rica40325-feedback-dpo-2

developer_uid: rica40325

submission_id: rica40325-feedback-dpo-2_v2

model_name: rica40325-feedback-dpo-2_v1

model_group: rica40325/feedback_dpo_2

status: torndown

timestamp: 2024-09-10T05:51:39+00:00

num_battles: 10866

num_wins: 2165

celo_rating: 983.01

family_friendly_score: 0.0

submission_type: basic

model_repo: rica40325/feedback_dpo_2

model_architecture: LlamaForCausalLM

model_num_parameters: 8030261248.0

best_of: 1

max_input_tokens: 512

max_output_tokens: 64

latencies: [{'batch_size': 1, 'throughput': 1.0696030768684375, 'latency_mean': 0.9348372662067413, 'latency_p50': 0.9377423524856567, 'latency_p90': 1.035806918144226}, {'batch_size': 5, 'throughput': 3.457146532346779, 'latency_mean': 1.4363089048862456, 'latency_p50': 1.4287923574447632, 'latency_p90': 1.5925129175186157}, {'batch_size': 10, 'throughput': 5.228592055241544, 'latency_mean': 1.8933809340000152, 'latency_p50': 1.8895419836044312, 'latency_p90': 2.1416224241256714}, {'batch_size': 15, 'throughput': 6.053727629424411, 'latency_mean': 2.428587933778763, 'latency_p50': 2.4296354055404663, 'latency_p90': 2.7588167428970336}, {'batch_size': 20, 'throughput': 6.573328510316361, 'latency_mean': 2.9854100012779234, 'latency_p50': 2.9754165410995483, 'latency_p90': 3.3725621938705443}, {'batch_size': 25, 'throughput': 6.979885142782057, 'latency_mean': 3.505819847583771, 'latency_p50': 3.4739861488342285, 'latency_p90': 4.106622672080993}, {'batch_size': 30, 'throughput': 7.29009411717532, 'latency_mean': 4.020029604434967, 'latency_p50': 4.024428844451904, 'latency_p90': 4.733847403526306}, {'batch_size': 35, 'throughput': 7.283789355541202, 'latency_mean': 4.689255359172821, 'latency_p50': 4.688684701919556, 'latency_p90': 5.486804533004761}, {'batch_size': 40, 'throughput': 7.332831855736693, 'latency_mean': 5.2563091194629665, 'latency_p50': 5.177881479263306, 'latency_p90': 6.0971712350845335}]

