chaiml-albert-dpo-0912-v_2740

submission_id: chaiml-albert-dpo-0912-v_2740_v1

developer_uid: chai_backend_admin

alignment_samples: 11223

alignment_score: -0.8213995903335

best_of: 8

celo_rating: 1248.07

display_name: edit_bottom_model_strength

formatter: {'memory_template': '', 'prompt_template': '', 'bot_template': '{bot_name}: {message}\n', 'user_template': '{user_name}: {message}\n', 'response_template': '{bot_name}:', 'truncate_by_message': False}

generation_params: {'temperature': 0.9, 'top_p': 1.0, 'min_p': 0.05, 'top_k': 80, 'presence_penalty': 0.0, 'frequency_penalty': 0.0, 'stopping_words': ['\n', '</s>', '###', 'Bot:', 'User:', 'You:', '<|im_end|>'], 'max_input_tokens': 1024, 'best_of': 8, 'max_output_tokens': 64}

gpu_counts: {'NVIDIA RTX A5000': 1}

is_internal_developer: True

language_model: ChaiML/albert_dpo_0912_virgo_edit_variety_bottom_model_strength

latencies: [{'batch_size': 1, 'throughput': 0.6149533009295395, 'latency_mean': 1.6260778415203094, 'latency_p50': 1.6193301677703857, 'latency_p90': 1.809686255455017}, {'batch_size': 3, 'throughput': 1.0677133959492433, 'latency_mean': 2.7969468808174134, 'latency_p50': 2.7761799097061157, 'latency_p90': 3.098133111000061}, {'batch_size': 5, 'throughput': 1.2171183129729601, 'latency_mean': 4.093708029985428, 'latency_p50': 4.092479467391968, 'latency_p90': 4.661460828781128}, {'batch_size': 6, 'throughput': 1.2319268679940514, 'latency_mean': 4.84853304028511, 'latency_p50': 4.856054306030273, 'latency_p90': 5.411260485649109}, {'batch_size': 8, 'throughput': 1.2242859512899285, 'latency_mean': 6.498587642908096, 'latency_p50': 6.548216342926025, 'latency_p90': 7.415464234352112}, {'batch_size': 10, 'throughput': 1.1968975678713905, 'latency_mean': 8.311278023719787, 'latency_p50': 8.308359146118164, 'latency_p90': 9.304836773872374}]

