jic062-dpo-v1-5

submission_id: jic062-dpo-v1-5_v1

developer_uid: chace9580

alignment_samples: 12254

alignment_score: 0.2971390442947388

best_of: 16

celo_rating: 1244.73

display_name: jic062-dpo-v1-5_v1

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': 0.01, 'top_k': 80, 'presence_penalty': 0.0, 'frequency_penalty': 0.0, 'stopping_words': ['\n', '<|end_of_text|>', '|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-v1.5

latencies: [{'batch_size': 1, 'throughput': 0.9154971736919261, 'latency_mean': 1.0922110319137572, 'latency_p50': 1.0889726877212524, 'latency_p90': 1.2289143562316895}, {'batch_size': 3, 'throughput': 1.5745264558858147, 'latency_mean': 1.8991196370124817, 'latency_p50': 1.907800316810608, 'latency_p90': 2.105493855476379}, {'batch_size': 5, 'throughput': 1.7588071066449091, 'latency_mean': 2.831242549419403, 'latency_p50': 2.796622395515442, 'latency_p90': 3.203091764450073}, {'batch_size': 6, 'throughput': 1.7634317247526823, 'latency_mean': 3.3833115208148956, 'latency_p50': 3.3564313650131226, 'latency_p90': 3.8059626817703247}, {'batch_size': 8, 'throughput': 1.7477161642985801, 'latency_mean': 4.553514835834503, 'latency_p50': 4.56014621257782, 'latency_p90': 5.1842406511306764}, {'batch_size': 10, 'throughput': 1.7370002563856812, 'latency_mean': 5.715762542486191, 'latency_p50': 5.771531701087952, 'latency_p90': 6.441944813728332}]

