jic062-dpo-v1-6

submission_id: jic062-dpo-v1-6_v1

developer_uid: chace9580

best_of: 16

celo_rating: 1240.93

display_name: jic062-dpo-v1-6_v1

family_friendly_score: 0.0

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': 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-v1.6

latencies: [{'batch_size': 1, 'throughput': 0.8970719898034647, 'latency_mean': 1.1146573317050934, 'latency_p50': 1.1189790964126587, 'latency_p90': 1.2374611616134643}, {'batch_size': 3, 'throughput': 1.5786732531437992, 'latency_mean': 1.890986316204071, 'latency_p50': 1.909153699874878, 'latency_p90': 2.1054311275482176}, {'batch_size': 5, 'throughput': 1.738651771577124, 'latency_mean': 2.857687853574753, 'latency_p50': 2.8671857118606567, 'latency_p90': 3.240846347808838}, {'batch_size': 6, 'throughput': 1.7539463326220663, 'latency_mean': 3.4002345323562624, 'latency_p50': 3.4146770238876343, 'latency_p90': 3.7797909498214723}, {'batch_size': 8, 'throughput': 1.729421693037199, 'latency_mean': 4.58982803106308, 'latency_p50': 4.58635687828064, 'latency_p90': 5.080252552032471}, {'batch_size': 10, 'throughput': 1.7622064506139392, 'latency_mean': 5.621050796508789, 'latency_p50': 5.58858585357666, 'latency_p90': 6.577765011787415}]