max_input_tokens: 1024

max_output_tokens: 64

model_architecture: MistralForCausalLM

model_group: ChaiML/albert_dpo_0912_v

model_name: edit_bottom_model_strength

model_num_parameters: 12772070400.0

model_repo: ChaiML/albert_dpo_0912_virgo_edit_variety_bottom_model_strength

model_size: 13B

num_battles: 11223

num_wins: 5645

propriety_score: 0.7532082922013821

propriety_total_count: 1013.0

ranking_group: single

status: inactive

submission_type: basic

throughput_3p7s: 1.19

timestamp: 2024-09-13T19:18:48+00:00

us_pacific_date: 2024-09-13

win_ratio: 0.5029849416377082

Download Preference Data

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-albert-dpo-0912-v-2740-v1-mkmlizer
Waiting for job on chaiml-albert-dpo-0912-v-2740-v1-mkmlizer to finish
chaiml-albert-dpo-0912-v-2740-v1-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
chaiml-albert-dpo-0912-v-2740-v1-mkmlizer: ║     _____            __           __                                ║
chaiml-albert-dpo-0912-v-2740-v1-mkmlizer: ║    / _/ /_ ___    __/ /  ___ ___ / /                                ║
chaiml-albert-dpo-0912-v-2740-v1-mkmlizer: ║   / _/ / // / |/|/ / _ \/ -_) -_) /                                 ║
chaiml-albert-dpo-0912-v-2740-v1-mkmlizer: ║  /_//_/\_, /|__,__/_//_/\__/\__/_/                                  ║
chaiml-albert-dpo-0912-v-2740-v1-mkmlizer: ║       /___/                                                         ║
chaiml-albert-dpo-0912-v-2740-v1-mkmlizer: ║                                                                     ║
chaiml-albert-dpo-0912-v-2740-v1-mkmlizer: ║  Version: 0.10.1                                                    ║
chaiml-albert-dpo-0912-v-2740-v1-mkmlizer: ║  Copyright 2023 MK ONE TECHNOLOGIES Inc.                            ║
chaiml-albert-dpo-0912-v-2740-v1-mkmlizer: ║  https://mk1.ai                                                     ║
chaiml-albert-dpo-0912-v-2740-v1-mkmlizer: ║                                                                     ║
chaiml-albert-dpo-0912-v-2740-v1-mkmlizer: ║  The license key for the current software has been verified as      ║
chaiml-albert-dpo-0912-v-2740-v1-mkmlizer: ║  belonging to:                                                      ║
chaiml-albert-dpo-0912-v-2740-v1-mkmlizer: ║                                                                     ║
chaiml-albert-dpo-0912-v-2740-v1-mkmlizer: ║  Chai Research Corp.                                                ║
chaiml-albert-dpo-0912-v-2740-v1-mkmlizer: ║  Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f                   ║
chaiml-albert-dpo-0912-v-2740-v1-mkmlizer: ║  Expiration: 2024-10-15 23:59:59                                    ║
chaiml-albert-dpo-0912-v-2740-v1-mkmlizer: ║                                                                     ║
chaiml-albert-dpo-0912-v-2740-v1-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
Connection pool is full, discarding connection: %s. Connection pool size: %s
Connection pool is full, discarding connection: %s. Connection pool size: %s
Connection pool is full, discarding connection: %s. Connection pool size: %s
Connection pool is full, discarding connection: %s. Connection pool size: %s
Connection pool is full, discarding connection: %s. Connection pool size: %s
Connection pool is full, discarding connection: %s. Connection pool size: %s
Connection pool is full, discarding connection: %s. Connection pool size: %s
