rica40325-feedback-dpo-5

developer_uid: rica40325

submission_id: rica40325-feedback-dpo-5_v2

model_name: rica40325-feedback-dpo-5_v1

model_group: rica40325/feedback_dpo_5

status: torndown

timestamp: 2024-09-10T08:35:51+00:00

num_battles: 11935

num_wins: 1070

celo_rating: 792.59

family_friendly_score: 0.0

submission_type: basic

model_repo: rica40325/feedback_dpo_5

model_architecture: LlamaForCausalLM

model_num_parameters: 8030261248.0

best_of: 1

max_input_tokens: 512

max_output_tokens: 64

latencies: [{'batch_size': 1, 'throughput': 1.0704963367113298, 'latency_mean': 0.9340853941440582, 'latency_p50': 0.9367431402206421, 'latency_p90': 1.0392061471939087}, {'batch_size': 5, 'throughput': 3.511312614183687, 'latency_mean': 1.4130074179172516, 'latency_p50': 1.4120358228683472, 'latency_p90': 1.5735148191452026}, {'batch_size': 10, 'throughput': 5.122838166157106, 'latency_mean': 1.9125505411624908, 'latency_p50': 1.9230626821517944, 'latency_p90': 2.158908987045288}, {'batch_size': 15, 'throughput': 6.046078213225946, 'latency_mean': 2.424727724790573, 'latency_p50': 2.3988263607025146, 'latency_p90': 2.7779078245162965}, {'batch_size': 20, 'throughput': 6.6272958403997375, 'latency_mean': 2.9647460770606995, 'latency_p50': 2.934940218925476, 'latency_p90': 3.3975889444351197}, {'batch_size': 25, 'throughput': 7.042409813862316, 'latency_mean': 3.4766038930416108, 'latency_p50': 3.468196392059326, 'latency_p90': 4.035438656806946}, {'batch_size': 30, 'throughput': 7.2202109388217375, 'latency_mean': 4.061782602071762, 'latency_p50': 4.0162636041641235, 'latency_p90': 4.8127221107482905}, {'batch_size': 35, 'throughput': 7.485637282910613, 'latency_mean': 4.563777828216553, 'latency_p50': 4.571220517158508, 'latency_p90': 5.474064254760742}, {'batch_size': 40, 'throughput': 7.411394380260749, 'latency_mean': 5.2154414224624634, 'latency_p50': 5.135430693626404, 'latency_p90': 6.149071526527404}]

