jic062-dpo-v1-4-nemo-c500

submission_id: jic062-dpo-v1-4-nemo-c500_v1

developer_uid: chace9580

alignment_samples: 14437

alignment_score: -0.993831830782994

best_of: 8

celo_rating: 1251.22

display_name: jic062-dpo-v1-4-nemo-c500_v1

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}

generation_params: {'temperature': 0.75, 'top_p': 1.0, 'min_p': 0.1, 'top_k': 40, 'presence_penalty': 0.0, 'frequency_penalty': 0.0, 'stopping_words': ['\n', '[/INST]'], 'max_input_tokens': 512, 'best_of': 8, 'max_output_tokens': 64}

gpu_counts: {'NVIDIA RTX A5000': 1}

is_internal_developer: False

language_model: jic062/dpo-v1.4-Nemo-c500

latencies: [{'batch_size': 1, 'throughput': 0.6961450887536549, 'latency_mean': 1.4363904452323915, 'latency_p50': 1.4281710386276245, 'latency_p90': 1.6013994216918945}, {'batch_size': 3, 'throughput': 1.318350721089663, 'latency_mean': 2.2681731367111206, 'latency_p50': 2.2731012105941772, 'latency_p90': 2.530795192718506}, {'batch_size': 5, 'throughput': 1.5404597220970047, 'latency_mean': 3.2276690924167633, 'latency_p50': 3.2049684524536133, 'latency_p90': 3.6570931911468505}, {'batch_size': 6, 'throughput': 1.6204383147979933, 'latency_mean': 3.6843768739700318, 'latency_p50': 3.695119261741638, 'latency_p90': 4.225161528587341}, {'batch_size': 8, 'throughput': 1.5728174563994215, 'latency_mean': 5.042973309755325, 'latency_p50': 5.074453353881836, 'latency_p90': 5.692192101478577}, {'batch_size': 10, 'throughput': 1.5166059140497983, 'latency_mean': 6.557505003213882, 'latency_p50': 6.628609657287598, 'latency_p90': 7.369110774993897}]

