jic062-dpo-v1-4

submission_id: jic062-dpo-v1-4_v2

developer_uid: chace9580

alignment_samples: 16822

alignment_score: 0.5143518947284255

best_of: 16

celo_rating: 1242.15

display_name: jic062-dpo-v1-4_v2

formatter: {'memory_template': "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\n{bot_name}'s Persona: {memory}\n\n", 'prompt_template': '{prompt}<|eot_id|>', 'bot_template': '<|start_header_id|>assistant<|end_header_id|>\n\n{bot_name}: {message}<|eot_id|>', 'user_template': '<|start_header_id|>user<|end_header_id|>\n\n{user_name}: {message}<|eot_id|>', 'response_template': '<|start_header_id|>assistant<|end_header_id|>\n\n{bot_name}:', 'truncate_by_message': False}

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', '|eot_id|'], 'max_input_tokens': 512, 'best_of': 16, 'max_output_tokens': 64}

gpu_counts: {'NVIDIA RTX A5000': 1}

is_internal_developer: False

language_model: jic062/dpo-v1.4

latencies: [{'batch_size': 1, 'throughput': 0.9025086717610313, 'latency_mean': 1.107960786819458, 'latency_p50': 1.1087754964828491, 'latency_p90': 1.2297024726867676}, {'batch_size': 3, 'throughput': 1.6267585819747885, 'latency_mean': 1.8371954596042632, 'latency_p50': 1.8272883892059326, 'latency_p90': 2.073454165458679}, {'batch_size': 5, 'throughput': 1.7829504740291724, 'latency_mean': 2.7949455118179323, 'latency_p50': 2.8041346073150635, 'latency_p90': 3.1563604354858397}, {'batch_size': 6, 'throughput': 1.7834603419898771, 'latency_mean': 3.3492347204685213, 'latency_p50': 3.371678352355957, 'latency_p90': 3.7574491024017336}, {'batch_size': 8, 'throughput': 1.7934190592498733, 'latency_mean': 4.43087975859642, 'latency_p50': 4.47354793548584, 'latency_p90': 5.0014440536499025}, {'batch_size': 10, 'throughput': 1.776054366766454, 'latency_mean': 5.586217359304428, 'latency_p50': 5.551625609397888, 'latency_p90': 6.534202933311462}]

