jic062-dpo-v1-8

developer_uid: chace9580

submission_id: jic062-dpo-v1-8_v2

model_name: jic062-dpo-v1-8_v2

model_group: jic062/dpo-v1.8

status: torndown

timestamp: 2024-09-25T03:39:27+00:00

num_battles: 3141

num_wins: 1578

celo_rating: 1252.16

family_friendly_score: 0.5641447368421053

family_friendly_standard_error: 0.008973914464603881

submission_type: basic

model_repo: jic062/dpo-v1.8

model_architecture: MistralForCausalLM

model_num_parameters: 12772070400.0

best_of: 8

max_input_tokens: 1024

max_output_tokens: 64

latencies: [{'batch_size': 1, 'throughput': 0.615699297781844, 'latency_mean': 1.62407443523407, 'latency_p50': 1.6246799230575562, 'latency_p90': 1.7805296897888183}, {'batch_size': 3, 'throughput': 1.0844477041132443, 'latency_mean': 2.7628300309181215, 'latency_p50': 2.778220534324646, 'latency_p90': 3.009187412261963}, {'batch_size': 5, 'throughput': 1.2318620415216939, 'latency_mean': 4.042426750659943, 'latency_p50': 4.0553590059280396, 'latency_p90': 4.690313696861267}, {'batch_size': 6, 'throughput': 1.2529397065722738, 'latency_mean': 4.7585270369052886, 'latency_p50': 4.765086531639099, 'latency_p90': 5.3423206329345705}, {'batch_size': 8, 'throughput': 1.2467363860836933, 'latency_mean': 6.373973944187164, 'latency_p50': 6.375430226325989, 'latency_p90': 7.3303824186325075}, {'batch_size': 10, 'throughput': 1.2060351207368953, 'latency_mean': 8.235136895179748, 'latency_p50': 8.130862593650818, 'latency_p90': 9.382778072357178}]

gpu_counts: {'NVIDIA RTX A5000': 1}

display_name: jic062-dpo-v1-8_v2

ineligible_reason: num_battles<5000

is_internal_developer: False

language_model: jic062/dpo-v1.8

model_size: 13B

ranking_group: single

throughput_3p7s: 1.21

us_pacific_date: 2024-09-24

win_ratio: 0.5023877745940784

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', '[/INST]'], 'max_input_tokens': 1024, 'best_of': 8, 'max_output_tokens': 64}