gpu_counts: {'NVIDIA RTX A5000': 1}

display_name: rica40325-feedback-dpo-5_v1

is_internal_developer: False

language_model: rica40325/feedback_dpo_5

model_size: 8B

ranking_group: single

throughput_3p7s: 7.27

us_pacific_date: 2024-09-10

win_ratio: 0.0896522832006703

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
Starting job with name rica40325-feedback-dpo-5-v2-mkmlizer
Waiting for job on rica40325-feedback-dpo-5-v2-mkmlizer to finish
rica40325-feedback-dpo-5-v2-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
rica40325-feedback-dpo-5-v2-mkmlizer: ║     _____            __           __                                ║
rica40325-feedback-dpo-5-v2-mkmlizer: ║    / _/ /_ ___    __/ /  ___ ___ / /                                ║
rica40325-feedback-dpo-5-v2-mkmlizer: ║   / _/ / // / |/|/ / _ \/ -_) -_) /                                 ║
rica40325-feedback-dpo-5-v2-mkmlizer: ║  /_//_/\_, /|__,__/_//_/\__/\__/_/                                  ║
rica40325-feedback-dpo-5-v2-mkmlizer: ║       /___/                                                         ║
rica40325-feedback-dpo-5-v2-mkmlizer: ║                                                                     ║
rica40325-feedback-dpo-5-v2-mkmlizer: ║  Version: 0.10.1                                                    ║
rica40325-feedback-dpo-5-v2-mkmlizer: ║  Copyright 2023 MK ONE TECHNOLOGIES Inc.                            ║
rica40325-feedback-dpo-5-v2-mkmlizer: ║  https://mk1.ai                                                     ║
rica40325-feedback-dpo-5-v2-mkmlizer: ║                                                                     ║
rica40325-feedback-dpo-5-v2-mkmlizer: ║  The license key for the current software has been verified as      ║
rica40325-feedback-dpo-5-v2-mkmlizer: ║  belonging to:                                                      ║
rica40325-feedback-dpo-5-v2-mkmlizer: ║                                                                     ║
rica40325-feedback-dpo-5-v2-mkmlizer: ║  Chai Research Corp.                                                ║
rica40325-feedback-dpo-5-v2-mkmlizer: ║  Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f                   ║
rica40325-feedback-dpo-5-v2-mkmlizer: ║  Expiration: 2024-10-15 23:59:59                                    ║
rica40325-feedback-dpo-5-v2-mkmlizer: ║                                                                     ║
rica40325-feedback-dpo-5-v2-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
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
rica40325-feedback-dpo-5-v2-mkmlizer: Downloaded to shared memory in 36.946s
rica40325-feedback-dpo-5-v2-mkmlizer: quantizing model to /dev/shm/model_cache, profile:s0, folder:/tmp/tmpqew_wuyl, device:0
rica40325-feedback-dpo-5-v2-mkmlizer: Saving flywheel model at /dev/shm/model_cache
rica40325-feedback-dpo-5-v2-mkmlizer: quantized model in 29.071s
rica40325-feedback-dpo-5-v2-mkmlizer: Processed model rica40325/feedback_dpo_5 in 66.018s
Failed to get response for submission blend_sinam_2024-09-09: ('http://chaiml-elo-alignment-run-3-v34-predictor.tenant-chaiml-guanaco.k.chaiverse.com/v1/models/GPT-J-6B-lit-v2:predict', 'read tcp 127.0.0.1:57090->127.0.0.1:8080: read: connection reset by peer\n')
rica40325-feedback-dpo-5-v2-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
rica40325-feedback-dpo-5-v2-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/rica40325-feedback-dpo-5-v2
rica40325-feedback-dpo-5-v2-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/rica40325-feedback-dpo-5-v2/config.json
rica40325-feedback-dpo-5-v2-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/rica40325-feedback-dpo-5-v2/special_tokens_map.json
rica40325-feedback-dpo-5-v2-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/rica40325-feedback-dpo-5-v2/tokenizer_config.json
rica40325-feedback-dpo-5-v2-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/rica40325-feedback-dpo-5-v2/tokenizer.json
rica40325-feedback-dpo-5-v2-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/rica40325-feedback-dpo-5-v2/flywheel_model.0.safetensors
rica40325-feedback-dpo-5-v2-mkmlizer: 
Loading 0:   0%|          | 0/291 [00:00<?, ?it/s]
Loading 0:   2%|▏         | 5/291 [00:00<00:11, 25.91it/s]
Loading 0:   4%|▍         | 12/291 [00:00<00:07, 36.39it/s]
Loading 0:   5%|▌         | 16/291 [00:00<00:08, 32.83it/s]
Loading 0:   7%|▋         | 21/291 [00:00<00:07, 36.46it/s]
Loading 0:   9%|▊         | 25/291 [00:00<00:07, 34.25it/s]
Loading 0:  11%|█         | 31/291 [00:00<00:06, 40.16it/s]
Loading 0:  12%|█▏        | 36/291 [00:01<00:10, 23.45it/s]
Loading 0:  14%|█▍        | 41/291 [00:01<00:09, 25.11it/s]
Loading 0:  16%|█▋        | 48/291 [00:01<00:07, 32.09it/s]
Loading 0:  18%|█▊        | 52/291 [00:01<00:07, 31.42it/s]
Loading 0:  20%|█▉        | 57/291 [00:01<00:06, 34.28it/s]
Loading 0:  21%|██        | 61/291 [00:01<00:06, 33.32it/s]
Loading 0:  23%|██▎       | 66/291 [00:02<00:06, 36.04it/s]
Loading 0:  24%|██▍       | 70/291 [00:02<00:06, 33.04it/s]
Loading 0:  25%|██▌       | 74/291 [00:02<00:06, 32.80it/s]
Loading 0:  27%|██▋       | 78/291 [00:02<00:06, 32.59it/s]
Loading 0:  28%|██▊       | 82/291 [00:02<00:08, 23.59it/s]
Loading 0:  29%|██▉       | 85/291 [00:02<00:08, 24.35it/s]
Loading 0:  31%|███       | 90/291 [00:02<00:06, 28.78it/s]
Loading 0:  32%|███▏      | 94/291 [00:03<00:06, 29.10it/s]
Loading 0:  34%|███▍      | 99/291 [00:03<00:05, 32.18it/s]
Loading 0:  35%|███▌      | 103/291 [00:03<00:05, 32.35it/s]
Loading 0:  37%|███▋      | 108/291 [00:03<00:05, 35.90it/s]
Loading 0:  38%|███▊      | 112/291 [00:03<00:05, 35.01it/s]
Loading 0:  40%|███▉      | 116/291 [00:03<00:04, 35.71it/s]
Loading 0:  42%|████▏     | 122/291 [00:03<00:04, 40.47it/s]
Loading 0:  44%|████▎     | 127/291 [00:03<00:04, 37.92it/s]
Loading 0:  46%|████▌     | 133/291 [00:04<00:04, 32.60it/s]
Loading 0:  47%|████▋     | 137/291 [00:04<00:04, 31.63it/s]
Loading 0:  48%|████▊     | 141/291 [00:04<00:05, 29.28it/s]
Loading 0:  51%|█████     | 147/291 [00:04<00:04, 34.07it/s]
Loading 0:  52%|█████▏    | 151/291 [00:04<00:04, 32.03it/s]
Loading 0:  54%|█████▎    | 156/291 [00:04<00:03, 34.69it/s]
Loading 0:  55%|█████▍    | 160/291 [00:04<00:03, 33.23it/s]
Loading 0:  57%|█████▋    | 165/291 [00:05<00:03, 35.81it/s]
Loading 0:  58%|█████▊    | 169/291 [00:05<00:03, 33.05it/s]
Loading 0:  60%|█████▉    | 174/291 [00:05<00:03, 34.84it/s]
Loading 0:  61%|██████    | 178/291 [00:05<00:03, 33.33it/s]
Loading 0:  63%|██████▎   | 184/291 [00:05<00:02, 39.30it/s]
Loading 0:  65%|██████▍   | 189/291 [00:06<00:04, 23.42it/s]
Loading 0:  67%|██████▋   | 194/291 [00:06<00:03, 25.04it/s]
Loading 0:  68%|██████▊   | 199/291 [00:06<00:03, 29.23it/s]
Loading 0:  70%|██████▉   | 203/291 [00:06<00:03, 27.52it/s]
Loading 0:  72%|███████▏  | 210/291 [00:06<00:02, 34.34it/s]
Loading 0:  74%|███████▎  | 214/291 [00:06<00:02, 32.87it/s]
Loading 0:  75%|███████▌  | 219/291 [00:06<00:02, 35.33it/s]
Loading 0:  77%|███████▋  | 223/291 [00:06<00:02, 33.83it/s]
Loading 0:  78%|███████▊  | 227/291 [00:07<00:01, 34.09it/s]
Loading 0:  79%|███████▉  | 231/291 [00:07<00:01, 33.87it/s]
Loading 0:  81%|████████  | 235/291 [00:07<00:02, 24.80it/s]
Loading 0:  82%|████████▏ | 239/291 [00:07<00:02, 24.44it/s]
Loading 0:  85%|████████▍ | 246/291 [00:07<00:01, 32.06it/s]
Loading 0:  86%|████████▌ | 250/291 [00:07<00:01, 31.30it/s]
Loading 0:  88%|████████▊ | 255/291 [00:08<00:01, 34.07it/s]
Loading 0:  89%|████████▉ | 259/291 [00:08<00:00, 32.54it/s]
Loading 0:  91%|█████████ | 264/291 [00:08<00:00, 35.15it/s]
Loading 0:  92%|█████████▏| 268/291 [00:08<00:00, 33.04it/s]
Loading 0:  94%|█████████▍| 273/291 [00:08<00:00, 35.60it/s]
Loading 0:  95%|█████████▌| 277/291 [00:08<00:00, 33.79it/s]
Loading 0:  97%|█████████▋| 281/291 [00:08<00:00, 33.93it/s]
Loading 0:  98%|█████████▊| 286/291 [00:14<00:01,  2.58it/s]
Loading 0:  99%|█████████▉| 289/291 [00:14<00:00,  3.20it/s]
                                                            
