rica40325-feedback-dpo-10

developer_uid: rica40325

submission_id: rica40325-feedback-dpo-10_v2

model_name: rica40325-feedback-dpo-10_v1

model_group: rica40325/feedback_dpo_1

status: torndown

timestamp: 2024-09-11T08:22:05+00:00

num_battles: 12868

num_wins: 6174

celo_rating: 1228.45

family_friendly_score: 0.0

submission_type: basic

model_repo: rica40325/feedback_dpo_10

model_architecture: LlamaForCausalLM

model_num_parameters: 8030261248.0

best_of: 8

max_input_tokens: 512

max_output_tokens: 64

latencies: [{'batch_size': 1, 'throughput': 0.9850711833029422, 'latency_mean': 1.015062974691391, 'latency_p50': 1.0110405683517456, 'latency_p90': 1.1405688762664794}, {'batch_size': 5, 'throughput': 2.6699645872118563, 'latency_mean': 1.8652181494235993, 'latency_p50': 1.8558623790740967, 'latency_p90': 2.06534206867218}, {'batch_size': 10, 'throughput': 3.1514048027376917, 'latency_mean': 3.1375916981697083, 'latency_p50': 3.1663548946380615, 'latency_p90': 3.5436832189559935}, {'batch_size': 15, 'throughput': 3.36217207636068, 'latency_mean': 4.408860654830932, 'latency_p50': 4.409454584121704, 'latency_p90': 4.990269899368286}, {'batch_size': 20, 'throughput': 3.345782400532887, 'latency_mean': 5.8601353490352635, 'latency_p50': 5.902622938156128, 'latency_p90': 6.546109509468079}]

gpu_counts: {'NVIDIA RTX A5000': 1}

display_name: rica40325-feedback-dpo-10_v1

is_internal_developer: False

language_model: rica40325/feedback_dpo_10

model_size: 8B

ranking_group: single

throughput_3p7s: 3.3

us_pacific_date: 2024-09-11

win_ratio: 0.4797948399129624

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'], 'max_input_tokens': 512, 'best_of': 8, 'max_output_tokens': 64}