formatter: {'memory_template': '[INST]system\n{memory}[/INST]\n', 'prompt_template': '[INST]user\n{prompt}[/INST]\n', 'bot_template': '[INST]assistant\n{bot_name}: {message}[/INST]\n', 'user_template': '[INST]user\n{user_name}: {message}[/INST]\n', 'response_template': '[INST]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 jic062-dpo-v1-8-v2-mkmlizer
Waiting for job on jic062-dpo-v1-8-v2-mkmlizer to finish
jic062-dpo-v1-8-v2-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
jic062-dpo-v1-8-v2-mkmlizer: ║     _____            __           __                                ║
jic062-dpo-v1-8-v2-mkmlizer: ║    / _/ /_ ___    __/ /  ___ ___ / /                                ║
jic062-dpo-v1-8-v2-mkmlizer: ║   / _/ / // / |/|/ / _ \/ -_) -_) /                                 ║
jic062-dpo-v1-8-v2-mkmlizer: ║  /_//_/\_, /|__,__/_//_/\__/\__/_/                                  ║
jic062-dpo-v1-8-v2-mkmlizer: ║       /___/                                                         ║
jic062-dpo-v1-8-v2-mkmlizer: ║                                                                     ║
jic062-dpo-v1-8-v2-mkmlizer: ║  Version: 0.11.12                                                   ║
jic062-dpo-v1-8-v2-mkmlizer: ║  Copyright 2023 MK ONE TECHNOLOGIES Inc.                            ║
jic062-dpo-v1-8-v2-mkmlizer: ║  https://mk1.ai                                                     ║
jic062-dpo-v1-8-v2-mkmlizer: ║                                                                     ║
jic062-dpo-v1-8-v2-mkmlizer: ║  The license key for the current software has been verified as      ║
jic062-dpo-v1-8-v2-mkmlizer: ║  belonging to:                                                      ║
jic062-dpo-v1-8-v2-mkmlizer: ║                                                                     ║
jic062-dpo-v1-8-v2-mkmlizer: ║  Chai Research Corp.                                                ║
jic062-dpo-v1-8-v2-mkmlizer: ║  Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f                   ║
jic062-dpo-v1-8-v2-mkmlizer: ║  Expiration: 2024-10-15 23:59:59                                    ║
jic062-dpo-v1-8-v2-mkmlizer: ║                                                                     ║
jic062-dpo-v1-8-v2-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
Connection pool is full, discarding connection: %s. Connection pool size: %s
jic062-dpo-v1-8-v2-mkmlizer: Downloaded to shared memory in 29.506s
jic062-dpo-v1-8-v2-mkmlizer: quantizing model to /dev/shm/model_cache, profile:s0, folder:/tmp/tmpo68xvlhp, device:0
jic062-dpo-v1-8-v2-mkmlizer: Saving flywheel model at /dev/shm/model_cache
jic062-dpo-v1-8-v2-mkmlizer: quantized model in 35.379s
jic062-dpo-v1-8-v2-mkmlizer: Processed model jic062/dpo-v1.8 in 64.885s
jic062-dpo-v1-8-v2-mkmlizer: creating bucket guanaco-mkml-models
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
jic062-dpo-v1-8-v2-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
jic062-dpo-v1-8-v2-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/jic062-dpo-v1-8-v2
jic062-dpo-v1-8-v2-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/jic062-dpo-v1-8-v2/config.json
jic062-dpo-v1-8-v2-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/jic062-dpo-v1-8-v2/special_tokens_map.json
jic062-dpo-v1-8-v2-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/jic062-dpo-v1-8-v2/tokenizer_config.json
jic062-dpo-v1-8-v2-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/jic062-dpo-v1-8-v2/tokenizer.json
jic062-dpo-v1-8-v2-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/jic062-dpo-v1-8-v2/flywheel_model.0.safetensors
jic062-dpo-v1-8-v2-mkmlizer: 
Loading 0:   0%|          | 0/363 [00:00<?, ?it/s]
Loading 0:   1%|▏         | 5/363 [00:00<00:11, 32.12it/s]
Loading 0:   4%|▎         | 13/363 [00:00<00:06, 53.12it/s]