chaiml-albert-dpo-0912-v-2740-v1-mkmlizer: Downloaded to shared memory in 96.446s
chaiml-albert-dpo-0912-v-2740-v1-mkmlizer: quantizing model to /dev/shm/model_cache, profile:s0, folder:/tmp/tmpp2wp9zo2, device:0
chaiml-albert-dpo-0912-v-2740-v1-mkmlizer: Saving flywheel model at /dev/shm/model_cache
Connection pool is full, discarding connection: %s. Connection pool size: %s
Connection pool is full, discarding connection: %s. Connection pool size: %s
Connection pool is full, discarding connection: %s. Connection pool size: %s
Connection pool is full, discarding connection: %s. Connection pool size: %s
Connection pool is full, discarding connection: %s. Connection pool size: %s
Connection pool is full, discarding connection: %s. Connection pool size: %s
Connection pool is full, discarding connection: %s. Connection pool size: %s
Connection pool is full, discarding connection: %s. Connection pool size: %s
Connection pool is full, discarding connection: %s. Connection pool size: %s
Connection pool is full, discarding connection: %s. Connection pool size: %s
chaiml-albert-dpo-0912-v-2740-v1-mkmlizer: quantized model in 40.336s
chaiml-albert-dpo-0912-v-2740-v1-mkmlizer: Processed model ChaiML/albert_dpo_0912_virgo_edit_variety_bottom_model_strength in 136.782s
chaiml-albert-dpo-0912-v-2740-v1-mkmlizer: creating bucket guanaco-mkml-models
chaiml-albert-dpo-0912-v-2740-v1-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/chaiml-albert-dpo-0912-v-2740-v1/config.json
chaiml-albert-dpo-0912-v-2740-v1-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/chaiml-albert-dpo-0912-v-2740-v1/special_tokens_map.json
chaiml-albert-dpo-0912-v-2740-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/chaiml-albert-dpo-0912-v-2740-v1/tokenizer_config.json
chaiml-albert-dpo-0912-v-2740-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/chaiml-albert-dpo-0912-v-2740-v1/tokenizer.json
chaiml-albert-dpo-0912-v-2740-v1-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/chaiml-albert-dpo-0912-v-2740-v1/flywheel_model.0.safetensors
chaiml-albert-dpo-0912-v-2740-v1-mkmlizer: 
Loading 0:   0%|          | 0/363 [00:00<?, ?it/s]
Loading 0:   1%|▏         | 5/363 [00:00<00:14, 24.19it/s]
Loading 0:   3%|▎         | 10/363 [00:00<00:12, 29.09it/s]
Loading 0:   4%|▍         | 14/363 [00:00<00:13, 25.43it/s]
Loading 0:   6%|▌         | 21/363 [00:00<00:09, 35.87it/s]
Loading 0:   7%|▋         | 25/363 [00:00<00:13, 24.48it/s]
Loading 0:   8%|▊         | 29/363 [00:01<00:12, 26.48it/s]
Loading 0:   9%|▉         | 33/363 [00:01<00:13, 24.75it/s]
Loading 0:  10%|█         | 37/363 [00:01<00:11, 27.61it/s]
Loading 0:  11%|█▏        | 41/363 [00:01<00:13, 24.47it/s]
Loading 0:  13%|█▎        | 48/363 [00:01<00:09, 31.50it/s]
Loading 0:  14%|█▍        | 52/363 [00:01<00:10, 29.78it/s]
Loading 0:  15%|█▌        | 56/363 [00:02<00:10, 28.81it/s]
Loading 0:  17%|█▋        | 61/363 [00:02<00:11, 25.92it/s]
Loading 0:  18%|█▊        | 64/363 [00:02<00:13, 22.94it/s]
Loading 0:  20%|█▉        | 71/363 [00:02<00:09, 30.06it/s]
Loading 0:  21%|██        | 75/363 [00:02<00:09, 29.84it/s]
Loading 0:  22%|██▏       | 79/363 [00:02<00:09, 28.97it/s]
Loading 0:  23%|██▎       | 84/363 [00:03<00:09, 29.99it/s]
Loading 0:  24%|██▍       | 88/363 [00:03<00:09, 28.65it/s]
Loading 0:  26%|██▌       | 93/363 [00:03<00:09, 29.78it/s]
Loading 0:  27%|██▋       | 97/363 [00:03<00:09, 28.85it/s]
Loading 0:  28%|██▊       | 101/363 [00:03<00:10, 24.07it/s]
Loading 0:  29%|██▊       | 104/363 [00:03<00:12, 21.07it/s]