formatter: {'memory_template': "{bot_name}'s Persona: {memory}\n####\n", 'prompt_template': '{prompt}\n<START>\n', 'bot_template': '{bot_name}: {message}\n', 'user_template': '{user_name}: {message}\n', 'response_template': '{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 rica40325-feedback-dpo-10-v2-mkmlizer
Waiting for job on rica40325-feedback-dpo-10-v2-mkmlizer to finish
rica40325-feedback-dpo-10-v2-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
rica40325-feedback-dpo-10-v2-mkmlizer: ║     _____            __           __                                ║
rica40325-feedback-dpo-10-v2-mkmlizer: ║    / _/ /_ ___    __/ /  ___ ___ / /                                ║
rica40325-feedback-dpo-10-v2-mkmlizer: ║   / _/ / // / |/|/ / _ \/ -_) -_) /                                 ║
rica40325-feedback-dpo-10-v2-mkmlizer: ║  /_//_/\_, /|__,__/_//_/\__/\__/_/                                  ║
rica40325-feedback-dpo-10-v2-mkmlizer: ║       /___/                                                         ║
rica40325-feedback-dpo-10-v2-mkmlizer: ║                                                                     ║
rica40325-feedback-dpo-10-v2-mkmlizer: ║  Version: 0.10.1                                                    ║
rica40325-feedback-dpo-10-v2-mkmlizer: ║  Copyright 2023 MK ONE TECHNOLOGIES Inc.                            ║
rica40325-feedback-dpo-10-v2-mkmlizer: ║  https://mk1.ai                                                     ║
rica40325-feedback-dpo-10-v2-mkmlizer: ║                                                                     ║
rica40325-feedback-dpo-10-v2-mkmlizer: ║  The license key for the current software has been verified as      ║
rica40325-feedback-dpo-10-v2-mkmlizer: ║  belonging to:                                                      ║
rica40325-feedback-dpo-10-v2-mkmlizer: ║                                                                     ║
rica40325-feedback-dpo-10-v2-mkmlizer: ║  Chai Research Corp.                                                ║
rica40325-feedback-dpo-10-v2-mkmlizer: ║  Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f                   ║
rica40325-feedback-dpo-10-v2-mkmlizer: ║  Expiration: 2024-10-15 23:59:59                                    ║
rica40325-feedback-dpo-10-v2-mkmlizer: ║                                                                     ║
rica40325-feedback-dpo-10-v2-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
rica40325-feedback-dpo-10-v2-mkmlizer: Downloaded to shared memory in 54.254s
rica40325-feedback-dpo-10-v2-mkmlizer: quantizing model to /dev/shm/model_cache, profile:s0, folder:/tmp/tmpigov767w, device:0
rica40325-feedback-dpo-10-v2-mkmlizer: Saving flywheel model at /dev/shm/model_cache
rica40325-feedback-dpo-10-v2-mkmlizer: quantized model in 29.381s
rica40325-feedback-dpo-10-v2-mkmlizer: Processed model rica40325/feedback_dpo_10 in 83.635s
rica40325-feedback-dpo-10-v2-mkmlizer: creating bucket guanaco-mkml-models
rica40325-feedback-dpo-10-v2-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
rica40325-feedback-dpo-10-v2-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/rica40325-feedback-dpo-10-v2
rica40325-feedback-dpo-10-v2-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/rica40325-feedback-dpo-10-v2/config.json
rica40325-feedback-dpo-10-v2-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/rica40325-feedback-dpo-10-v2/special_tokens_map.json
rica40325-feedback-dpo-10-v2-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/rica40325-feedback-dpo-10-v2/tokenizer_config.json
rica40325-feedback-dpo-10-v2-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/rica40325-feedback-dpo-10-v2/tokenizer.json
Connection pool is full, discarding connection: %s. Connection pool size: %s
Connection pool is full, discarding connection: %s. Connection pool size: %s
rica40325-feedback-dpo-10-v2-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/rica40325-feedback-dpo-10-v2/flywheel_model.0.safetensors
