alexxi19-ft-v1-nemo-base-dpo

developer_uid: leosheng

submission_id: alexxi19-ft-v1-nemo-base-dpo_v4

model_name: alexxi19-ft-v1-nemo-base-dpo_v4

model_group: alexxi19/ft-v1-nemo-base

status: torndown

timestamp: 2025-02-21T07:07:02+00:00

num_battles: 8059

num_wins: 3739

celo_rating: 1243.15

family_friendly_score: 0.5636

family_friendly_standard_error: 0.007013630158484265

submission_type: basic

model_repo: alexxi19/ft-v1-nemo-base-dpo

model_architecture: MistralForCausalLM

model_num_parameters: 12772070400.0

best_of: 8

max_input_tokens: 1024

max_output_tokens: 64

latencies: [{'batch_size': 1, 'throughput': 0.6017065257612098, 'latency_mean': 1.66185156583786, 'latency_p50': 1.6774935722351074, 'latency_p90': 1.8257150888442992}, {'batch_size': 3, 'throughput': 1.1063228718493774, 'latency_mean': 2.7098269414901734, 'latency_p50': 2.7130420207977295, 'latency_p90': 2.9510035514831543}, {'batch_size': 5, 'throughput': 1.339601483550617, 'latency_mean': 3.7152580571174623, 'latency_p50': 3.717002511024475, 'latency_p90': 4.135295820236206}, {'batch_size': 6, 'throughput': 1.4086907741423496, 'latency_mean': 4.235752679109574, 'latency_p50': 4.259005069732666, 'latency_p90': 4.777411603927613}, {'batch_size': 8, 'throughput': 1.456935026168203, 'latency_mean': 5.450377551317215, 'latency_p50': 5.488942980766296, 'latency_p90': 6.096507453918457}, {'batch_size': 10, 'throughput': 1.5071797689874684, 'latency_mean': 6.590978096723557, 'latency_p50': 6.5294575691223145, 'latency_p90': 7.418133425712585}]

gpu_counts: {'NVIDIA RTX A5000': 1}

display_name: alexxi19-ft-v1-nemo-base-dpo_v4

is_internal_developer: False

language_model: alexxi19/ft-v1-nemo-base-dpo

model_size: 13B

ranking_group: single

throughput_3p7s: 1.34

us_pacific_date: 2025-02-20

win_ratio: 0.4639533440873558

generation_params: {'temperature': 1.0, 'top_p': 1.0, 'min_p': 0.0, 'top_k': 40, 'presence_penalty': 0.0, 'frequency_penalty': 0.0, 'stopping_words': ['\n'], 'max_input_tokens': 1024, 'best_of': 8, 'max_output_tokens': 64}