Loading 0:  31%|███       | 111/363 [00:04<00:08, 28.54it/s]
Loading 0:  32%|███▏      | 115/363 [00:04<00:08, 28.07it/s]
Loading 0:  33%|███▎      | 120/363 [00:04<00:07, 30.54it/s]
Loading 0:  34%|███▍      | 124/363 [00:04<00:08, 29.44it/s]
Loading 0:  36%|███▌      | 129/363 [00:04<00:07, 32.16it/s]
Loading 0:  37%|███▋      | 133/363 [00:04<00:07, 30.37it/s]
Loading 0:  38%|███▊      | 137/363 [00:04<00:07, 29.62it/s]
Loading 0:  39%|███▉      | 142/363 [00:05<00:08, 26.39it/s]
Loading 0:  40%|███▉      | 145/363 [00:05<00:08, 24.80it/s]
Loading 0:  41%|████      | 149/363 [00:05<00:08, 23.89it/s]
Loading 0:  43%|████▎     | 156/363 [00:05<00:06, 30.90it/s]
Loading 0:  44%|████▍     | 160/363 [00:05<00:06, 29.79it/s]
Loading 0:  45%|████▌     | 165/363 [00:05<00:06, 32.26it/s]
Loading 0:  47%|████▋     | 169/363 [00:06<00:06, 30.84it/s]
Loading 0:  48%|████▊     | 174/363 [00:06<00:06, 31.39it/s]
Loading 0:  49%|████▉     | 178/363 [00:06<00:06, 29.53it/s]
Loading 0:  50%|█████     | 182/363 [00:06<00:07, 25.02it/s]
Loading 0:  51%|█████     | 185/363 [00:06<00:08, 21.96it/s]
Loading 0:  53%|█████▎    | 192/363 [00:06<00:05, 29.27it/s]
Loading 0:  54%|█████▍    | 196/363 [00:07<00:05, 28.27it/s]
Loading 0:  55%|█████▌    | 201/363 [00:07<00:05, 31.50it/s]
Loading 0:  56%|█████▋    | 205/363 [00:07<00:05, 30.32it/s]
Loading 0:  58%|█████▊    | 210/363 [00:07<00:04, 32.95it/s]
Loading 0:  59%|█████▉    | 214/363 [00:07<00:04, 30.73it/s]
Loading 0:  60%|██████    | 218/363 [00:07<00:04, 31.28it/s]
Loading 0:  61%|██████▏   | 223/363 [00:07<00:05, 27.34it/s]
Loading 0:  62%|██████▏   | 226/363 [00:08<00:05, 25.25it/s]
Loading 0:  63%|██████▎   | 230/363 [00:08<00:05, 23.86it/s]
Loading 0:  65%|██████▌   | 237/363 [00:08<00:04, 31.11it/s]
Loading 0:  66%|██████▋   | 241/363 [00:08<00:04, 29.69it/s]
Loading 0:  68%|██████▊   | 246/363 [00:08<00:03, 32.32it/s]
Loading 0:  69%|██████▉   | 250/363 [00:08<00:03, 31.11it/s]
Loading 0:  70%|███████   | 255/363 [00:08<00:03, 32.77it/s]
Loading 0:  71%|███████▏  | 259/363 [00:09<00:03, 31.36it/s]
Loading 0:  72%|███████▏  | 263/363 [00:09<00:03, 26.61it/s]
Loading 0:  73%|███████▎  | 266/363 [00:09<00:04, 22.81it/s]
Loading 0:  75%|███████▌  | 273/363 [00:09<00:03, 29.87it/s]
Loading 0:  76%|███████▋  | 277/363 [00:09<00:02, 29.11it/s]
Loading 0:  78%|███████▊  | 282/363 [00:09<00:02, 31.84it/s]
Loading 0:  79%|███████▉  | 286/363 [00:10<00:02, 30.54it/s]
Loading 0:  80%|████████  | 291/363 [00:10<00:02, 32.93it/s]
Loading 0:  81%|████████▏ | 295/363 [00:10<00:02, 30.40it/s]
Loading 0:  82%|████████▏ | 299/363 [00:10<00:02, 30.16it/s]
Loading 0:  84%|████████▎ | 304/363 [00:10<00:02, 25.49it/s]
Loading 0:  85%|████████▍ | 307/363 [00:10<00:02, 23.83it/s]
Loading 0:  86%|████████▌ | 311/363 [00:11<00:02, 22.52it/s]
Loading 0:  88%|████████▊ | 318/363 [00:11<00:01, 28.91it/s]
Loading 0:  89%|████████▊ | 322/363 [00:11<00:01, 28.65it/s]
Loading 0:  90%|█████████ | 327/363 [00:11<00:01, 31.87it/s]
Loading 0:  91%|█████████ | 331/363 [00:11<00:01, 30.54it/s]
Loading 0:  93%|█████████▎| 336/363 [00:11<00:00, 33.25it/s]
Loading 0:  94%|█████████▎| 340/363 [00:11<00:00, 31.27it/s]
Loading 0:  95%|█████████▍| 344/363 [00:18<00:09,  2.00it/s]
Loading 0:  96%|█████████▌| 348/363 [00:18<00:05,  2.69it/s]
Loading 0:  97%|█████████▋| 353/363 [00:19<00:02,  3.91it/s]
Loading 0:  98%|█████████▊| 357/363 [00:19<00:01,  5.06it/s]
                                                            
