-gy-exp61-dpo-exp-38838-v1/nvidia/flywheel_model.0.safetensors
chaiml-gy-exp61-dpo-exp-38838-v1-mkmlizer: 
Loading 0:   0%|          | 0/363 [00:00<?, ?it/s]
Loading 0:   1%|          | 4/363 [00:00<00:10, 33.25it/s]
Loading 0:   2%|▏         | 8/363 [00:00<00:13, 25.88it/s]
Loading 0:   3%|▎         | 12/363 [00:00<00:13, 26.92it/s]
Loading 0:   4%|▍         | 15/363 [00:00<00:15, 22.22it/s]
Loading 0:   6%|▌         | 21/363 [00:00<00:12, 27.76it/s]
Loading 0:   7%|▋         | 24/363 [00:00<00:14, 23.44it/s]
Loading 0:   8%|▊         | 29/363 [00:01<00:11, 29.30it/s]
Loading 0:   9%|▉         | 33/363 [00:01<00:14, 22.61it/s]
Loading 0:  10%|▉         | 36/363 [00:01<00:16, 19.24it/s]
Loading 0:  11%|█         | 40/363 [00:01<00:14, 22.37it/s]
Loading 0:  12%|█▏        | 43/363 [00:01<00:14, 22.41it/s]
Loading 0:  13%|█▎        | 46/363 [00:02<00:19, 16.66it/s]
Loading 0:  13%|█▎        | 49/363 [00:02<00:16, 18.70it/s]
Loading 0:  14%|█▍        | 52/363 [00:02<00:16, 19.38it/s]
Loading 0:  16%|█▌        | 57/363 [00:02<00:13, 23.21it/s]
Loading 0:  17%|█▋        | 60/363 [00:02<00:15, 20.01it/s]
Loading 0:  18%|█▊        | 64/363 [00:02<00:12, 23.83it/s]
Loading 0:  19%|█▉        | 69/363 [00:02<00:10, 26.74it/s]
Loading 0:  20%|█▉        | 72/363 [00:03<00:15, 19.35it/s]
Loading 0:  21%|██        | 75/363 [00:03<00:14, 19.91it/s]
Loading 0:  21%|██▏       | 78/363 [00:03<00:13, 21.66it/s]
Loading 0:  22%|██▏       | 81/363 [00:03<00:13, 20.20it/s]
Loading 0:  24%|██▎       | 86/363 [00:03<00:11, 24.45it/s]
Loading 0:  25%|██▍       | 89/363 [00:03<00:11, 24.26it/s]
Loading 0:  25%|██▌       | 92/363 [00:04<00:14, 19.26it/s]
Loading 0:  27%|██▋       | 97/363 [00:04<00:10, 25.12it/s]
Loading 0:  28%|██▊       | 100/363 [00:04<00:10, 26.14it/s]
Loading 0:  28%|██▊       | 103/363 [00:04<00:10, 24.83it/s]
Loading 0:  29%|██▉       | 107/363 [00:04<00:13, 19.65it/s]
Loading 0:  31%|███       | 111/363 [00:04<00:10, 23.33it/s]
Loading 0:  31%|███▏      | 114/363 [00:05<00:11, 21.19it/s]
Loading 0:  33%|███▎      | 120/363 [00:05<00:09, 26.62it/s]
Loading 0:  34%|███▍      | 123/363 [00:05<00:10, 23.70it/s]
Loading 0:  36%|███▌      | 129/363 [00:05<00:07, 29.27it/s]
Loading 0:  37%|███▋      | 133/363 [00:05<00:08, 27.11it/s]
Loading 0:  38%|███▊      | 138/363 [00:05<00:07, 29.37it/s]
Loading 0:  39%|███▉      | 142/363 [00:06<00:08, 26.53it/s]
Loading 0:  40%|████      | 146/363 [00:06<00:07, 29.22it/s]
Loading 0:  41%|████▏     | 150/363 [00:06<00:09, 22.90it/s]
Loading 0:  42%|████▏     | 153/363 [00:06<00:10, 19.26it/s]
Loading 0:  43%|████▎     | 157/363 [00:06<00:09, 22.36it/s]
Loading 0:  44%|████▍     | 160/363 [00:06<00:08, 22.72it/s]
Loading 0:  45%|████▌     | 165/363 [00:07<00:07, 26.62it/s]
Loading 0:  46%|████▋     | 168/363 [00:07<00:08, 23.65it/s]