max_input_tokens: 512

max_output_tokens: 64

model_architecture: MistralForCausalLM

model_group: jic062/dpo-v1.4-Nemo-c50

model_name: jic062-dpo-v1-4-nemo-c500_v1

model_num_parameters: 12772070400.0

model_repo: jic062/dpo-v1.4-Nemo-c500

model_size: 13B

num_battles: 14433

num_wins: 7347

propriety_score: 0.7063829787234043

propriety_total_count: 1175.0

ranking_group: single

status: inactive

submission_type: basic

throughput_3p7s: 1.63

timestamp: 2024-09-13T03:38:21+00:00

us_pacific_date: 2024-09-12

win_ratio: 0.5090417792558719

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 jic062-dpo-v1-4-nemo-c500-v1-mkmlizer
Waiting for job on jic062-dpo-v1-4-nemo-c500-v1-mkmlizer to finish
Failed to get response for submission riverise-0912-1056-sft-9k_v2: ('http://riverise-0912-1056-sft-9k-v2-predictor.tenant-chaiml-guanaco.k.chaiverse.com/v1/models/GPT-J-6B-lit-v2:predict', 'read tcp 127.0.0.1:38114->127.0.0.1:8080: read: connection reset by peer\n')
jic062-dpo-v1-4-nemo-c500-v1-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
jic062-dpo-v1-4-nemo-c500-v1-mkmlizer: ║     _____            __           __                                ║
jic062-dpo-v1-4-nemo-c500-v1-mkmlizer: ║    / _/ /_ ___    __/ /  ___ ___ / /                                ║
jic062-dpo-v1-4-nemo-c500-v1-mkmlizer: ║   / _/ / // / |/|/ / _ \/ -_) -_) /                                 ║
jic062-dpo-v1-4-nemo-c500-v1-mkmlizer: ║  /_//_/\_, /|__,__/_//_/\__/\__/_/                                  ║
jic062-dpo-v1-4-nemo-c500-v1-mkmlizer: ║       /___/                                                         ║
jic062-dpo-v1-4-nemo-c500-v1-mkmlizer: ║                                                                     ║
jic062-dpo-v1-4-nemo-c500-v1-mkmlizer: ║  Version: 0.10.1                                                    ║
jic062-dpo-v1-4-nemo-c500-v1-mkmlizer: ║  Copyright 2023 MK ONE TECHNOLOGIES Inc.                            ║
jic062-dpo-v1-4-nemo-c500-v1-mkmlizer: ║  https://mk1.ai                                                     ║
jic062-dpo-v1-4-nemo-c500-v1-mkmlizer: ║                                                                     ║
jic062-dpo-v1-4-nemo-c500-v1-mkmlizer: ║  The license key for the current software has been verified as      ║
jic062-dpo-v1-4-nemo-c500-v1-mkmlizer: ║  belonging to:                                                      ║
jic062-dpo-v1-4-nemo-c500-v1-mkmlizer: ║                                                                     ║
jic062-dpo-v1-4-nemo-c500-v1-mkmlizer: ║  Chai Research Corp.                                                ║
jic062-dpo-v1-4-nemo-c500-v1-mkmlizer: ║  Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f                   ║
jic062-dpo-v1-4-nemo-c500-v1-mkmlizer: ║  Expiration: 2024-10-15 23:59:59                                    ║
jic062-dpo-v1-4-nemo-c500-v1-mkmlizer: ║                                                                     ║
jic062-dpo-v1-4-nemo-c500-v1-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
jic062-dpo-v1-4-nemo-c500-v1-mkmlizer: Downloaded to shared memory in 44.791s
jic062-dpo-v1-4-nemo-c500-v1-mkmlizer: quantizing model to /dev/shm/model_cache, profile:s0, folder:/tmp/tmp4kydct6e, device:0
jic062-dpo-v1-4-nemo-c500-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
jic062-dpo-v1-4-nemo-c500-v1-mkmlizer: quantized model in 36.553s
jic062-dpo-v1-4-nemo-c500-v1-mkmlizer: Processed model jic062/dpo-v1.4-Nemo-c500 in 81.345s
jic062-dpo-v1-4-nemo-c500-v1-mkmlizer: creating bucket guanaco-mkml-models
jic062-dpo-v1-4-nemo-c500-v1-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
jic062-dpo-v1-4-nemo-c500-v1-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/jic062-dpo-v1-4-nemo-c500-v1
jic062-dpo-v1-4-nemo-c500-v1-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/jic062-dpo-v1-4-nemo-c500-v1/config.json
jic062-dpo-v1-4-nemo-c500-v1-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/jic062-dpo-v1-4-nemo-c500-v1/special_tokens_map.json
jic062-dpo-v1-4-nemo-c500-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/jic062-dpo-v1-4-nemo-c500-v1/tokenizer_config.json
jic062-dpo-v1-4-nemo-c500-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/jic062-dpo-v1-4-nemo-c500-v1/tokenizer.json
jic062-dpo-v1-4-nemo-c500-v1-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/jic062-dpo-v1-4-nemo-c500-v1/flywheel_model.0.safetensors