Loading 0:   5%|▌         | 19/363 [00:00<00:06, 50.30it/s]
Loading 0:   7%|▋         | 25/363 [00:00<00:06, 50.59it/s]
Loading 0:   9%|▊         | 31/363 [00:00<00:06, 52.87it/s]
Loading 0:  10%|█         | 37/363 [00:00<00:06, 50.25it/s]
Loading 0:  12%|█▏        | 43/363 [00:00<00:06, 50.18it/s]
Loading 0:  13%|█▎        | 49/363 [00:00<00:06, 51.74it/s]
Loading 0:  15%|█▌        | 55/363 [00:01<00:06, 49.59it/s]
Loading 0:  17%|█▋        | 61/363 [00:01<00:08, 35.15it/s]
Loading 0:  18%|█▊        | 66/363 [00:01<00:08, 35.95it/s]
Loading 0:  20%|█▉        | 72/363 [00:01<00:07, 40.71it/s]
Loading 0:  21%|██▏       | 78/363 [00:01<00:06, 41.67it/s]
Loading 0:  23%|██▎       | 83/363 [00:01<00:06, 41.85it/s]
Loading 0:  25%|██▍       | 90/363 [00:01<00:05, 47.64it/s]
Loading 0:  26%|██▋       | 96/363 [00:02<00:05, 46.69it/s]
Loading 0:  28%|██▊       | 101/363 [00:02<00:05, 45.10it/s]
Loading 0:  29%|██▉       | 106/363 [00:02<00:05, 46.02it/s]
Loading 0:  31%|███       | 113/363 [00:02<00:05, 44.36it/s]
Loading 0:  33%|███▎      | 118/363 [00:02<00:05, 44.09it/s]
Loading 0:  34%|███▍      | 125/363 [00:02<00:04, 50.55it/s]
Loading 0:  36%|███▌      | 131/363 [00:02<00:04, 51.84it/s]
Loading 0:  38%|███▊      | 137/363 [00:02<00:04, 46.23it/s]
Loading 0:  39%|███▉      | 142/363 [00:03<00:06, 34.55it/s]
Loading 0:  40%|████      | 146/363 [00:03<00:06, 35.27it/s]
Loading 0:  41%|████▏     | 150/363 [00:03<00:06, 34.72it/s]
Loading 0:  43%|████▎     | 157/363 [00:03<00:05, 40.94it/s]
Loading 0:  45%|████▍     | 163/363 [00:03<00:04, 41.00it/s]
Loading 0:  46%|████▋     | 168/363 [00:03<00:04, 41.36it/s]
Loading 0:  48%|████▊     | 175/363 [00:03<00:04, 46.69it/s]
Loading 0:  50%|████▉     | 181/363 [00:04<00:04, 45.42it/s]
Loading 0:  51%|█████     | 186/363 [00:04<00:03, 44.25it/s]
Loading 0:  53%|█████▎    | 193/363 [00:04<00:03, 48.52it/s]
Loading 0:  55%|█████▍    | 199/363 [00:04<00:03, 46.50it/s]
Loading 0:  56%|█████▌    | 204/363 [00:04<00:03, 44.77it/s]
Loading 0:  58%|█████▊    | 211/363 [00:04<00:03, 48.58it/s]
Loading 0:  60%|█████▉    | 217/363 [00:04<00:03, 47.19it/s]
Loading 0:  61%|██████▏   | 223/363 [00:05<00:03, 35.39it/s]
Loading 0:  63%|██████▎   | 227/363 [00:05<00:03, 35.87it/s]
Loading 0:  64%|██████▎   | 231/363 [00:05<00:03, 35.04it/s]
Loading 0:  65%|██████▌   | 237/363 [00:05<00:03, 40.36it/s]
Loading 0:  67%|██████▋   | 242/363 [00:05<00:02, 42.56it/s]
Loading 0:  68%|██████▊   | 247/363 [00:05<00:02, 44.43it/s]
Loading 0:  70%|██████▉   | 253/363 [00:05<00:02, 44.29it/s]
Loading 0:  71%|███████   | 258/363 [00:05<00:02, 42.91it/s]
Loading 0:  73%|███████▎  | 265/363 [00:06<00:02, 48.16it/s]
Loading 0:  75%|███████▍  | 271/363 [00:06<00:01, 46.99it/s]
Loading 0:  76%|███████▌  | 276/363 [00:06<00:01, 45.31it/s]
Loading 0:  78%|███████▊  | 283/363 [00:06<00:01, 49.62it/s]
Loading 0:  80%|███████▉  | 289/363 [00:06<00:01, 46.11it/s]
Loading 0:  81%|████████  | 294/363 [00:06<00:01, 43.72it/s]
Loading 0:  83%|████████▎ | 300/363 [00:06<00:01, 47.72it/s]
Loading 0:  84%|████████▍ | 305/363 [00:13<00:22,  2.61it/s]
Loading 0:  85%|████████▌ | 309/363 [00:13<00:16,  3.34it/s]
Loading 0:  86%|████████▌ | 313/363 [00:13<00:11,  4.32it/s]
Loading 0:  88%|████████▊ | 320/363 [00:13<00:06,  6.76it/s]
Loading 0:  90%|████████▉ | 326/363 [00:14<00:03,  9.30it/s]
Loading 0:  91%|█████████ | 331/363 [00:14<00:02, 11.80it/s]
Loading 0:  93%|█████████▎| 338/363 [00:14<00:01, 16.50it/s]
Loading 0:  95%|█████████▍| 344/363 [00:14<00:00, 20.42it/s]
Loading 0:  96%|█████████▌| 349/363 [00:14<00:00, 23.86it/s]
Loading 0:  98%|█████████▊| 356/363 [00:14<00:00, 30.80it/s]
Loading 0: 100%|█████████▉| 362/363 [00:14<00:00, 34.00it/s]
                                                            