rica40325-feedback-dpo-10-v2-mkmlizer: 
Loading 0:   0%|          | 0/291 [00:00<?, ?it/s]
Loading 0:   2%|▏         | 5/291 [00:00<00:11, 25.57it/s]
Loading 0:   4%|▍         | 12/291 [00:00<00:07, 35.77it/s]
Loading 0:   5%|▌         | 16/291 [00:00<00:08, 33.36it/s]
Loading 0:   7%|▋         | 21/291 [00:00<00:07, 34.39it/s]
Loading 0:   9%|▊         | 25/291 [00:00<00:08, 31.49it/s]
Loading 0:  11%|█         | 31/291 [00:00<00:06, 37.88it/s]
Loading 0:  12%|█▏        | 35/291 [00:01<00:10, 24.26it/s]
Loading 0:  13%|█▎        | 39/291 [00:01<00:10, 25.03it/s]
Loading 0:  14%|█▍        | 42/291 [00:01<00:10, 24.37it/s]
Loading 0:  16%|█▋        | 48/291 [00:01<00:07, 30.41it/s]
Loading 0:  18%|█▊        | 52/291 [00:01<00:07, 30.37it/s]
Loading 0:  20%|█▉        | 57/291 [00:01<00:07, 33.35it/s]
Loading 0:  21%|██        | 61/291 [00:02<00:07, 31.98it/s]
Loading 0:  23%|██▎       | 66/291 [00:02<00:06, 34.98it/s]
Loading 0:  24%|██▍       | 70/291 [00:02<00:06, 33.68it/s]
Loading 0:  25%|██▌       | 74/291 [00:02<00:06, 33.92it/s]
Loading 0:  27%|██▋       | 78/291 [00:02<00:06, 33.15it/s]
Loading 0:  28%|██▊       | 82/291 [00:02<00:08, 23.36it/s]
Loading 0:  29%|██▉       | 85/291 [00:02<00:08, 24.08it/s]
Loading 0:  31%|███       | 90/291 [00:03<00:07, 28.21it/s]
Loading 0:  32%|███▏      | 94/291 [00:03<00:06, 29.56it/s]
Loading 0:  34%|███▍      | 99/291 [00:03<00:05, 32.26it/s]
Loading 0:  35%|███▌      | 103/291 [00:03<00:05, 31.63it/s]
Loading 0:  37%|███▋      | 108/291 [00:03<00:05, 33.55it/s]
Loading 0:  38%|███▊      | 112/291 [00:03<00:05, 32.51it/s]
Loading 0:  40%|███▉      | 116/291 [00:03<00:05, 33.78it/s]
Loading 0:  42%|████▏     | 122/291 [00:03<00:04, 38.79it/s]
Loading 0:  44%|████▎     | 127/291 [00:04<00:04, 34.77it/s]
Loading 0:  46%|████▌     | 133/291 [00:04<00:05, 29.74it/s]
Loading 0:  47%|████▋     | 137/291 [00:04<00:05, 29.64it/s]
Loading 0:  48%|████▊     | 141/291 [00:04<00:05, 28.33it/s]
Loading 0:  51%|█████     | 147/291 [00:04<00:04, 33.78it/s]
Loading 0:  52%|█████▏    | 151/291 [00:04<00:04, 31.83it/s]
Loading 0:  54%|█████▎    | 156/291 [00:05<00:04, 33.73it/s]
Loading 0:  55%|█████▍    | 160/291 [00:05<00:04, 32.03it/s]
Loading 0:  57%|█████▋    | 165/291 [00:05<00:03, 33.65it/s]
Loading 0:  58%|█████▊    | 169/291 [00:05<00:03, 31.46it/s]
Loading 0:  60%|█████▉    | 174/291 [00:05<00:03, 34.26it/s]
Loading 0:  61%|██████    | 178/291 [00:05<00:03, 33.63it/s]
Loading 0:  64%|██████▎   | 185/291 [00:05<00:02, 38.47it/s]
Loading 0:  65%|██████▍   | 189/291 [00:06<00:04, 25.43it/s]
Loading 0:  67%|██████▋   | 194/291 [00:06<00:03, 26.26it/s]
Loading 0:  69%|██████▉   | 201/291 [00:06<00:02, 32.78it/s]
Loading 0:  70%|███████   | 205/291 [00:06<00:02, 32.55it/s]
Loading 0:  72%|███████▏  | 210/291 [00:06<00:02, 35.37it/s]
Loading 0:  74%|███████▎  | 214/291 [00:06<00:02, 34.32it/s]
Loading 0:  75%|███████▌  | 219/291 [00:06<00:01, 37.45it/s]
Loading 0:  77%|███████▋  | 223/291 [00:07<00:02, 33.35it/s]
Loading 0:  78%|███████▊  | 227/291 [00:07<00:01, 32.54it/s]
Loading 0:  79%|███████▉  | 231/291 [00:07<00:01, 32.27it/s]
Loading 0:  81%|████████  | 235/291 [00:07<00:02, 25.44it/s]
Loading 0:  82%|████████▏ | 239/291 [00:07<00:02, 25.02it/s]
Loading 0:  85%|████████▍ | 246/291 [00:07<00:01, 33.03it/s]
Loading 0:  86%|████████▌ | 250/291 [00:07<00:01, 32.82it/s]
Loading 0:  88%|████████▊ | 255/291 [00:08<00:00, 36.11it/s]
Loading 0:  89%|████████▉ | 259/291 [00:08<00:00, 34.30it/s]
Loading 0:  91%|█████████ | 264/291 [00:08<00:00, 36.56it/s]
Loading 0:  92%|█████████▏| 268/291 [00:08<00:00, 34.18it/s]
Loading 0:  94%|█████████▍| 273/291 [00:08<00:00, 36.11it/s]
Loading 0:  95%|█████████▌| 277/291 [00:08<00:00, 33.76it/s]
Loading 0:  97%|█████████▋| 281/291 [00:08<00:00, 34.49it/s]
Loading 0:  98%|█████████▊| 286/291 [00:14<00:01,  2.59it/s]
Loading 0:  99%|█████████▉| 289/291 [00:14<00:00,  3.22it/s]
                                                            
