chaiml-qwen25-14b-ft-sf_35028

developer_uid: azuruce

submission_id: chaiml-qwen25-14b-ft-sf_35028_v6

model_name: nis-best-config

model_group: ChaiML/qwen25-14B-ft-sft

status: torndown

timestamp: 2025-04-11T19:39:14+00:00

num_battles: 7204

num_wins: 3641

celo_rating: 1272.68

family_friendly_score: 0.5416000000000001

family_friendly_standard_error: 0.007046551497009016

submission_type: basic

model_repo: ChaiML/qwen25-14B-ft-sft-exp2-del12k

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': 0.47506959756454203, 'latency_mean': 2.1048895180225373, 'latency_p50': 2.1173077821731567, 'latency_p90': 2.298700976371765}, {'batch_size': 3, 'throughput': 0.8827307845430681, 'latency_mean': 3.389930386543274, 'latency_p50': 3.3882375955581665, 'latency_p90': 3.7330159187316894}, {'batch_size': 5, 'throughput': 1.0780834042084098, 'latency_mean': 4.619201074838639, 'latency_p50': 4.602953791618347, 'latency_p90': 5.230712938308716}, {'batch_size': 6, 'throughput': 1.128990606901588, 'latency_mean': 5.2724354255199435, 'latency_p50': 5.315587878227234, 'latency_p90': 5.89898567199707}, {'batch_size': 10, 'throughput': 1.2299338747178843, 'latency_mean': 8.06317032814026, 'latency_p50': 8.017940044403076, 'latency_p90': 9.264218997955322}]

gpu_counts: {'NVIDIA RTX A5000': 1}

display_name: nis-best-config

is_internal_developer: False

language_model: ChaiML/qwen25-14B-ft-sft-exp2-del12k

model_size: 15B

ranking_group: single

throughput_3p7s: 0.95

us_pacific_date: 2025-04-11

win_ratio: 0.5054136590782898

generation_params: {'temperature': 0.75, 'top_p': 0.96, 'min_p': 0.02, 'top_k': 60, 'presence_penalty': 0.2, 'frequency_penalty': 0.2, 'stopping_words': ['\n'], 'max_input_tokens': 1024, 'best_of': 8, 'max_output_tokens': 64}