Loading 0:  48%|████▊     | 174/363 [00:07<00:06, 28.15it/s]
Loading 0:  49%|████▉     | 177/363 [00:07<00:07, 24.52it/s]
Loading 0:  50%|█████     | 182/363 [00:07<00:06, 27.95it/s]
Loading 0:  52%|█████▏    | 187/363 [00:07<00:06, 25.16it/s]
Loading 0:  52%|█████▏    | 190/363 [00:08<00:07, 22.75it/s]
Loading 0:  53%|█████▎    | 193/363 [00:08<00:07, 23.46it/s]
Loading 0:  54%|█████▍    | 196/363 [00:08<00:07, 23.44it/s]
Loading 0:  55%|█████▌    | 200/363 [00:23<00:06, 23.44it/s]
Loading 0:  55%|█████▌    | 201/363 [00:23<03:04,  1.14s/it]
Loading 0:  56%|█████▌    | 203/363 [00:23<02:32,  1.05it/s]
Loading 0:  57%|█████▋    | 208/363 [00:23<01:31,  1.69it/s]
Loading 0:  58%|█████▊    | 211/363 [00:23<01:10,  2.15it/s]
Loading 0:  59%|█████▉    | 214/363 [00:23<00:52,  2.83it/s]
Loading 0:  60%|██████    | 218/363 [00:23<00:35,  4.06it/s]
Loading 0:  61%|██████    | 221/363 [00:23<00:27,  5.17it/s]
Loading 0:  62%|██████▏   | 224/363 [00:24<00:22,  6.13it/s]
Loading 0:  63%|██████▎   | 228/363 [00:24<00:15,  8.62it/s]
Loading 0:  64%|██████▎   | 231/363 [00:24<00:13,  9.93it/s]
Loading 0:  65%|██████▌   | 237/363 [00:24<00:08, 14.95it/s]
Loading 0:  66%|██████▌   | 240/363 [00:24<00:07, 15.51it/s]
Loading 0:  68%|██████▊   | 246/363 [00:24<00:05, 21.31it/s]
Loading 0:  69%|██████▉   | 250/363 [00:25<00:05, 21.73it/s]
Loading 0:  70%|███████   | 255/363 [00:25<00:04, 25.42it/s]
Loading 0:  71%|███████▏  | 259/363 [00:25<00:04, 24.67it/s]
Loading 0:  73%|███████▎  | 264/363 [00:25<00:03, 29.43it/s]
Loading 0:  74%|███████▍  | 268/363 [00:25<00:04, 22.53it/s]
Loading 0:  75%|███████▍  | 271/363 [00:25<00:04, 21.09it/s]
Loading 0:  75%|███████▌  | 274/363 [00:26<00:04, 22.12it/s]
Loading 0:  76%|███████▋  | 277/363 [00:26<00:03, 22.55it/s]
Loading 0:  78%|███████▊  | 282/363 [00:26<00:03, 26.35it/s]
Loading 0:  79%|███████▊  | 285/363 [00:26<00:03, 22.56it/s]
Loading 0:  80%|███████▉  | 289/363 [00:26<00:02, 25.83it/s]
Loading 0:  80%|████████  | 292/363 [00:26<00:02, 26.02it/s]
Loading 0:  81%|████████▏ | 295/363 [00:26<00:02, 25.37it/s]
Loading 0:  82%|████████▏ | 299/363 [00:27<00:02, 26.46it/s]
Loading 0:  84%|████████▎ | 304/363 [00:27<00:02, 23.66it/s]
Loading 0:  85%|████████▍ | 307/363 [00:27<00:02, 21.75it/s]
Loading 0:  85%|████████▌ | 310/363 [00:27<00:02, 22.84it/s]
Loading 0:  86%|████████▌ | 313/363 [00:27<00:02, 23.37it/s]
Loading 0:  88%|████████▊ | 318/363 [00:27<00:01, 27.71it/s]
Loading 0:  88%|████████▊ | 321/363 [00:27<00:01, 24.18it/s]
Loading 0:  90%|█████████ | 327/363 [00:28<00:01, 29.97it/s]
Loading 0:  91%|█████████ | 331/363 [00:28<00:01, 27.80it/s]
Loading 0:  92%|█████████▏| 335/363 [00:28<00:00, 29.08it/s]
Loading 0:  93%|█████████▎| 339/363 [00:28<00:00, 28.48it/s]
Loading 0:  94%|█████████▍| 342/363 [00:28<00:01, 16.82it/s]