max_input_tokens: 512

max_output_tokens: 64

model_architecture: LlamaForCausalLM

model_group: jic062/dpo-v1.5

model_name: jic062-dpo-v1-5_v1

model_num_parameters: 8030261248.0

model_repo: jic062/dpo-v1.5

model_size: 8B

num_battles: 12254

num_wins: 5972

propriety_score: 0.75

propriety_total_count: 1076.0

ranking_group: single

status: inactive

submission_type: basic

throughput_3p7s: 1.77

timestamp: 2024-09-14T18:21:29+00:00

us_pacific_date: 2024-09-14

win_ratio: 0.48735106903868125

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-v1-5-v1-mkmlizer
Waiting for job on jic062-dpo-v1-5-v1-mkmlizer to finish
jic062-dpo-v1-5-v1-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
jic062-dpo-v1-5-v1-mkmlizer: ║     _____            __           __                                ║
jic062-dpo-v1-5-v1-mkmlizer: ║    / _/ /_ ___    __/ /  ___ ___ / /                                ║
jic062-dpo-v1-5-v1-mkmlizer: ║   / _/ / // / |/|/ / _ \/ -_) -_) /                                 ║
jic062-dpo-v1-5-v1-mkmlizer: ║  /_//_/\_, /|__,__/_//_/\__/\__/_/                                  ║
jic062-dpo-v1-5-v1-mkmlizer: ║       /___/                                                         ║
jic062-dpo-v1-5-v1-mkmlizer: ║                                                                     ║
jic062-dpo-v1-5-v1-mkmlizer: ║  Version: 0.10.1                                                    ║
jic062-dpo-v1-5-v1-mkmlizer: ║  Copyright 2023 MK ONE TECHNOLOGIES Inc.                            ║
jic062-dpo-v1-5-v1-mkmlizer: ║  https://mk1.ai                                                     ║
jic062-dpo-v1-5-v1-mkmlizer: ║                                                                     ║
jic062-dpo-v1-5-v1-mkmlizer: ║  The license key for the current software has been verified as      ║
jic062-dpo-v1-5-v1-mkmlizer: ║  belonging to:                                                      ║
jic062-dpo-v1-5-v1-mkmlizer: ║                                                                     ║
jic062-dpo-v1-5-v1-mkmlizer: ║  Chai Research Corp.                                                ║
jic062-dpo-v1-5-v1-mkmlizer: ║  Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f                   ║
jic062-dpo-v1-5-v1-mkmlizer: ║  Expiration: 2024-10-15 23:59:59                                    ║
jic062-dpo-v1-5-v1-mkmlizer: ║                                                                     ║
jic062-dpo-v1-5-v1-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
jic062-dpo-v1-5-v1-mkmlizer: Downloaded to shared memory in 43.507s
jic062-dpo-v1-5-v1-mkmlizer: quantizing model to /dev/shm/model_cache, profile:s0, folder:/tmp/tmp044_5vj_, device:0
jic062-dpo-v1-5-v1-mkmlizer: Saving flywheel model at /dev/shm/model_cache
Connection pool is full, discarding connection: %s. Connection pool size: %s
jic062-dpo-v1-5-v1-mkmlizer: quantized model in 26.425s
jic062-dpo-v1-5-v1-mkmlizer: Processed model jic062/dpo-v1.5 in 69.933s
jic062-dpo-v1-5-v1-mkmlizer: creating bucket guanaco-mkml-models
jic062-dpo-v1-5-v1-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
jic062-dpo-v1-5-v1-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/jic062-dpo-v1-5-v1
jic062-dpo-v1-5-v1-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/jic062-dpo-v1-5-v1/config.json
jic062-dpo-v1-5-v1-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/jic062-dpo-v1-5-v1/special_tokens_map.json
jic062-dpo-v1-5-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/jic062-dpo-v1-5-v1/tokenizer_config.json
jic062-dpo-v1-5-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/jic062-dpo-v1-5-v1/tokenizer.json
jic062-dpo-v1-5-v1-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/jic062-dpo-v1-5-v1/flywheel_model.0.safetensors
jic062-dpo-v1-5-v1-mkmlizer: 
Loading 0:   0%|          | 0/291 [00:00<?, ?it/s]
Loading 0:   2%|▏         | 5/291 [00:00<00:07, 36.31it/s]
Loading 0:   5%|▍         | 14/291 [00:00<00:05, 49.31it/s]
Loading 0:   8%|▊         | 22/291 [00:00<00:04, 57.47it/s]
Loading 0:  10%|▉         | 28/291 [00:00<00:05, 50.29it/s]
Loading 0:  12%|█▏        | 34/291 [00:00<00:05, 49.60it/s]
Loading 0:  14%|█▍        | 41/291 [00:00<00:05, 47.44it/s]
Loading 0:  17%|█▋        | 50/291 [00:00<00:04, 50.90it/s]
Loading 0:  20%|██        | 59/291 [00:01<00:04, 52.49it/s]
Loading 0:  23%|██▎       | 68/291 [00:01<00:04, 52.92it/s]
Loading 0:  26%|██▌       | 76/291 [00:01<00:03, 57.85it/s]
Loading 0:  28%|██▊       | 82/291 [00:01<00:04, 50.87it/s]
Loading 0:  30%|███       | 88/291 [00:01<00:05, 36.73it/s]
Loading 0:  32%|███▏      | 93/291 [00:01<00:05, 39.10it/s]
Loading 0:  34%|███▎      | 98/291 [00:02<00:04, 41.19it/s]
Loading 0:  36%|███▌      | 104/291 [00:02<00:04, 40.54it/s]
Loading 0:  38%|███▊      | 112/291 [00:02<00:03, 48.24it/s]
Loading 0:  41%|████      | 118/291 [00:02<00:03, 46.32it/s]
Loading 0:  42%|████▏     | 123/291 [00:02<00:03, 46.21it/s]
Loading 0:  44%|████▍     | 128/291 [00:02<00:03, 42.82it/s]
Loading 0:  46%|████▌     | 133/291 [00:02<00:04, 37.19it/s]
Loading 0:  48%|████▊     | 139/291 [00:03<00:03, 42.08it/s]
Loading 0:  49%|████▉     | 144/291 [00:03<00:03, 43.24it/s]
Loading 0:  51%|█████     | 149/291 [00:03<00:03, 36.10it/s]
Loading 0:  54%|█████▎    | 156/291 [00:03<00:03, 43.28it/s]
Loading 0:  55%|█████▌    | 161/291 [00:03<00:03, 43.08it/s]
Loading 0:  57%|█████▋    | 166/291 [00:03<00:02, 42.58it/s]
Loading 0:  59%|█████▉    | 172/291 [00:03<00:02, 42.23it/s]
Loading 0:  62%|██████▏   | 179/291 [00:03<00:02, 48.04it/s]
Loading 0:  64%|██████▎   | 185/291 [00:04<00:02, 50.89it/s]
Loading 0:  66%|██████▌   | 191/291 [00:04<00:02, 34.00it/s]
Loading 0:  67%|██████▋   | 196/291 [00:04<00:02, 35.63it/s]
Loading 0:  69%|██████▉   | 202/291 [00:04<00:02, 39.57it/s]
Loading 0:  71%|███████   | 207/291 [00:04<00:02, 41.21it/s]
Loading 0:  73%|███████▎  | 212/291 [00:04<00:02, 36.25it/s]
Loading 0:  75%|███████▌  | 219/291 [00:04<00:01, 42.68it/s]
Loading 0:  77%|███████▋  | 224/291 [00:05<00:01, 44.09it/s]
Loading 0:  79%|███████▉  | 230/291 [00:05<00:01, 39.72it/s]
Loading 0:  82%|████████▏ | 238/291 [00:05<00:01, 47.53it/s]
Loading 0:  84%|████████▍ | 244/291 [00:05<00:01, 46.35it/s]
Loading 0:  86%|████████▌ | 249/291 [00:05<00:00, 44.59it/s]
Loading 0:  88%|████████▊ | 255/291 [00:05<00:00, 46.99it/s]
Loading 0:  89%|████████▉ | 260/291 [00:05<00:00, 46.32it/s]
Loading 0:  91%|█████████ | 265/291 [00:05<00:00, 45.93it/s]
Loading 0:  93%|█████████▎| 271/291 [00:06<00:00, 45.26it/s]
Loading 0:  95%|█████████▍| 276/291 [00:06<00:00, 45.65it/s]
Loading 0:  97%|█████████▋| 281/291 [00:06<00:00, 45.03it/s]
Loading 0:  98%|█████████▊| 286/291 [00:06<00:00, 38.84it/s]
Loading 0: 100%|██████████| 291/291 [00:11<00:00,  3.01it/s]
                                                            