formatter: {'memory_template': '', 'prompt_template': '', '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-qwen25-14b-ft-sf-35028-v6-mkmlizer
Waiting for job on chaiml-qwen25-14b-ft-sf-35028-v6-mkmlizer to finish
chaiml-qwen25-14b-ft-sf-35028-v6-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
chaiml-qwen25-14b-ft-sf-35028-v6-mkmlizer: ║     _____            __           __                                ║
chaiml-qwen25-14b-ft-sf-35028-v6-mkmlizer: ║    / _/ /_ ___    __/ /  ___ ___ / /                                ║
chaiml-qwen25-14b-ft-sf-35028-v6-mkmlizer: ║   / _/ / // / |/|/ / _ \/ -_) -_) /                                 ║
chaiml-qwen25-14b-ft-sf-35028-v6-mkmlizer: ║  /_//_/\_, /|__,__/_//_/\__/\__/_/                                  ║
chaiml-qwen25-14b-ft-sf-35028-v6-mkmlizer: ║       /___/                                                         ║
chaiml-qwen25-14b-ft-sf-35028-v6-mkmlizer: ║                                                                     ║
chaiml-qwen25-14b-ft-sf-35028-v6-mkmlizer: ║  Version: 0.12.8                                                    ║
chaiml-qwen25-14b-ft-sf-35028-v6-mkmlizer: ║  Copyright 2023 MK ONE TECHNOLOGIES Inc.                            ║
chaiml-qwen25-14b-ft-sf-35028-v6-mkmlizer: ║  https://mk1.ai                                                     ║
chaiml-qwen25-14b-ft-sf-35028-v6-mkmlizer: ║                                                                     ║
chaiml-qwen25-14b-ft-sf-35028-v6-mkmlizer: ║  The license key for the current software has been verified as      ║
chaiml-qwen25-14b-ft-sf-35028-v6-mkmlizer: ║  belonging to:                                                      ║
chaiml-qwen25-14b-ft-sf-35028-v6-mkmlizer: ║                                                                     ║
chaiml-qwen25-14b-ft-sf-35028-v6-mkmlizer: ║  Chai Research Corp.                                                ║
chaiml-qwen25-14b-ft-sf-35028-v6-mkmlizer: ║  Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f                   ║
chaiml-qwen25-14b-ft-sf-35028-v6-mkmlizer: ║  Expiration: 2025-04-15 23:59:59                                    ║
chaiml-qwen25-14b-ft-sf-35028-v6-mkmlizer: ║                                                                     ║
chaiml-qwen25-14b-ft-sf-35028-v6-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
chaiml-qwen25-14b-ft-sf-35028-v6-mkmlizer: Downloaded to shared memory in 41.203s
chaiml-qwen25-14b-ft-sf-35028-v6-mkmlizer: quantizing model to /dev/shm/model_cache, profile:s0, folder:/tmp/tmpwg8jj6n4, device:0
chaiml-qwen25-14b-ft-sf-35028-v6-mkmlizer: Saving flywheel model at /dev/shm/model_cache
Failed to get response for submission jellywibble-tyler-james_92283_v1: HTTPConnectionPool(host='jellywibble-tyler-james-92283-v1-predictor.tenant-chaiml-guanaco.k.chaiverse.com', port=80): Read timed out. (read timeout=12.0)
chaiml-qwen25-14b-ft-sf-35028-v6-mkmlizer: quantized model in 35.886s
chaiml-qwen25-14b-ft-sf-35028-v6-mkmlizer: Processed model ChaiML/qwen25-14B-ft-sft-exp2-del12k in 77.089s
chaiml-qwen25-14b-ft-sf-35028-v6-mkmlizer: creating bucket guanaco-mkml-models
chaiml-qwen25-14b-ft-sf-35028-v6-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
chaiml-qwen25-14b-ft-sf-35028-v6-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/chaiml-qwen25-14b-ft-sf-35028-v6
chaiml-qwen25-14b-ft-sf-35028-v6-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/chaiml-qwen25-14b-ft-sf-35028-v6/config.json
chaiml-qwen25-14b-ft-sf-35028-v6-mkmlizer: cp /dev/shm/model_cache/added_tokens.json s3://guanaco-mkml-models/chaiml-qwen25-14b-ft-sf-35028-v6/added_tokens.json
chaiml-qwen25-14b-ft-sf-35028-v6-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/chaiml-qwen25-14b-ft-sf-35028-v6/special_tokens_map.json
chaiml-qwen25-14b-ft-sf-35028-v6-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/chaiml-qwen25-14b-ft-sf-35028-v6/tokenizer_config.json
chaiml-qwen25-14b-ft-sf-35028-v6-mkmlizer: cp /dev/shm/model_cache/merges.txt s3://guanaco-mkml-models/chaiml-qwen25-14b-ft-sf-35028-v6/merges.txt
chaiml-qwen25-14b-ft-sf-35028-v6-mkmlizer: cp /dev/shm/model_cache/vocab.json s3://guanaco-mkml-models/chaiml-qwen25-14b-ft-sf-35028-v6/vocab.json
chaiml-qwen25-14b-ft-sf-35028-v6-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/chaiml-qwen25-14b-ft-sf-35028-v6/tokenizer.json
chaiml-qwen25-14b-ft-sf-35028-v6-mkmlizer: cp /dev/shm/model_cache/flywheel_model.1.safetensors s3://guanaco-mkml-models/chaiml-qwen25-14b-ft-sf-35028-v6/flywheel_model.1.safetensors
chaiml-qwen25-14b-ft-sf-35028-v6-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/chaiml-qwen25-14b-ft-sf-35028-v6/flywheel_model.0.safetensors
chaiml-qwen25-14b-ft-sf-35028-v6-mkmlizer: 
Loading 0:   0%|          | 0/579 [00:00<?, ?it/s]
Loading 0:   3%|▎         | 15/579 [00:00<00:03, 148.70it/s]
Loading 0:   5%|▌         | 30/579 [00:00<00:04, 135.43it/s]
Loading 0:   9%|▉         | 51/579 [00:00<00:03, 156.98it/s]
Loading 0:  12%|█▏        | 67/579 [00:00<00:03, 147.86it/s]
Loading 0:  14%|█▍        | 82/579 [00:00<00:05, 98.49it/s] 
Loading 0:  17%|█▋        | 96/579 [00:00<00:04, 104.22it/s]
Loading 0:  20%|██        | 118/579 [00:00<00:03, 126.82it/s]
Loading 0:  23%|██▎       | 132/579 [00:01<00:03, 125.09it/s]
Loading 0:  27%|██▋       | 154/579 [00:01<00:02, 144.09it/s]
Loading 0:  29%|██▉       | 170/579 [00:01<00:02, 141.60it/s]
Loading 0:  33%|███▎      | 189/579 [00:01<00:03, 106.13it/s]
Loading 0:  35%|███▍      | 202/579 [00:01<00:04, 91.02it/s] 
Loading 0:  38%|███▊      | 219/579 [00:01<00:03, 104.25it/s]
Loading 0:  40%|████      | 233/579 [00:01<00:03, 108.17it/s]
Loading 0:  44%|████▍     | 255/579 [00:02<00:02, 128.23it/s]
Loading 0:  46%|████▋     | 269/579 [00:02<00:02, 125.87it/s]
Loading 0:  50%|████▉     | 287/579 [00:02<00:02, 138.82it/s]
Loading 0:  52%|█████▏    | 302/579 [00:02<00:02, 99.83it/s] 
Loading 0:  55%|█████▍    | 316/579 [00:02<00:02, 104.91it/s]
Loading 0:  57%|█████▋    | 332/579 [00:02<00:02, 116.86it/s]
Loading 0:  61%|██████    | 351/579 [00:02<00:01, 128.73it/s]
Loading 0:  63%|██████▎   | 366/579 [00:03<00:01, 128.77it/s]
Loading 0:  67%|██████▋   | 387/579 [00:03<00:01, 144.29it/s]
Loading 0:  70%|██████▉   | 405/579 [00:03<00:01, 111.48it/s]
Loading 0:  72%|███████▏  | 418/579 [00:03<00:01, 108.62it/s]
Loading 0:  75%|███████▌  | 435/579 [00:03<00:01, 121.02it/s]
Loading 0:  78%|███████▊  | 449/579 [00:03<00:01, 120.81it/s]
Loading 0:  81%|████████▏ | 471/579 [00:03<00:00, 138.64it/s]
Loading 0:  84%|████████▍ | 486/579 [00:03<00:00, 135.95it/s]
Loading 0:  88%|████████▊ | 508/579 [00:04<00:00, 155.32it/s]
Loading 0:  88%|████████▊ | 512/579 [00:25<00:00, 155.32it/s]
Loading 0:  89%|████████▊ | 513/579 [00:25<00:29,  2.23it/s] 
Loading 0:  90%|█████████ | 522/579 [00:25<00:20,  2.81it/s]
Loading 0:  94%|█████████▍| 544/579 [00:25<00:07,  4.83it/s]
Loading 0:  97%|█████████▋| 560/579 [00:25<00:02,  6.85it/s]
                                                            