Loading 0:  96%|█████████▌| 347/363 [00:29<00:00, 20.63it/s]
Loading 0:  96%|█████████▋| 350/363 [00:29<00:00, 21.53it/s]
Loading 0:  98%|█████████▊| 355/363 [00:29<00:00, 26.32it/s]
Loading 0:  99%|█████████▉| 359/363 [00:29<00:00, 24.85it/s]
                                                            
Job chaiml-gy-exp61-dpo-exp-38838-v1-mkmlizer completed after 165.96s with status: succeeded
Stopping job with name chaiml-gy-exp61-dpo-exp-38838-v1-mkmlizer
Pipeline stage MKMLizer completed in 166.49s
run pipeline stage %s
Running pipeline stage MKMLTemplater
Pipeline stage MKMLTemplater completed in 0.16s
run pipeline stage %s
Running pipeline stage MKMLDeployer
Creating inference service chaiml-gy-exp61-dpo-exp-38838-v1
Waiting for inference service chaiml-gy-exp61-dpo-exp-38838-v1 to be ready
Failed to get response for submission chaiml-qwen3-14base-sft_43480_v2: HTTPConnectionPool(host='chaiml-qwen3-14base-sft-43480-v2-predictor.tenant-chaiml-guanaco.k.chaiverse.com', port=80): Read timed out. (read timeout=12.0)
Inference service chaiml-gy-exp61-dpo-exp-38838-v1 ready after 221.07634615898132s
Pipeline stage MKMLDeployer completed in 221.68s
run pipeline stage %s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.567626476287842s
Received healthy response to inference request in 1.8555505275726318s
Received healthy response to inference request in 1.993596076965332s
Received healthy response to inference request in 1.9239354133605957s
Received healthy response to inference request in 1.8939895629882812s
5 requests
0 failed requests
5th percentile: 1.8632383346557617
10th percentile: 1.8709261417388916
20th percentile: 1.8863017559051514
30th percentile: 1.8999787330627442
40th percentile: 1.91195707321167
50th percentile: 1.9239354133605957
60th percentile: 1.9517996788024903
70th percentile: 1.9796639442443849
80th percentile: 2.108402156829834
90th percentile: 2.338014316558838
95th percentile: 2.45282039642334
99th percentile: 2.5446652603149413
mean time: 2.0469396114349365
Pipeline stage StressChecker completed in 11.55s
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 0.69s
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 0.68s
Shutdown handler de-registered
chaiml-gy-exp61-dpo-exp_38838_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 0.13s
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 chaiml-gy-exp61-dpo-exp-38838-v1-profiler
Waiting for inference service chaiml-gy-exp61-dpo-exp-38838-v1-profiler to be ready
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 6370.33s
Shutdown handler de-registered
chaiml-gy-exp61-dpo-exp_38838_v1 status is now inactive due to auto deactivation removed underperforming models
chaiml-gy-exp61-dpo-exp_38838_v1 status is now torndown due to DeploymentManager action