max_input_tokens: 512

max_output_tokens: 64

model_architecture: LlamaForCausalLM

model_group: jic062/dpo-v1.4

model_name: jic062-dpo-v1-4_v2

model_num_parameters: 8030261248.0

model_repo: jic062/dpo-v1.4

model_size: 8B

num_battles: 16821

num_wins: 8442

propriety_score: 0.7487821851078637

propriety_total_count: 1437.0

ranking_group: single

status: inactive

submission_type: basic

throughput_3p7s: 1.8

timestamp: 2024-09-08T18:10:32+00:00

us_pacific_date: 2024-09-08

win_ratio: 0.50187265917603

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-v2-mkmlizer
Waiting for job on jic062-dpo-v1-4-v2-mkmlizer to finish
jic062-dpo-v1-4-v2-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
jic062-dpo-v1-4-v2-mkmlizer: ║     _____            __           __                                ║
jic062-dpo-v1-4-v2-mkmlizer: ║    / _/ /_ ___    __/ /  ___ ___ / /                                ║
jic062-dpo-v1-4-v2-mkmlizer: ║   / _/ / // / |/|/ / _ \/ -_) -_) /                                 ║
jic062-dpo-v1-4-v2-mkmlizer: ║  /_//_/\_, /|__,__/_//_/\__/\__/_/                                  ║
jic062-dpo-v1-4-v2-mkmlizer: ║       /___/                                                         ║
jic062-dpo-v1-4-v2-mkmlizer: ║                                                                     ║
jic062-dpo-v1-4-v2-mkmlizer: ║  Version: 0.10.1                                                    ║
jic062-dpo-v1-4-v2-mkmlizer: ║  Copyright 2023 MK ONE TECHNOLOGIES Inc.                            ║
jic062-dpo-v1-4-v2-mkmlizer: ║  https://mk1.ai                                                     ║
jic062-dpo-v1-4-v2-mkmlizer: ║                                                                     ║
jic062-dpo-v1-4-v2-mkmlizer: ║  The license key for the current software has been verified as      ║
jic062-dpo-v1-4-v2-mkmlizer: ║  belonging to:                                                      ║
jic062-dpo-v1-4-v2-mkmlizer: ║                                                                     ║
jic062-dpo-v1-4-v2-mkmlizer: ║  Chai Research Corp.                                                ║
jic062-dpo-v1-4-v2-mkmlizer: ║  Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f                   ║
jic062-dpo-v1-4-v2-mkmlizer: ║  Expiration: 2024-10-15 23:59:59                                    ║
jic062-dpo-v1-4-v2-mkmlizer: ║                                                                     ║
jic062-dpo-v1-4-v2-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
jic062-dpo-v1-4-v2-mkmlizer: Downloaded to shared memory in 20.518s
jic062-dpo-v1-4-v2-mkmlizer: quantizing model to /dev/shm/model_cache, profile:s0, folder:/tmp/tmpc9wfemkr, device:0
jic062-dpo-v1-4-v2-mkmlizer: Saving flywheel model at /dev/shm/model_cache
jic062-dpo-v1-4-v2-mkmlizer: quantized model in 25.914s
jic062-dpo-v1-4-v2-mkmlizer: Processed model jic062/dpo-v1.4 in 46.432s
jic062-dpo-v1-4-v2-mkmlizer: creating bucket guanaco-mkml-models
jic062-dpo-v1-4-v2-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
jic062-dpo-v1-4-v2-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/jic062-dpo-v1-4-v2
jic062-dpo-v1-4-v2-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/jic062-dpo-v1-4-v2/config.json
jic062-dpo-v1-4-v2-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/jic062-dpo-v1-4-v2/special_tokens_map.json
jic062-dpo-v1-4-v2-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/jic062-dpo-v1-4-v2/tokenizer_config.json
jic062-dpo-v1-4-v2-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/jic062-dpo-v1-4-v2/tokenizer.json
jic062-dpo-v1-4-v2-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/jic062-dpo-v1-4-v2/flywheel_model.0.safetensors
jic062-dpo-v1-4-v2-mkmlizer: 
Loading 0:   0%|          | 0/291 [00:00<?, ?it/s]
Loading 0:   2%|▏         | 5/291 [00:00<00:08, 33.86it/s]
Loading 0:   4%|▍         | 13/291 [00:00<00:05, 53.77it/s]
Loading 0:   7%|▋         | 19/291 [00:00<00:05, 49.40it/s]
Loading 0:   9%|▊         | 25/291 [00:00<00:05, 52.81it/s]
Loading 0:  11%|█         | 32/291 [00:00<00:05, 48.92it/s]
Loading 0:  14%|█▍        | 41/291 [00:00<00:04, 51.85it/s]
Loading 0:  17%|█▋        | 50/291 [00:00<00:04, 53.20it/s]
Loading 0:  20%|██        | 59/291 [00:01<00:04, 52.31it/s]
Loading 0:  23%|██▎       | 68/291 [00:01<00:04, 51.43it/s]
Loading 0:  26%|██▌       | 75/291 [00:01<00:03, 54.58it/s]
Loading 0:  28%|██▊       | 81/291 [00:01<00:03, 54.39it/s]
Loading 0:  30%|██▉       | 87/291 [00:01<00:05, 35.86it/s]
Loading 0:  32%|███▏      | 93/291 [00:01<00:04, 40.21it/s]
Loading 0:  34%|███▍      | 99/291 [00:02<00:04, 43.90it/s]
Loading 0:  36%|███▌      | 105/291 [00:02<00:04, 42.01it/s]
Loading 0:  38%|███▊      | 112/291 [00:02<00:03, 46.55it/s]
Loading 0:  41%|████      | 118/291 [00:02<00:03, 45.58it/s]
Loading 0:  42%|████▏     | 123/291 [00:02<00:03, 44.70it/s]
Loading 0:  45%|████▍     | 130/291 [00:02<00:03, 49.81it/s]
Loading 0:  47%|████▋     | 136/291 [00:02<00:03, 48.47it/s]
Loading 0:  49%|████▉     | 142/291 [00:02<00:03, 48.90it/s]
Loading 0:  51%|█████     | 148/291 [00:03<00:02, 51.05it/s]
Loading 0:  53%|█████▎    | 154/291 [00:03<00:02, 46.87it/s]
Loading 0:  55%|█████▍    | 159/291 [00:03<00:02, 46.48it/s]
Loading 0:  57%|█████▋    | 165/291 [00:03<00:02, 48.09it/s]
Loading 0:  58%|█████▊    | 170/291 [00:03<00:02, 46.21it/s]
Loading 0:  61%|██████    | 177/291 [00:03<00:02, 46.55it/s]
Loading 0:  63%|██████▎   | 182/291 [00:03<00:02, 46.14it/s]
Loading 0:  64%|██████▍   | 187/291 [00:04<00:02, 36.18it/s]
Loading 0:  66%|██████▌   | 192/291 [00:04<00:02, 37.99it/s]
Loading 0:  68%|██████▊   | 199/291 [00:04<00:02, 41.13it/s]
Loading 0:  70%|███████   | 204/291 [00:04<00:02, 42.57it/s]
Loading 0:  73%|███████▎  | 211/291 [00:04<00:01, 49.11it/s]
Loading 0:  75%|███████▍  | 217/291 [00:04<00:01, 48.30it/s]
Loading 0:  77%|███████▋  | 223/291 [00:04<00:01, 50.04it/s]
Loading 0:  79%|███████▊  | 229/291 [00:04<00:01, 52.14it/s]
Loading 0:  81%|████████  | 235/291 [00:05<00:01, 48.69it/s]
Loading 0:  82%|████████▏ | 240/291 [00:05<00:01, 47.47it/s]
Loading 0:  85%|████████▍ | 247/291 [00:05<00:00, 52.70it/s]
Loading 0:  87%|████████▋ | 253/291 [00:05<00:00, 49.58it/s]
Loading 0:  89%|████████▉ | 259/291 [00:05<00:00, 48.90it/s]
Loading 0:  91%|█████████ | 264/291 [00:05<00:00, 48.33it/s]
Loading 0:  92%|█████████▏| 269/291 [00:05<00:00, 47.38it/s]
Loading 0:  95%|█████████▍| 275/291 [00:05<00:00, 40.16it/s]
Loading 0:  97%|█████████▋| 282/291 [00:06<00:00, 41.42it/s]
Loading 0:  99%|█████████▊| 287/291 [00:11<00:01,  3.28it/s]
                                                            