gpu_counts: {'NVIDIA RTX A5000': 1}

display_name: rica40325-feedback-dpo-2_v1

is_internal_developer: False

language_model: rica40325/feedback_dpo_2

model_size: 8B

ranking_group: single

throughput_3p7s: 7.27

us_pacific_date: 2024-09-09

win_ratio: 0.199245352475612

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': 1, '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
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
Starting job with name rica40325-feedback-dpo-2-v2-mkmlizer
Waiting for job on rica40325-feedback-dpo-2-v2-mkmlizer to finish
rica40325-feedback-dpo-2-v2-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
rica40325-feedback-dpo-2-v2-mkmlizer: ║     _____            __           __                                ║
rica40325-feedback-dpo-2-v2-mkmlizer: ║    / _/ /_ ___    __/ /  ___ ___ / /                                ║
rica40325-feedback-dpo-2-v2-mkmlizer: ║   / _/ / // / |/|/ / _ \/ -_) -_) /                                 ║
rica40325-feedback-dpo-2-v2-mkmlizer: ║  /_//_/\_, /|__,__/_//_/\__/\__/_/                                  ║
rica40325-feedback-dpo-2-v2-mkmlizer: ║       /___/                                                         ║
rica40325-feedback-dpo-2-v2-mkmlizer: ║                                                                     ║
rica40325-feedback-dpo-2-v2-mkmlizer: ║  Version: 0.10.1                                                    ║
rica40325-feedback-dpo-2-v2-mkmlizer: ║  Copyright 2023 MK ONE TECHNOLOGIES Inc.                            ║
rica40325-feedback-dpo-2-v2-mkmlizer: ║  https://mk1.ai                                                     ║
rica40325-feedback-dpo-2-v2-mkmlizer: ║                                                                     ║
rica40325-feedback-dpo-2-v2-mkmlizer: ║  The license key for the current software has been verified as      ║
rica40325-feedback-dpo-2-v2-mkmlizer: ║  belonging to:                                                      ║
rica40325-feedback-dpo-2-v2-mkmlizer: ║                                                                     ║
rica40325-feedback-dpo-2-v2-mkmlizer: ║  Chai Research Corp.                                                ║
rica40325-feedback-dpo-2-v2-mkmlizer: ║  Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f                   ║
rica40325-feedback-dpo-2-v2-mkmlizer: ║  Expiration: 2024-10-15 23:59:59                                    ║
rica40325-feedback-dpo-2-v2-mkmlizer: ║                                                                     ║
rica40325-feedback-dpo-2-v2-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
Retrying (%r) after connection broken by '%r': %s
rica40325-feedback-dpo-2-v2-mkmlizer: Downloaded to shared memory in 62.548s
rica40325-feedback-dpo-2-v2-mkmlizer: quantizing model to /dev/shm/model_cache, profile:s0, folder:/tmp/tmpmtyu0qsl, device:0
rica40325-feedback-dpo-2-v2-mkmlizer: Saving flywheel model at /dev/shm/model_cache
Retrying (%r) after connection broken by '%r': %s
Retrying (%r) after connection broken by '%r': %s
rica40325-feedback-dpo-2-v2-mkmlizer: quantized model in 29.106s
rica40325-feedback-dpo-2-v2-mkmlizer: Processed model rica40325/feedback_dpo_2 in 91.654s
rica40325-feedback-dpo-2-v2-mkmlizer: creating bucket guanaco-mkml-models
rica40325-feedback-dpo-2-v2-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
rica40325-feedback-dpo-2-v2-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/rica40325-feedback-dpo-2-v2
rica40325-feedback-dpo-2-v2-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/rica40325-feedback-dpo-2-v2/config.json
rica40325-feedback-dpo-2-v2-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/rica40325-feedback-dpo-2-v2/special_tokens_map.json
rica40325-feedback-dpo-2-v2-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/rica40325-feedback-dpo-2-v2/tokenizer_config.json
rica40325-feedback-dpo-2-v2-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/rica40325-feedback-dpo-2-v2/tokenizer.json