Job rica40325-feedback-dpo-10-v2-mkmlizer completed after 104.89s with status: succeeded
Stopping job with name rica40325-feedback-dpo-10-v2-mkmlizer
Pipeline stage MKMLizer completed in 106.10s
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 rica40325-feedback-dpo-10-v2
Waiting for inference service rica40325-feedback-dpo-10-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
Inference service rica40325-feedback-dpo-10-v2 ready after 161.35511088371277s
Pipeline stage MKMLDeployer completed in 161.80s
run pipeline stage %s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.37652850151062s
Received healthy response to inference request in 2.210474729537964s
Received healthy response to inference request in 1.4512887001037598s
Received healthy response to inference request in 1.737461805343628s
Received healthy response to inference request in 1.9667718410491943s
5 requests
0 failed requests
5th percentile: 1.5085233211517335
10th percentile: 1.565757942199707
20th percentile: 1.6802271842956542
30th percentile: 1.7833238124847413
40th percentile: 1.8750478267669677
50th percentile: 1.9667718410491943
60th percentile: 2.064252996444702
70th percentile: 2.16173415184021
80th percentile: 2.243685483932495
90th percentile: 2.3101069927215576
95th percentile: 2.343317747116089
99th percentile: 2.369886350631714
mean time: 1.9485051155090332
Pipeline stage StressChecker completed in 10.52s
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 6.17s
Shutdown handler de-registered
rica40325-feedback-dpo-10_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 0.11s
run pipeline stage %s
Running pipeline stage MKMLProfilerDeployer
Creating inference service rica40325-feedback-dpo-10-v2-profiler
Waiting for inference service rica40325-feedback-dpo-10-v2-profiler to be ready
Inference service rica40325-feedback-dpo-10-v2-profiler ready after 160.50564217567444s
Pipeline stage MKMLProfilerDeployer completed in 160.86s
run pipeline stage %s
Running pipeline stage MKMLProfilerRunner
kubectl cp /code/guanaco/guanaco_inference_services/src/inference_scripts tenant-chaiml-guanaco/rica40325-feedback-d1783b54485310643a6ffb6731e43b9f9-deplolrhkl:/code/chaiverse_profiler_1726043407 --namespace tenant-chaiml-guanaco
kubectl exec -it rica40325-feedback-d1783b54485310643a6ffb6731e43b9f9-deplolrhkl --namespace tenant-chaiml-guanaco -- sh -c 'cd /code/chaiverse_profiler_1726043407 && 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_1726043407/summary.json'
kubectl exec -it rica40325-feedback-d1783b54485310643a6ffb6731e43b9f9-deplolrhkl --namespace tenant-chaiml-guanaco -- bash -c 'cat /code/chaiverse_profiler_1726043407/summary.json'
Pipeline stage MKMLProfilerRunner completed in 466.27s
run pipeline stage %s
Running pipeline stage MKMLProfilerDeleter
Checking if service rica40325-feedback-dpo-10-v2-profiler is running
Tearing down inference service rica40325-feedback-dpo-10-v2-profiler
Service rica40325-feedback-dpo-10-v2-profiler has been torndown
Pipeline stage MKMLProfilerDeleter completed in 1.76s
Shutdown handler de-registered
rica40325-feedback-dpo-10_v2 status is now inactive due to auto deactivation removed underperforming models
rica40325-feedback-dpo-10_v2 status is now torndown due to DeploymentManager action