formatter: {'memory_template': '<|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 alexxi19-ft-v1-nemo-base-dpo-v4-mkmlizer
Waiting for job on alexxi19-ft-v1-nemo-base-dpo-v4-mkmlizer to finish
alexxi19-ft-v1-nemo-base-dpo-v4-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
alexxi19-ft-v1-nemo-base-dpo-v4-mkmlizer: ║     _____            __           __                                ║
alexxi19-ft-v1-nemo-base-dpo-v4-mkmlizer: ║    / _/ /_ ___    __/ /  ___ ___ / /                                ║
alexxi19-ft-v1-nemo-base-dpo-v4-mkmlizer: ║   / _/ / // / |/|/ / _ \/ -_) -_) /                                 ║
alexxi19-ft-v1-nemo-base-dpo-v4-mkmlizer: ║  /_//_/\_, /|__,__/_//_/\__/\__/_/                                  ║
alexxi19-ft-v1-nemo-base-dpo-v4-mkmlizer: ║       /___/                                                         ║
alexxi19-ft-v1-nemo-base-dpo-v4-mkmlizer: ║                                                                     ║
alexxi19-ft-v1-nemo-base-dpo-v4-mkmlizer: ║  Version: 0.12.8                                                    ║
alexxi19-ft-v1-nemo-base-dpo-v4-mkmlizer: ║  Copyright 2023 MK ONE TECHNOLOGIES Inc.                            ║
alexxi19-ft-v1-nemo-base-dpo-v4-mkmlizer: ║  https://mk1.ai                                                     ║
alexxi19-ft-v1-nemo-base-dpo-v4-mkmlizer: ║                                                                     ║
alexxi19-ft-v1-nemo-base-dpo-v4-mkmlizer: ║  The license key for the current software has been verified as      ║
alexxi19-ft-v1-nemo-base-dpo-v4-mkmlizer: ║  belonging to:                                                      ║
alexxi19-ft-v1-nemo-base-dpo-v4-mkmlizer: ║                                                                     ║
alexxi19-ft-v1-nemo-base-dpo-v4-mkmlizer: ║  Chai Research Corp.                                                ║
alexxi19-ft-v1-nemo-base-dpo-v4-mkmlizer: ║  Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f                   ║
alexxi19-ft-v1-nemo-base-dpo-v4-mkmlizer: ║  Expiration: 2025-04-15 23:59:59                                    ║
alexxi19-ft-v1-nemo-base-dpo-v4-mkmlizer: ║                                                                     ║
alexxi19-ft-v1-nemo-base-dpo-v4-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
alexxi19-ft-v1-nemo-base-dpo-v4-mkmlizer: Downloaded to shared memory in 46.097s
alexxi19-ft-v1-nemo-base-dpo-v4-mkmlizer: quantizing model to /dev/shm/model_cache, profile:s0, folder:/tmp/tmp80nqbnub, device:0
alexxi19-ft-v1-nemo-base-dpo-v4-mkmlizer: Saving flywheel model at /dev/shm/model_cache
alexxi19-ft-v1-nemo-base-dpo-v4-mkmlizer: quantized model in 37.585s
alexxi19-ft-v1-nemo-base-dpo-v4-mkmlizer: Processed model alexxi19/ft-v1-nemo-base-dpo in 83.682s
alexxi19-ft-v1-nemo-base-dpo-v4-mkmlizer: creating bucket guanaco-mkml-models
alexxi19-ft-v1-nemo-base-dpo-v4-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
alexxi19-ft-v1-nemo-base-dpo-v4-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/alexxi19-ft-v1-nemo-base-dpo-v4
alexxi19-ft-v1-nemo-base-dpo-v4-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/alexxi19-ft-v1-nemo-base-dpo-v4/special_tokens_map.json
alexxi19-ft-v1-nemo-base-dpo-v4-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/alexxi19-ft-v1-nemo-base-dpo-v4/config.json
alexxi19-ft-v1-nemo-base-dpo-v4-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/alexxi19-ft-v1-nemo-base-dpo-v4/tokenizer_config.json
alexxi19-ft-v1-nemo-base-dpo-v4-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/alexxi19-ft-v1-nemo-base-dpo-v4/tokenizer.json
alexxi19-ft-v1-nemo-base-dpo-v4-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/alexxi19-ft-v1-nemo-base-dpo-v4/flywheel_model.0.safetensors
alexxi19-ft-v1-nemo-base-dpo-v4-mkmlizer: 
Loading 0:   0%|          | 0/363 [00:00<?, ?it/s]
Loading 0:   1%|▏         | 5/363 [00:00<00:12, 29.15it/s]
Loading 0:   4%|▎         | 13/363 [00:00<00:07, 48.03it/s]
Loading 0:   5%|▌         | 19/363 [00:00<00:08, 41.22it/s]
Loading 0:   7%|▋         | 24/363 [00:00<00:08, 39.33it/s]
Loading 0:   8%|▊         | 30/363 [00:00<00:07, 44.68it/s]