rica40325-feedback-dpo-2-v2-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/rica40325-feedback-dpo-2-v2/flywheel_model.0.safetensors
rica40325-feedback-dpo-2-v2-mkmlizer: 
Loading 0:   0%|          | 0/291 [00:00<?, ?it/s]
Loading 0:   2%|▏         | 5/291 [00:00<00:10, 26.26it/s]
Loading 0:   4%|▍         | 12/291 [00:00<00:07, 36.16it/s]
Loading 0:   5%|▌         | 16/291 [00:00<00:08, 33.12it/s]
Loading 0:   7%|▋         | 21/291 [00:00<00:07, 36.36it/s]
Loading 0:   9%|▊         | 25/291 [00:00<00:07, 34.28it/s]
Loading 0:  11%|█         | 31/291 [00:00<00:06, 40.42it/s]
Loading 0:  12%|█▏        | 36/291 [00:01<00:10, 23.54it/s]
Loading 0:  14%|█▍        | 41/291 [00:01<00:09, 25.16it/s]
Loading 0:  16%|█▋        | 48/291 [00:01<00:07, 31.85it/s]
Loading 0:  18%|█▊        | 52/291 [00:01<00:07, 31.29it/s]
Loading 0:  20%|█▉        | 57/291 [00:01<00:07, 33.32it/s]
Loading 0:  21%|██        | 61/291 [00:01<00:07, 32.53it/s]
Loading 0:  23%|██▎       | 66/291 [00:02<00:06, 35.99it/s]
Loading 0:  24%|██▍       | 70/291 [00:02<00:06, 33.07it/s]
Loading 0:  25%|██▌       | 74/291 [00:02<00:06, 32.17it/s]
Loading 0:  27%|██▋       | 78/291 [00:02<00:06, 32.45it/s]
Loading 0:  28%|██▊       | 82/291 [00:02<00:08, 24.00it/s]
Loading 0:  29%|██▉       | 85/291 [00:02<00:08, 25.16it/s]
Loading 0:  31%|███       | 90/291 [00:02<00:06, 29.63it/s]
Loading 0:  32%|███▏      | 94/291 [00:03<00:06, 29.71it/s]
Loading 0:  34%|███▍      | 99/291 [00:03<00:05, 33.44it/s]
Loading 0:  35%|███▌      | 103/291 [00:03<00:05, 32.12it/s]
Loading 0:  37%|███▋      | 108/291 [00:03<00:05, 34.13it/s]
Loading 0:  38%|███▊      | 112/291 [00:03<00:05, 32.40it/s]
Loading 0:  40%|███▉      | 116/291 [00:03<00:05, 32.43it/s]
Loading 0:  42%|████▏     | 122/291 [00:03<00:04, 37.27it/s]
Loading 0:  44%|████▎     | 127/291 [00:03<00:04, 35.17it/s]
Loading 0:  46%|████▌     | 133/291 [00:04<00:05, 31.19it/s]
Loading 0:  47%|████▋     | 137/291 [00:04<00:05, 30.73it/s]
Loading 0:  48%|████▊     | 141/291 [00:04<00:05, 27.78it/s]
Loading 0:  51%|█████     | 147/291 [00:04<00:04, 32.03it/s]
Loading 0:  52%|█████▏    | 151/291 [00:04<00:04, 31.03it/s]
Loading 0:  54%|█████▎    | 156/291 [00:04<00:03, 34.56it/s]
Loading 0:  55%|█████▍    | 160/291 [00:05<00:03, 33.08it/s]
Loading 0:  57%|█████▋    | 165/291 [00:05<00:03, 35.96it/s]
Loading 0:  58%|█████▊    | 169/291 [00:05<00:03, 33.85it/s]
Loading 0:  60%|█████▉    | 174/291 [00:05<00:03, 36.60it/s]
Loading 0:  61%|██████    | 178/291 [00:05<00:03, 33.51it/s]
Loading 0:  63%|██████▎   | 183/291 [00:05<00:02, 36.44it/s]
Loading 0:  64%|██████▍   | 187/291 [00:05<00:04, 24.66it/s]
Loading 0:  66%|██████▌   | 191/291 [00:06<00:03, 25.76it/s]
Loading 0:  67%|██████▋   | 194/291 [00:06<00:04, 23.17it/s]
Loading 0:  69%|██████▉   | 201/291 [00:06<00:02, 30.48it/s]
Loading 0:  70%|███████   | 205/291 [00:06<00:02, 29.70it/s]
Loading 0:  72%|███████▏  | 210/291 [00:06<00:02, 32.50it/s]
Loading 0:  74%|███████▎  | 214/291 [00:06<00:02, 31.25it/s]
Loading 0:  75%|███████▌  | 219/291 [00:06<00:02, 32.77it/s]
Loading 0:  77%|███████▋  | 223/291 [00:07<00:02, 30.76it/s]
Loading 0:  78%|███████▊  | 227/291 [00:07<00:02, 31.42it/s]
Loading 0:  79%|███████▉  | 231/291 [00:07<00:01, 31.75it/s]
Loading 0:  81%|████████  | 235/291 [00:07<00:02, 24.33it/s]
Loading 0:  82%|████████▏ | 239/291 [00:07<00:02, 24.23it/s]
Loading 0:  85%|████████▍ | 246/291 [00:07<00:01, 31.53it/s]
Loading 0:  86%|████████▌ | 250/291 [00:08<00:01, 30.73it/s]
Loading 0:  88%|████████▊ | 255/291 [00:08<00:01, 32.20it/s]
Loading 0:  89%|████████▉ | 259/291 [00:08<00:01, 31.23it/s]
Loading 0:  91%|█████████ | 264/291 [00:08<00:00, 33.89it/s]
Loading 0:  92%|█████████▏| 268/291 [00:08<00:00, 32.29it/s]
Loading 0:  94%|█████████▍| 273/291 [00:08<00:00, 35.20it/s]
Loading 0:  95%|█████████▌| 277/291 [00:08<00:00, 33.82it/s]
Loading 0:  97%|█████████▋| 281/291 [00:08<00:00, 34.51it/s]
Loading 0:  98%|█████████▊| 286/291 [00:14<00:01,  2.61it/s]
Loading 0:  99%|█████████▉| 289/291 [00:14<00:00,  3.24it/s]
                                                            