jic062-dpo-v1-4-nemo-c500-v1-mkmlizer: 
Loading 0:   0%|          | 0/363 [00:00<?, ?it/s]
Loading 0:   1%|▏         | 5/363 [00:00<00:11, 32.08it/s]
Loading 0:   4%|▎         | 13/363 [00:00<00:06, 51.35it/s]
Loading 0:   5%|▌         | 19/363 [00:00<00:08, 42.94it/s]
Loading 0:   7%|▋         | 24/363 [00:00<00:08, 40.74it/s]
Loading 0:   9%|▊         | 31/363 [00:00<00:07, 46.06it/s]
Loading 0:  10%|█         | 37/363 [00:00<00:07, 45.75it/s]
Loading 0:  12%|█▏        | 42/363 [00:00<00:07, 44.65it/s]
Loading 0:  13%|█▎        | 49/363 [00:01<00:06, 50.96it/s]
Loading 0:  15%|█▌        | 55/363 [00:01<00:06, 48.94it/s]
Loading 0:  17%|█▋        | 61/363 [00:01<00:08, 33.70it/s]
Loading 0:  18%|█▊        | 66/363 [00:01<00:08, 33.34it/s]
Loading 0:  20%|█▉        | 71/363 [00:01<00:08, 35.30it/s]
Loading 0:  21%|██        | 75/363 [00:01<00:08, 33.77it/s]
Loading 0:  22%|██▏       | 80/363 [00:02<00:07, 36.12it/s]
Loading 0:  23%|██▎       | 84/363 [00:02<00:07, 34.97it/s]
Loading 0:  25%|██▍       | 89/363 [00:02<00:07, 38.01it/s]
Loading 0:  26%|██▌       | 93/363 [00:02<00:07, 36.76it/s]
Loading 0:  27%|██▋       | 98/363 [00:02<00:06, 38.76it/s]
Loading 0:  28%|██▊       | 102/363 [00:02<00:07, 36.72it/s]
Loading 0:  29%|██▉       | 106/363 [00:02<00:07, 36.09it/s]
Loading 0:  31%|███       | 112/363 [00:02<00:06, 40.46it/s]
Loading 0:  32%|███▏      | 117/363 [00:02<00:06, 38.86it/s]
Loading 0:  34%|███▍      | 123/363 [00:03<00:06, 39.51it/s]
Loading 0:  35%|███▍      | 127/363 [00:03<00:06, 36.94it/s]
Loading 0:  37%|███▋      | 134/363 [00:03<00:05, 44.09it/s]
Loading 0:  38%|███▊      | 139/363 [00:03<00:05, 43.41it/s]
Loading 0:  40%|███▉      | 144/363 [00:03<00:07, 27.83it/s]
Loading 0:  41%|████      | 149/363 [00:03<00:07, 29.23it/s]
Loading 0:  43%|████▎     | 157/363 [00:04<00:05, 37.64it/s]
Loading 0:  45%|████▍     | 162/363 [00:04<00:05, 39.50it/s]
Loading 0:  46%|████▌     | 167/363 [00:04<00:05, 33.39it/s]
Loading 0:  48%|████▊     | 174/363 [00:04<00:04, 39.93it/s]
Loading 0:  49%|████▉     | 179/363 [00:04<00:04, 39.91it/s]
Loading 0:  51%|█████     | 184/363 [00:04<00:04, 39.79it/s]
Loading 0:  52%|█████▏    | 189/363 [00:04<00:04, 41.77it/s]
Loading 0:  53%|█████▎    | 194/363 [00:05<00:04, 35.17it/s]
Loading 0:  55%|█████▌    | 201/363 [00:05<00:03, 40.80it/s]
Loading 0:  57%|█████▋    | 206/363 [00:05<00:03, 41.03it/s]
Loading 0:  58%|█████▊    | 211/363 [00:05<00:03, 42.53it/s]
Loading 0:  60%|█████▉    | 216/363 [00:05<00:03, 44.18it/s]
Loading 0:  61%|██████    | 222/363 [00:05<00:03, 42.71it/s]
Loading 0:  63%|██████▎   | 227/363 [00:05<00:04, 31.69it/s]
Loading 0:  64%|██████▎   | 231/363 [00:06<00:04, 31.56it/s]
Loading 0:  65%|██████▌   | 237/363 [00:06<00:03, 35.81it/s]
Loading 0:  66%|██████▋   | 241/363 [00:06<00:03, 34.14it/s]
Loading 0:  68%|██████▊   | 246/363 [00:06<00:03, 36.12it/s]
Loading 0:  69%|██████▉   | 250/363 [00:06<00:03, 35.07it/s]
Loading 0:  70%|███████   | 255/363 [00:06<00:02, 37.30it/s]
Loading 0:  71%|███████▏  | 259/363 [00:06<00:02, 35.99it/s]
Loading 0:  73%|███████▎  | 264/363 [00:06<00:02, 37.63it/s]
Loading 0:  74%|███████▍  | 268/363 [00:07<00:02, 37.64it/s]
Loading 0:  75%|███████▌  | 273/363 [00:07<00:02, 40.00it/s]
Loading 0:  77%|███████▋  | 278/363 [00:07<00:02, 40.38it/s]
Loading 0:  78%|███████▊  | 283/363 [00:07<00:01, 40.31it/s]
Loading 0:  79%|███████▉  | 288/363 [00:07<00:01, 42.56it/s]
Loading 0:  81%|████████  | 293/363 [00:07<00:01, 37.22it/s]
Loading 0:  82%|████████▏ | 299/363 [00:07<00:01, 41.76it/s]
Loading 0:  84%|████████▎ | 304/363 [00:14<00:24,  2.43it/s]
Loading 0:  85%|████████▍ | 308/363 [00:14<00:17,  3.17it/s]
Loading 0:  86%|████████▌ | 312/363 [00:14<00:12,  4.15it/s]
Loading 0:  88%|████████▊ | 319/363 [00:14<00:06,  6.60it/s]
Loading 0:  89%|████████▉ | 324/363 [00:15<00:04,  8.73it/s]
Loading 0:  91%|█████████ | 329/363 [00:15<00:02, 11.36it/s]
Loading 0:  92%|█████████▏| 334/363 [00:15<00:01, 14.66it/s]
Loading 0:  93%|█████████▎| 339/363 [00:15<00:01, 16.97it/s]
Loading 0:  95%|█████████▌| 346/363 [00:15<00:00, 23.43it/s]
Loading 0:  97%|█████████▋| 351/363 [00:15<00:00, 26.97it/s]
Loading 0:  98%|█████████▊| 357/363 [00:15<00:00, 29.00it/s]
                                                            