max_input_tokens: 512

max_output_tokens: 64

model_architecture: LlamaForCausalLM

model_group: jic062/dpo-v1.6

model_name: jic062-dpo-v1-6_v1

model_num_parameters: 8030261248.0

model_repo: jic062/dpo-v1.6

model_size: 8B

num_battles: 10000

num_wins: 4826

ranking_group: single

status: torndown

submission_type: basic

throughput_3p7s: 1.76

timestamp: 2024-09-15T18:09:33+00:00

us_pacific_date: 2024-09-15

win_ratio: 0.4826

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-v1-6-v1-mkmlizer
Waiting for job on jic062-dpo-v1-6-v1-mkmlizer to finish
jic062-dpo-v1-6-v1-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
jic062-dpo-v1-6-v1-mkmlizer: ║     _____            __           __                                ║
jic062-dpo-v1-6-v1-mkmlizer: ║    / _/ /_ ___    __/ /  ___ ___ / /                                ║
jic062-dpo-v1-6-v1-mkmlizer: ║   / _/ / // / |/|/ / _ \/ -_) -_) /                                 ║
jic062-dpo-v1-6-v1-mkmlizer: ║  /_//_/\_, /|__,__/_//_/\__/\__/_/                                  ║
jic062-dpo-v1-6-v1-mkmlizer: ║       /___/                                                         ║
jic062-dpo-v1-6-v1-mkmlizer: ║                                                                     ║
jic062-dpo-v1-6-v1-mkmlizer: ║  Version: 0.10.1                                                    ║
jic062-dpo-v1-6-v1-mkmlizer: ║  Copyright 2023 MK ONE TECHNOLOGIES Inc.                            ║
jic062-dpo-v1-6-v1-mkmlizer: ║  https://mk1.ai                                                     ║
jic062-dpo-v1-6-v1-mkmlizer: ║                                                                     ║
jic062-dpo-v1-6-v1-mkmlizer: ║  The license key for the current software has been verified as      ║
jic062-dpo-v1-6-v1-mkmlizer: ║  belonging to:                                                      ║
jic062-dpo-v1-6-v1-mkmlizer: ║                                                                     ║
jic062-dpo-v1-6-v1-mkmlizer: ║  Chai Research Corp.                                                ║
jic062-dpo-v1-6-v1-mkmlizer: ║  Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f                   ║
jic062-dpo-v1-6-v1-mkmlizer: ║  Expiration: 2024-10-15 23:59:59                                    ║
jic062-dpo-v1-6-v1-mkmlizer: ║                                                                     ║
jic062-dpo-v1-6-v1-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
jic062-dpo-v1-6-v1-mkmlizer: Downloaded to shared memory in 37.047s
jic062-dpo-v1-6-v1-mkmlizer: quantizing model to /dev/shm/model_cache, profile:s0, folder:/tmp/tmpfks9x8my, device:0
jic062-dpo-v1-6-v1-mkmlizer: Saving flywheel model at /dev/shm/model_cache
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', '')
jic062-dpo-v1-6-v1-mkmlizer: quantized model in 26.583s
jic062-dpo-v1-6-v1-mkmlizer: Processed model jic062/dpo-v1.6 in 63.630s
jic062-dpo-v1-6-v1-mkmlizer: creating bucket guanaco-mkml-models
jic062-dpo-v1-6-v1-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
jic062-dpo-v1-6-v1-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/jic062-dpo-v1-6-v1
jic062-dpo-v1-6-v1-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/jic062-dpo-v1-6-v1/config.json
jic062-dpo-v1-6-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/jic062-dpo-v1-6-v1/tokenizer_config.json
jic062-dpo-v1-6-v1-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/jic062-dpo-v1-6-v1/special_tokens_map.json
jic062-dpo-v1-6-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/jic062-dpo-v1-6-v1/tokenizer.json
jic062-dpo-v1-6-v1-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/jic062-dpo-v1-6-v1/flywheel_model.0.safetensors
jic062-dpo-v1-6-v1-mkmlizer: 
Loading 0:   0%|          | 0/291 [00:00<?, ?it/s]
Loading 0:   2%|▏         | 5/291 [00:00<00:08, 32.47it/s]
Loading 0:   4%|▍         | 12/291 [00:00<00:05, 49.00it/s]
Loading 0:   6%|▌         | 18/291 [00:00<00:05, 52.30it/s]
Loading 0:   8%|▊         | 24/291 [00:00<00:05, 46.66it/s]
Loading 0:  11%|█         | 32/291 [00:00<00:05, 48.11it/s]
Loading 0:  14%|█▍        | 41/291 [00:00<00:05, 49.36it/s]
Loading 0:  16%|█▋        | 48/291 [00:00<00:04, 54.05it/s]
Loading 0:  19%|█▊        | 54/291 [00:01<00:04, 54.19it/s]
Loading 0:  21%|██        | 60/291 [00:01<00:04, 47.46it/s]
Loading 0:  23%|██▎       | 66/291 [00:01<00:04, 50.02it/s]
Loading 0:  25%|██▍       | 72/291 [00:01<00:04, 52.10it/s]
Loading 0:  27%|██▋       | 78/291 [00:01<00:04, 45.00it/s]
Loading 0:  29%|██▊       | 83/291 [00:01<00:05, 36.00it/s]
Loading 0:  30%|███       | 88/291 [00:01<00:05, 36.50it/s]
Loading 0:  32%|███▏      | 94/291 [00:02<00:04, 40.25it/s]
Loading 0:  34%|███▍      | 100/291 [00:02<00:04, 38.57it/s]
Loading 0:  36%|███▌      | 105/291 [00:02<00:04, 39.40it/s]
Loading 0:  38%|███▊      | 112/291 [00:02<00:03, 46.06it/s]
Loading 0:  40%|████      | 117/291 [00:02<00:03, 46.20it/s]
Loading 0:  42%|████▏     | 122/291 [00:02<00:04, 36.89it/s]
Loading 0:  44%|████▍     | 129/291 [00:02<00:03, 43.54it/s]
Loading 0:  46%|████▌     | 134/291 [00:03<00:03, 43.00it/s]
Loading 0:  48%|████▊     | 139/291 [00:03<00:03, 42.13it/s]
Loading 0:  49%|████▉     | 144/291 [00:03<00:03, 43.38it/s]
Loading 0:  51%|█████     | 149/291 [00:03<00:03, 36.35it/s]
Loading 0:  54%|█████▎    | 156/291 [00:03<00:03, 42.42it/s]
Loading 0:  55%|█████▌    | 161/291 [00:03<00:02, 43.44it/s]
Loading 0:  57%|█████▋    | 167/291 [00:03<00:03, 40.55it/s]
Loading 0:  60%|██████    | 176/291 [00:03<00:02, 51.93it/s]
Loading 0:  63%|██████▎   | 182/291 [00:04<00:02, 45.75it/s]
Loading 0:  64%|██████▍   | 187/291 [00:04<00:02, 36.75it/s]
Loading 0:  66%|██████▌   | 192/291 [00:04<00:02, 36.97it/s]
Loading 0:  68%|██████▊   | 197/291 [00:04<00:02, 37.65it/s]
Loading 0:  69%|██████▉   | 202/291 [00:04<00:02, 39.68it/s]
Loading 0:  71%|███████▏  | 208/291 [00:04<00:02, 38.59it/s]
Loading 0:  73%|███████▎  | 213/291 [00:04<00:01, 39.31it/s]
Loading 0:  76%|███████▌  | 220/291 [00:05<00:01, 45.62it/s]
Loading 0:  78%|███████▊  | 226/291 [00:05<00:01, 44.97it/s]
Loading 0:  79%|███████▉  | 231/291 [00:05<00:01, 44.68it/s]
Loading 0:  82%|████████▏ | 238/291 [00:05<00:01, 49.09it/s]
Loading 0:  84%|████████▍ | 244/291 [00:05<00:01, 45.27it/s]
Loading 0:  86%|████████▌ | 249/291 [00:05<00:00, 44.80it/s]
Loading 0:  88%|████████▊ | 256/291 [00:05<00:00, 49.81it/s]
Loading 0:  90%|█████████ | 262/291 [00:05<00:00, 47.76it/s]
Loading 0:  92%|█████████▏| 267/291 [00:06<00:00, 47.52it/s]
Loading 0:  94%|█████████▍| 274/291 [00:06<00:00, 52.27it/s]
Loading 0:  96%|█████████▌| 280/291 [00:06<00:00, 47.68it/s]
Loading 0:  98%|█████████▊| 285/291 [00:06<00:00, 44.67it/s]
Loading 0: 100%|█████████▉| 290/291 [00:11<00:00,  3.22it/s]
                                                            