Job rica40325-feedback-dpo-2-v2-mkmlizer completed after 114.8s with status: succeeded
Stopping job with name rica40325-feedback-dpo-2-v2-mkmlizer
Pipeline stage MKMLizer completed in 115.74s
run pipeline stage %s
Running pipeline stage MKMLTemplater
Pipeline stage MKMLTemplater completed in 0.08s
run pipeline stage %s
Running pipeline stage MKMLDeployer
Creating inference service rica40325-feedback-dpo-2-v2
Waiting for inference service rica40325-feedback-dpo-2-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
Connection pool is full, discarding connection: %s. Connection pool size: %s
Failed to get response for submission mistralai-mixtral-8x7b_3473_v131: ('http://mistralai-mixtral-8x7b-3473-v131-predictor.tenant-chaiml-guanaco.k.chaiverse.com/v1/models/GPT-J-6B-lit-v2:predict', 'read tcp 127.0.0.1:43612->127.0.0.1:8080: read: connection reset by peer\n')
Connection pool is full, discarding connection: %s. Connection pool size: %s
Inference service rica40325-feedback-dpo-2-v2 ready after 241.74012804031372s
Pipeline stage MKMLDeployer completed in 242.75s
run pipeline stage %s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.0094704627990723s
Received healthy response to inference request in 1.4077284336090088s
Received healthy response to inference request in 1.1898407936096191s
Received healthy response to inference request in 0.6570541858673096s
Received healthy response to inference request in 0.9098339080810547s
5 requests
0 failed requests
5th percentile: 0.7076101303100586
10th percentile: 0.7581660747528076
20th percentile: 0.8592779636383057
30th percentile: 0.9658352851867675
40th percentile: 1.0778380393981934
50th percentile: 1.1898407936096191
60th percentile: 1.276995849609375
70th percentile: 1.3641509056091308
80th percentile: 1.5280768394470217
90th percentile: 1.768773651123047
95th percentile: 1.8891220569610594
99th percentile: 1.9854007816314696
mean time: 1.234785556793213
Pipeline stage StressChecker completed in 7.90s
run pipeline stage %s
Running pipeline stage TriggerMKMLProfilingPipeline
run_pipeline:run_in_cloud %s
starting trigger_guanaco_pipeline args=%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
Connection pool is full, discarding connection: %s. Connection pool size: %s
Connection pool is full, discarding connection: %s. Connection pool size: %s
Pipeline stage TriggerMKMLProfilingPipeline completed in 5.77s
Shutdown handler de-registered
rica40325-feedback-dpo-2_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 0.11s
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 rica40325-feedback-dpo-2-v2-profiler
Waiting for inference service rica40325-feedback-dpo-2-v2-profiler to be ready
Inference service rica40325-feedback-dpo-2-v2-profiler ready after 160.39529943466187s
Pipeline stage MKMLProfilerDeployer completed in 160.79s
run pipeline stage %s
Running pipeline stage MKMLProfilerRunner
kubectl cp /code/guanaco/guanaco_inference_services/src/inference_scripts tenant-chaiml-guanaco/rica40325-feedback-dpo-2-v2-profiler-predictor-00001-deplokd68f:/code/chaiverse_profiler_1725948073 --namespace tenant-chaiml-guanaco
kubectl exec -it rica40325-feedback-dpo-2-v2-profiler-predictor-00001-deplokd68f --namespace tenant-chaiml-guanaco -- sh -c 'cd /code/chaiverse_profiler_1725948073 && python profiles.py profile --best_of_n 1 --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_1725948073/summary.json'
kubectl exec -it rica40325-feedback-dpo-2-v2-profiler-predictor-00001-deplokd68f --namespace tenant-chaiml-guanaco -- bash -c 'cat /code/chaiverse_profiler_1725948073/summary.json'
Pipeline stage MKMLProfilerRunner completed in 462.21s
run pipeline stage %s
Running pipeline stage MKMLProfilerDeleter
Checking if service rica40325-feedback-dpo-2-v2-profiler is running
Tearing down inference service rica40325-feedback-dpo-2-v2-profiler
Service rica40325-feedback-dpo-2-v2-profiler has been torndown
Pipeline stage MKMLProfilerDeleter completed in 1.82s
Shutdown handler de-registered
rica40325-feedback-dpo-2_v2 status is now inactive due to auto deactivation removed underperforming models
rica40325-feedback-dpo-2_v2 status is now torndown due to DeploymentManager action