Job chaiml-albert-dpo-0912-v-2740-v1-mkmlizer completed after 168.46s with status: succeeded
Stopping job with name chaiml-albert-dpo-0912-v-2740-v1-mkmlizer
Pipeline stage MKMLizer completed in 169.80s
run pipeline stage %s
Running pipeline stage MKMLTemplater
Pipeline stage MKMLTemplater completed in 0.10s
run pipeline stage %s
Running pipeline stage MKMLDeployer
Creating inference service chaiml-albert-dpo-0912-v-2740-v1
Waiting for inference service chaiml-albert-dpo-0912-v-2740-v1 to be ready
Inference service chaiml-albert-dpo-0912-v-2740-v1 ready after 181.08316946029663s
Pipeline stage MKMLDeployer completed in 181.43s
run pipeline stage %s
Running pipeline stage StressChecker
Received healthy response to inference request in 1.627079725265503s
Received healthy response to inference request in 2.2134618759155273s
Received healthy response to inference request in 1.6130154132843018s
Received healthy response to inference request in 2.402184247970581s
Received healthy response to inference request in 2.9794464111328125s
5 requests
0 failed requests
5th percentile: 1.615828275680542
10th percentile: 1.6186411380767822
20th percentile: 1.6242668628692627
30th percentile: 1.7443561553955078
40th percentile: 1.9789090156555176
50th percentile: 2.2134618759155273
60th percentile: 2.2889508247375487
70th percentile: 2.36443977355957
80th percentile: 2.5176366806030273
90th percentile: 2.74854154586792
95th percentile: 2.8639939785003663
99th percentile: 2.9563559246063233
mean time: 2.167037534713745
Pipeline stage StressChecker completed in 11.82s
run pipeline stage %s
Running pipeline stage TriggerMKMLProfilingPipeline
run_pipeline:run_in_cloud %s
starting trigger_guanaco_pipeline args=%s
Pipeline stage TriggerMKMLProfilingPipeline completed in 5.16s
Shutdown handler de-registered
chaiml-albert-dpo-0912-v_2740_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 0.14s
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-albert-dpo-0912-v-2740-v1-profiler
Waiting for inference service chaiml-albert-dpo-0912-v-2740-v1-profiler to be ready
Inference service chaiml-albert-dpo-0912-v-2740-v1-profiler ready after 170.42109322547913s
Pipeline stage MKMLProfilerDeployer completed in 170.80s
run pipeline stage %s
Running pipeline stage MKMLProfilerRunner
kubectl cp /code/guanaco/guanaco_inference_services/src/inference_scripts tenant-chaiml-guanaco/chaiml-albert-dpo-09d387afe9a2014e8f9ccb383143d5693a-deplovlfnl:/code/chaiverse_profiler_1726255705 --namespace tenant-chaiml-guanaco
kubectl exec -it chaiml-albert-dpo-09d387afe9a2014e8f9ccb383143d5693a-deplovlfnl --namespace tenant-chaiml-guanaco -- sh -c 'cd /code/chaiverse_profiler_1726255705 && python profiles.py profile --best_of_n 8 --auto_batch 5 --batches 1,5,10,15,20,25,30,35,40,45,50,55,60,65,70,75,80,85,90,95,100,105,110,115,120,125,130,135,140,145,150,155,160,165,170,175,180,185,190,195 --samples 200 --input_tokens 1024 --output_tokens 64 --summary /code/chaiverse_profiler_1726255705/summary.json'
kubectl exec -it chaiml-albert-dpo-09d387afe9a2014e8f9ccb383143d5693a-deplovlfnl --namespace tenant-chaiml-guanaco -- bash -c 'cat /code/chaiverse_profiler_1726255705/summary.json'
Pipeline stage MKMLProfilerRunner completed in 1174.26s
run pipeline stage %s
Running pipeline stage MKMLProfilerDeleter
Checking if service chaiml-albert-dpo-0912-v-2740-v1-profiler is running
Tearing down inference service chaiml-albert-dpo-0912-v-2740-v1-profiler
Service chaiml-albert-dpo-0912-v-2740-v1-profiler has been torndown
Pipeline stage MKMLProfilerDeleter completed in 2.42s
Shutdown handler de-registered
chaiml-albert-dpo-0912-v_2740_v1 status is now inactive due to auto deactivation removed underperforming models