Job jic062-dpo-v1-6-v1-mkmlizer completed after 90.02s with status: succeeded
Stopping job with name jic062-dpo-v1-6-v1-mkmlizer
Pipeline stage MKMLizer completed in 91.70s
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-v1-6-v1
Waiting for inference service jic062-dpo-v1-6-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
Connection pool is full, discarding connection: %s. Connection pool size: %s
Connection pool is full, discarding connection: %s. Connection pool size: %s
Failed to get response for submission blend_tagim_2024-09-14: ('http://chaiml-llama-8b-pairwis-8189-v19-predictor.tenant-chaiml-guanaco.k.chaiverse.com/v1/models/GPT-J-6B-lit-v2:predict', '{"error":"ValueError : [TypeError(\\"\'numpy.int64\' object is not iterable\\"), TypeError(\'vars() argument must have __dict__ attribute\')]"}')
Inference service jic062-dpo-v1-6-v1 ready after 170.97616744041443s
Pipeline stage MKMLDeployer completed in 171.33s
run pipeline stage %s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.3814096450805664s
Received healthy response to inference request in 1.8190336227416992s
Received healthy response to inference request in 2.025083541870117s
Received healthy response to inference request in 2.451071262359619s
Received healthy response to inference request in 1.8216655254364014s
5 requests
0 failed requests
5th percentile: 1.8195600032806396
10th percentile: 1.82008638381958
20th percentile: 1.821139144897461
30th percentile: 1.8623491287231446
40th percentile: 1.943716335296631
50th percentile: 2.025083541870117
60th percentile: 2.167613983154297
70th percentile: 2.3101444244384766
80th percentile: 2.395341968536377
90th percentile: 2.423206615447998
95th percentile: 2.4371389389038085
99th percentile: 2.448284797668457
mean time: 2.099652719497681
Pipeline stage StressChecker completed in 11.09s
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 5.95s
Shutdown handler de-registered
jic062-dpo-v1-6_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 0.11s
run pipeline stage %s
Running pipeline stage MKMLProfilerDeployer
Creating inference service jic062-dpo-v1-6-v1-profiler
Waiting for inference service jic062-dpo-v1-6-v1-profiler to be ready
Inference service jic062-dpo-v1-6-v1-profiler ready after 170.39563584327698s
Pipeline stage MKMLProfilerDeployer completed in 170.74s
run pipeline stage %s
Running pipeline stage MKMLProfilerRunner
kubectl cp /code/guanaco/guanaco_inference_services/src/inference_scripts tenant-chaiml-guanaco/jic062-dpo-v1-6-v1-profiler-predictor-00001-deployment-557fhksf:/code/chaiverse_profiler_1726424270 --namespace tenant-chaiml-guanaco
kubectl exec -it jic062-dpo-v1-6-v1-profiler-predictor-00001-deployment-557fhksf --namespace tenant-chaiml-guanaco -- sh -c 'cd /code/chaiverse_profiler_1726424270 && 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_1726424270/summary.json'
kubectl exec -it jic062-dpo-v1-6-v1-profiler-predictor-00001-deployment-557fhksf --namespace tenant-chaiml-guanaco -- bash -c 'cat /code/chaiverse_profiler_1726424270/summary.json'
Pipeline stage MKMLProfilerRunner completed in 817.44s
run pipeline stage %s
Running pipeline stage MKMLProfilerDeleter
Checking if service jic062-dpo-v1-6-v1-profiler is running
Tearing down inference service jic062-dpo-v1-6-v1-profiler
Service jic062-dpo-v1-6-v1-profiler has been torndown
Pipeline stage MKMLProfilerDeleter completed in 1.94s
Shutdown handler de-registered
jic062-dpo-v1-6_v1 status is now inactive due to auto deactivation removed underperforming models
jic062-dpo-v1-6_v1 status is now torndown due to DeploymentManager action