Job jic062-dpo-v1-4-nemo-c500-v1-mkmlizer completed after 107.36s with status: succeeded
Stopping job with name jic062-dpo-v1-4-nemo-c500-v1-mkmlizer
Pipeline stage MKMLizer completed in 108.28s
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 jic062-dpo-v1-4-nemo-c500-v1
Waiting for inference service jic062-dpo-v1-4-nemo-c500-v1 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
Inference service jic062-dpo-v1-4-nemo-c500-v1 ready after 180.70353364944458s
Pipeline stage MKMLDeployer completed in 181.06s
run pipeline stage %s
Running pipeline stage StressChecker
Received healthy response to inference request in 3.8186542987823486s
Received healthy response to inference request in 2.4412174224853516s
Received healthy response to inference request in 2.20790696144104s
Received healthy response to inference request in 1.9536373615264893s
Received healthy response to inference request in 2.497779607772827s
5 requests
0 failed requests
5th percentile: 2.0044912815093996
10th percentile: 2.0553452014923095
20th percentile: 2.1570530414581297
30th percentile: 2.2545690536499023
40th percentile: 2.347893238067627
50th percentile: 2.4412174224853516
60th percentile: 2.4638422966003417
70th percentile: 2.486467170715332
80th percentile: 2.7619545459747314
90th percentile: 3.29030442237854
95th percentile: 3.5544793605804443
99th percentile: 3.7658193111419678
mean time: 2.5838391304016115
Pipeline stage StressChecker completed in 13.95s
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 6.81s
Shutdown handler de-registered
jic062-dpo-v1-4-nemo-c500_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 0.11s
run pipeline stage %s
Running pipeline stage MKMLProfilerTemplater
Pipeline stage MKMLProfilerTemplater completed in 0.10s
run pipeline stage %s
Running pipeline stage MKMLProfilerDeployer
Creating inference service jic062-dpo-v1-4-nemo-c500-v1-profiler
Waiting for inference service jic062-dpo-v1-4-nemo-c500-v1-profiler to be ready
Inference service jic062-dpo-v1-4-nemo-c500-v1-profiler ready after 180.4048879146576s
Pipeline stage MKMLProfilerDeployer completed in 180.77s
run pipeline stage %s
Running pipeline stage MKMLProfilerRunner
kubectl cp /code/guanaco/guanaco_inference_services/src/inference_scripts tenant-chaiml-guanaco/jic062-dpo-v1-4-nemo1cefeab1464c6a46f0576e049064de69-deplo4crps:/code/chaiverse_profiler_1726199237 --namespace tenant-chaiml-guanaco
kubectl exec -it jic062-dpo-v1-4-nemo1cefeab1464c6a46f0576e049064de69-deplo4crps --namespace tenant-chaiml-guanaco -- sh -c 'cd /code/chaiverse_profiler_1726199237 && 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 512 --output_tokens 64 --summary /code/chaiverse_profiler_1726199237/summary.json'
kubectl exec -it jic062-dpo-v1-4-nemo1cefeab1464c6a46f0576e049064de69-deplo4crps --namespace tenant-chaiml-guanaco -- bash -c 'cat /code/chaiverse_profiler_1726199237/summary.json'
Pipeline stage MKMLProfilerRunner completed in 957.84s
run pipeline stage %s
Running pipeline stage MKMLProfilerDeleter
Checking if service jic062-dpo-v1-4-nemo-c500-v1-profiler is running
Tearing down inference service jic062-dpo-v1-4-nemo-c500-v1-profiler
Service jic062-dpo-v1-4-nemo-c500-v1-profiler has been torndown
Pipeline stage MKMLProfilerDeleter completed in 2.03s
Shutdown handler de-registered
jic062-dpo-v1-4-nemo-c500_v1 status is now inactive due to auto deactivation removed underperforming models