Loading 0:  10%|▉         | 35/363 [00:00<00:07, 42.53it/s]
Loading 0:  11%|█         | 40/363 [00:00<00:07, 42.15it/s]
Loading 0:  12%|█▏        | 45/363 [00:01<00:07, 42.57it/s]
Loading 0:  14%|█▍        | 50/363 [00:01<00:09, 34.70it/s]
Loading 0:  15%|█▌        | 56/363 [00:01<00:07, 40.12it/s]
Loading 0:  17%|█▋        | 61/363 [00:01<00:10, 29.27it/s]
Loading 0:  18%|█▊        | 65/363 [00:01<00:10, 29.15it/s]
Loading 0:  20%|█▉        | 71/363 [00:01<00:08, 35.21it/s]
Loading 0:  21%|██        | 76/363 [00:02<00:08, 35.15it/s]
Loading 0:  22%|██▏       | 81/363 [00:02<00:07, 36.83it/s]
Loading 0:  23%|██▎       | 85/363 [00:02<00:07, 36.67it/s]
Loading 0:  25%|██▍       | 89/363 [00:02<00:07, 37.03it/s]
Loading 0:  26%|██▌       | 93/363 [00:02<00:07, 35.69it/s]
Loading 0:  27%|██▋       | 99/363 [00:02<00:06, 39.85it/s]
Loading 0:  29%|██▊       | 104/363 [00:02<00:06, 40.86it/s]
Loading 0:  30%|███       | 109/363 [00:02<00:05, 43.01it/s]
Loading 0:  31%|███▏      | 114/363 [00:03<00:07, 35.01it/s]
Loading 0:  33%|███▎      | 118/363 [00:03<00:07, 32.84it/s]
Loading 0:  34%|███▍      | 125/363 [00:03<00:06, 39.55it/s]
Loading 0:  36%|███▌      | 130/363 [00:03<00:05, 38.89it/s]
Loading 0:  37%|███▋      | 135/363 [00:03<00:05, 39.13it/s]
Loading 0:  39%|███▊      | 140/363 [00:03<00:05, 40.19it/s]
Loading 0:  40%|███▉      | 145/363 [00:04<00:08, 25.24it/s]
Loading 0:  41%|████      | 149/363 [00:04<00:08, 25.87it/s]
Loading 0:  43%|████▎     | 156/363 [00:04<00:06, 32.49it/s]
Loading 0:  44%|████▍     | 160/363 [00:04<00:06, 32.50it/s]
Loading 0:  45%|████▌     | 165/363 [00:04<00:05, 35.73it/s]
Loading 0:  47%|████▋     | 169/363 [00:04<00:05, 34.91it/s]
Loading 0:  48%|████▊     | 174/363 [00:04<00:05, 36.79it/s]
Loading 0:  49%|████▉     | 178/363 [00:04<00:05, 35.61it/s]
Loading 0:  50%|█████     | 183/363 [00:05<00:04, 38.51it/s]
Loading 0:  52%|█████▏    | 187/363 [00:05<00:04, 36.68it/s]
Loading 0:  53%|█████▎    | 192/363 [00:05<00:04, 38.03it/s]
Loading 0:  54%|█████▍    | 196/363 [00:05<00:04, 36.40it/s]
Loading 0:  55%|█████▌    | 201/363 [00:05<00:04, 38.16it/s]
Loading 0:  56%|█████▋    | 205/363 [00:05<00:04, 36.58it/s]
Loading 0:  58%|█████▊    | 210/363 [00:05<00:04, 38.22it/s]
Loading 0:  59%|█████▉    | 214/363 [00:05<00:04, 36.56it/s]
Loading 0:  60%|██████    | 218/363 [00:06<00:03, 36.36it/s]
Loading 0:  61%|██████▏   | 223/363 [00:06<00:05, 26.93it/s]
Loading 0:  63%|██████▎   | 227/363 [00:06<00:04, 28.50it/s]
Loading 0:  64%|██████▎   | 231/363 [00:06<00:04, 28.37it/s]
Loading 0:  65%|██████▌   | 237/363 [00:06<00:03, 33.76it/s]
Loading 0:  66%|██████▋   | 241/363 [00:06<00:03, 33.54it/s]
Loading 0:  68%|██████▊   | 246/363 [00:06<00:03, 36.71it/s]
Loading 0:  69%|██████▉   | 250/363 [00:07<00:03, 35.53it/s]
Loading 0:  70%|███████   | 255/363 [00:07<00:02, 37.93it/s]
Loading 0:  71%|███████▏  | 259/363 [00:07<00:02, 35.68it/s]
Loading 0:  73%|███████▎  | 264/363 [00:07<00:02, 37.66it/s]
Loading 0:  74%|███████▍  | 268/363 [00:07<00:02, 35.98it/s]
Loading 0:  75%|███████▌  | 273/363 [00:07<00:02, 37.64it/s]
Loading 0:  76%|███████▋  | 277/363 [00:07<00:02, 35.16it/s]
Loading 0:  78%|███████▊  | 282/363 [00:07<00:02, 37.67it/s]
Loading 0:  79%|███████▉  | 286/363 [00:07<00:02, 36.21it/s]
Loading 0:  80%|████████  | 291/363 [00:08<00:01, 38.09it/s]
Loading 0:  81%|████████▏ | 295/363 [00:08<00:01, 36.25it/s]
Loading 0:  82%|████████▏ | 299/363 [00:08<00:01, 35.99it/s]
Loading 0:  84%|████████▎ | 304/363 [00:15<00:28,  2.10it/s]
Loading 0:  85%|████████▍ | 307/363 [00:15<00:21,  2.64it/s]
Loading 0:  86%|████████▌ | 312/363 [00:15<00:13,  3.89it/s]
Loading 0:  88%|████████▊ | 319/363 [00:15<00:06,  6.34it/s]
Loading 0:  89%|████████▉ | 324/363 [00:15<00:04,  8.44it/s]
Loading 0:  91%|█████████ | 329/363 [00:15<00:03, 11.10it/s]
Loading 0:  92%|█████████▏| 334/363 [00:15<00:02, 14.36it/s]
Loading 0:  93%|█████████▎| 339/363 [00:16<00:01, 16.45it/s]
Loading 0:  96%|█████████▌| 347/363 [00:16<00:00, 23.70it/s]
Loading 0:  97%|█████████▋| 352/363 [00:16<00:00, 27.29it/s]
Loading 0:  98%|█████████▊| 357/363 [00:16<00:00, 26.71it/s]
                                                            