Job jic062-dpo-v1-4-v2-mkmlizer completed after 74.42s with status: succeeded
Stopping job with name jic062-dpo-v1-4-v2-mkmlizer
Pipeline stage MKMLizer completed in 75.38s
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-v1-4-v2
Waiting for inference service jic062-dpo-v1-4-v2 to be ready
Inference service jic062-dpo-v1-4-v2 ready after 150.81845259666443s
Pipeline stage MKMLDeployer completed in 151.55s
run pipeline stage %s
Running pipeline stage StressChecker
Received healthy response to inference request in 1.8263318538665771s
Received healthy response to inference request in 1.560359239578247s
Received healthy response to inference request in 1.8154346942901611s
Received healthy response to inference request in 1.6394627094268799s
Received healthy response to inference request in 1.769486904144287s
5 requests
0 failed requests
5th percentile: 1.5761799335479736
10th percentile: 1.5920006275177
20th percentile: 1.6236420154571534
30th percentile: 1.6654675483703614
40th percentile: 1.7174772262573241
50th percentile: 1.769486904144287
60th percentile: 1.7878660202026366
70th percentile: 1.8062451362609864
80th percentile: 1.8176141262054444
90th percentile: 1.8219729900360107
95th percentile: 1.824152421951294
99th percentile: 1.8258959674835205
mean time: 1.7222150802612304
Pipeline stage StressChecker completed in 9.17s
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 5.48s
Shutdown handler de-registered
jic062-dpo-v1-4_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 0.12s
run pipeline stage %s
Running pipeline stage MKMLProfilerTemplater
Pipeline stage MKMLProfilerTemplater completed in 0.12s
run pipeline stage %s
Running pipeline stage MKMLProfilerDeployer
Creating inference service jic062-dpo-v1-4-v2-profiler
Waiting for inference service jic062-dpo-v1-4-v2-profiler to be ready
Inference service jic062-dpo-v1-4-v2-profiler ready after 150.33110165596008s
Pipeline stage MKMLProfilerDeployer completed in 150.68s
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-v2-profiler-predictor-00001-deployment-65bzlhdq:/code/chaiverse_profiler_1725819466 --namespace tenant-chaiml-guanaco
kubectl exec -it jic062-dpo-v1-4-v2-profiler-predictor-00001-deployment-65bzlhdq --namespace tenant-chaiml-guanaco -- sh -c 'cd /code/chaiverse_profiler_1725819466 && python profiles.py profile --best_of_n 16 --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_1725819466/summary.json'
kubectl exec -it jic062-dpo-v1-4-v2-profiler-predictor-00001-deployment-65bzlhdq --namespace tenant-chaiml-guanaco -- bash -c 'cat /code/chaiverse_profiler_1725819466/summary.json'
Pipeline stage MKMLProfilerRunner completed in 799.62s
run pipeline stage %s
Running pipeline stage MKMLProfilerDeleter
Checking if service jic062-dpo-v1-4-v2-profiler is running
Tearing down inference service jic062-dpo-v1-4-v2-profiler
Service jic062-dpo-v1-4-v2-profiler has been torndown
Pipeline stage MKMLProfilerDeleter completed in 1.74s
Shutdown handler de-registered
jic062-dpo-v1-4_v2 status is now inactive due to auto deactivation removed underperforming models