Job jic062-dpo-v1-8-v2-mkmlizer completed after 85.19s with status: succeeded
Stopping job with name jic062-dpo-v1-8-v2-mkmlizer
Connection pool is full, discarding connection: %s. Connection pool size: %s
Connection pool is full, discarding connection: %s. Connection pool size: %s
Pipeline stage MKMLizer completed in 86.39s
run pipeline stage %s
Running pipeline stage MKMLTemplater
Pipeline stage MKMLTemplater completed in 0.09s
run pipeline stage %s
Running pipeline stage MKMLDeployer
Creating inference service jic062-dpo-v1-8-v2
Waiting for inference service jic062-dpo-v1-8-v2 to be ready
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
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
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
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
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
Inference service jic062-dpo-v1-8-v2 ready after 212.9887261390686s
Pipeline stage MKMLDeployer completed in 214.81s
run pipeline stage %s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.777043104171753s
Received healthy response to inference request in 6.517951250076294s
Received healthy response to inference request in 2.059239625930786s
Received healthy response to inference request in 2.575983762741089s
Received healthy response to inference request in 2.2844128608703613s
5 requests
0 failed requests
5th percentile: 2.104274272918701
10th percentile: 2.149308919906616
20th percentile: 2.2393782138824463
30th percentile: 2.342727041244507
40th percentile: 2.4593554019927977
50th percentile: 2.575983762741089
60th percentile: 2.6564074993133544
70th percentile: 2.73683123588562
80th percentile: 3.525224733352662
90th percentile: 5.021587991714478
95th percentile: 5.769769620895385
99th percentile: 6.368314924240112
mean time: 3.2429261207580566
Pipeline stage StressChecker completed in 20.24s
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 6.32s
Shutdown handler de-registered
jic062-dpo-v1-8_v2 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 0.15s
run pipeline stage %s
Running pipeline stage MKMLProfilerTemplater
Pipeline stage MKMLProfilerTemplater completed in 0.16s
run pipeline stage %s
Running pipeline stage MKMLProfilerDeployer
Creating inference service jic062-dpo-v1-8-v2-profiler
Waiting for inference service jic062-dpo-v1-8-v2-profiler to be ready
Inference service jic062-dpo-v1-8-v2-profiler ready after 210.4765236377716s
Pipeline stage MKMLProfilerDeployer completed in 210.87s
run pipeline stage %s
Running pipeline stage MKMLProfilerRunner
kubectl cp /code/guanaco/guanaco_inference_services/src/inference_scripts tenant-chaiml-guanaco/jic062-dpo-v1-8-v2-profiler-predictor-00001-deployment-c75gpg8b:/code/chaiverse_profiler_1727236169 --namespace tenant-chaiml-guanaco
kubectl exec -it jic062-dpo-v1-8-v2-profiler-predictor-00001-deployment-c75gpg8b --namespace tenant-chaiml-guanaco -- sh -c 'cd /code/chaiverse_profiler_1727236169 && 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_1727236169/summary.json'
kubectl exec -it jic062-dpo-v1-8-v2-profiler-predictor-00001-deployment-c75gpg8b --namespace tenant-chaiml-guanaco -- bash -c 'cat /code/chaiverse_profiler_1727236169/summary.json'
Pipeline stage MKMLProfilerRunner completed in 1163.33s
run pipeline stage %s
Running pipeline stage MKMLProfilerDeleter
Checking if service jic062-dpo-v1-8-v2-profiler is running
Tearing down inference service jic062-dpo-v1-8-v2-profiler
Service jic062-dpo-v1-8-v2-profiler has been torndown
Pipeline stage MKMLProfilerDeleter completed in 2.21s
Shutdown handler de-registered
jic062-dpo-v1-8_v2 status is now inactive due to auto deactivation removed underperforming models
Deleting key koboldai-llama2-13b-estopia-v1/tokenizer.json from bucket guanaco-mkml-models
jic062-dpo-v1-8_v2 status is now torndown due to DeploymentManager action