Job alexxi19-ft-v1-nemo-base-dpo-v4-mkmlizer completed after 114.29s with status: succeeded
Stopping job with name alexxi19-ft-v1-nemo-base-dpo-v4-mkmlizer
Pipeline stage MKMLizer completed in 114.76s
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 alexxi19-ft-v1-nemo-base-dpo-v4
Waiting for inference service alexxi19-ft-v1-nemo-base-dpo-v4 to be ready
Inference service alexxi19-ft-v1-nemo-base-dpo-v4 ready after 220.9197483062744s
Pipeline stage MKMLDeployer completed in 221.43s
run pipeline stage %s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.3544938564300537s
Received healthy response to inference request in 1.5368635654449463s
Received healthy response to inference request in 1.5674941539764404s
Received healthy response to inference request in 1.5668754577636719s
Received healthy response to inference request in 1.9070746898651123s
5 requests
0 failed requests
5th percentile: 1.5428659439086914
10th percentile: 1.5488683223724364
20th percentile: 1.5608730792999268
30th percentile: 1.5669991970062256
40th percentile: 1.567246675491333
50th percentile: 1.5674941539764404
60th percentile: 1.703326368331909
70th percentile: 1.8391585826873778
80th percentile: 1.9965585231781007
90th percentile: 2.175526189804077
95th percentile: 2.2650100231170653
99th percentile: 2.336597089767456
mean time: 1.786560344696045
Pipeline stage StressChecker completed in 10.15s
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 0.61s
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 0.65s
Shutdown handler de-registered
alexxi19-ft-v1-nemo-base-dpo_v4 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 0.10s
run pipeline stage %s
Running pipeline stage MKMLProfilerTemplater
Pipeline stage MKMLProfilerTemplater completed in 0.09s
run pipeline stage %s
Running pipeline stage MKMLProfilerDeployer
Creating inference service alexxi19-ft-v1-nemo-base-dpo-v4-profiler
Waiting for inference service alexxi19-ft-v1-nemo-base-dpo-v4-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 3909.37s
Shutdown handler de-registered
alexxi19-ft-v1-nemo-base-dpo_v4 status is now inactive due to auto deactivation removed underperforming models
alexxi19-ft-v1-nemo-base-dpo_v4 status is now torndown due to DeploymentManager action