Job chaiml-qwen25-14b-ft-sf-35028-v6-mkmlizer completed after 105.77s with status: succeeded
Stopping job with name chaiml-qwen25-14b-ft-sf-35028-v6-mkmlizer
Pipeline stage MKMLizer completed in 106.24s
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-qwen25-14b-ft-sf-35028-v6
Waiting for inference service chaiml-qwen25-14b-ft-sf-35028-v6 to be ready
Failed to get response for submission jellywibble-tyler-james_92283_v1: HTTPConnectionPool(host='jellywibble-tyler-james-92283-v1-predictor.tenant-chaiml-guanaco.k.chaiverse.com', port=80): Read timed out. (read timeout=12.0)
Inference service chaiml-qwen25-14b-ft-sf-35028-v6 ready after 110.58231687545776s
Pipeline stage MKMLDeployer completed in 111.06s
run pipeline stage %s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.6105332374572754s
Received healthy response to inference request in 1.8856360912322998s
Received healthy response to inference request in 1.956897258758545s
Received healthy response to inference request in 2.2495181560516357s
Received healthy response to inference request in 1.8936643600463867s
5 requests
0 failed requests
5th percentile: 1.887241744995117
10th percentile: 1.8888473987579346
20th percentile: 1.8920587062835694
30th percentile: 1.9063109397888183
40th percentile: 1.9316040992736816
50th percentile: 1.956897258758545
60th percentile: 2.0739456176757813
70th percentile: 2.1909939765930173
80th percentile: 2.3217211723327638
90th percentile: 2.4661272048950194
95th percentile: 2.5383302211761474
99th percentile: 2.59609263420105
mean time: 2.1192498207092285
Pipeline stage StressChecker completed in 11.65s
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 0.62s
run pipeline stage %s
Running pipeline stage TriggerMKMLProfilingPipeline
run_pipeline:run_in_cloud %s
starting trigger_guanaco_pipeline args=%s
Failed to get response for submission jellywibble-tyler-james_92283_v1: HTTPConnectionPool(host='jellywibble-tyler-james-92283-v1-predictor.tenant-chaiml-guanaco.k.chaiverse.com', port=80): Read timed out. (read timeout=12.0)
triggered trigger_guanaco_pipeline args=%s
Pipeline stage TriggerMKMLProfilingPipeline completed in 0.68s
Shutdown handler de-registered
chaiml-qwen25-14b-ft-sf_35028_v6 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 0.08s
run pipeline stage %s
Running pipeline stage MKMLProfilerDeployer
Creating inference service chaiml-qwen25-14b-ft-sf-35028-v6-profiler
Waiting for inference service chaiml-qwen25-14b-ft-sf-35028-v6-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 5129.10s
Shutdown handler de-registered
chaiml-qwen25-14b-ft-sf_35028_v6 status is now inactive due to auto deactivation removed underperforming models
chaiml-qwen25-14b-ft-sf_35028_v6 status is now torndown due to DeploymentManager action