Job jic062-dpo-v1-5-v1-mkmlizer completed after 95.74s with status: succeeded
Stopping job with name jic062-dpo-v1-5-v1-mkmlizer
Pipeline stage MKMLizer completed in 96.89s
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 jic062-dpo-v1-5-v1
Waiting for inference service jic062-dpo-v1-5-v1 to be ready
Failed to get response for submission blend_hokok_2024-09-09: ('http://neversleep-noromaid-v0-8068-v150-predictor.tenant-chaiml-guanaco.k.chaiverse.com/v1/models/GPT-J-6B-lit-v2:predict', '')
Inference service jic062-dpo-v1-5-v1 ready after 161.47682094573975s
Pipeline stage MKMLDeployer completed in 162.68s
run pipeline stage %s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.275589942932129s
Received healthy response to inference request in 1.7532265186309814s
Received healthy response to inference request in 2.2423830032348633s
Received healthy response to inference request in 1.9470465183258057s
Received healthy response to inference request in 3.4277210235595703s
5 requests
0 failed requests
5th percentile: 1.7919905185699463
10th percentile: 1.8307545185089111
20th percentile: 1.9082825183868408
30th percentile: 2.006113815307617
40th percentile: 2.12424840927124
50th percentile: 2.2423830032348633
60th percentile: 2.2556657791137695
70th percentile: 2.268948554992676
80th percentile: 2.5060161590576175
90th percentile: 2.966868591308594
95th percentile: 3.1972948074340817
99th percentile: 3.3816357803344728
mean time: 2.3291934013366697
Pipeline stage StressChecker completed in 17.19s
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 6.99s
Shutdown handler de-registered
jic062-dpo-v1-5_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 0.12s
run pipeline stage %s
Running pipeline stage MKMLProfilerTemplater
Pipeline stage MKMLProfilerTemplater completed in 3.00s
run pipeline stage %s
Running pipeline stage MKMLProfilerDeployer
Creating inference service jic062-dpo-v1-5-v1-profiler
Waiting for inference service jic062-dpo-v1-5-v1-profiler to be ready
Inference service jic062-dpo-v1-5-v1-profiler ready after 170.38003253936768s
Pipeline stage MKMLProfilerDeployer completed in 172.58s
run pipeline stage %s
Running pipeline stage MKMLProfilerRunner
kubectl cp /code/guanaco/guanaco_inference_services/src/inference_scripts tenant-chaiml-guanaco/jic062-dpo-v1-5-v1-profiler-predictor-00001-deployment-6bf6b6rv:/code/chaiverse_profiler_1726338596 --namespace tenant-chaiml-guanaco
kubectl exec -it jic062-dpo-v1-5-v1-profiler-predictor-00001-deployment-6bf6b6rv --namespace tenant-chaiml-guanaco -- sh -c 'cd /code/chaiverse_profiler_1726338596 && 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_1726338596/summary.json'
kubectl exec -it jic062-dpo-v1-5-v1-profiler-predictor-00001-deployment-6bf6b6rv --namespace tenant-chaiml-guanaco -- bash -c 'cat /code/chaiverse_profiler_1726338596/summary.json'
Pipeline stage MKMLProfilerRunner completed in 809.96s
run pipeline stage %s
Running pipeline stage MKMLProfilerDeleter
Checking if service jic062-dpo-v1-5-v1-profiler is running
Tearing down inference service jic062-dpo-v1-5-v1-profiler
Service jic062-dpo-v1-5-v1-profiler has been torndown
Pipeline stage MKMLProfilerDeleter completed in 3.66s
Shutdown handler de-registered
jic062-dpo-v1-5_v1 status is now inactive due to auto deactivation removed underperforming models