Job rica40325-feedback-dpo-5-v2-mkmlizer completed after 84.1s with status: succeeded
Stopping job with name rica40325-feedback-dpo-5-v2-mkmlizer
Pipeline stage MKMLizer completed in 86.07s
run pipeline stage %s
Running pipeline stage MKMLTemplater
Pipeline stage MKMLTemplater completed in 0.28s
run pipeline stage %s
Running pipeline stage MKMLDeployer
Creating inference service rica40325-feedback-dpo-5-v2
Waiting for inference service rica40325-feedback-dpo-5-v2 to be ready
Failed to get response for submission trace2333-mistral-trail7_v1: ('http://trace2333-mistral-trail7-v1-predictor.tenant-chaiml-guanaco.k.chaiverse.com/v1/models/GPT-J-6B-lit-v2:predict', 'read tcp 127.0.0.1:60304->127.0.0.1:8080: read: connection reset by peer\n')
Inference service rica40325-feedback-dpo-5-v2 ready after 161.18867373466492s
Pipeline stage MKMLDeployer completed in 161.66s
run pipeline stage %s
Running pipeline stage StressChecker
Received healthy response to inference request in 1.9022469520568848s
Received healthy response to inference request in 1.8643932342529297s
Received healthy response to inference request in 0.5432348251342773s
Received healthy response to inference request in 2.1129510402679443s
Received healthy response to inference request in 0.530104398727417s
5 requests
0 failed requests
5th percentile: 0.5327304840087891
10th percentile: 0.5353565692901612
20th percentile: 0.5406087398529053
30th percentile: 0.8074665069580078
40th percentile: 1.3359298706054688
50th percentile: 1.8643932342529297
60th percentile: 1.8795347213745117
70th percentile: 1.8946762084960938
80th percentile: 1.9443877696990968
90th percentile: 2.0286694049835203
95th percentile: 2.0708102226257323
99th percentile: 2.104522876739502
mean time: 1.3905860900878906
Pipeline stage StressChecker completed in 7.73s
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 6.24s
Shutdown handler de-registered
rica40325-feedback-dpo-5_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 0.13s
run pipeline stage %s
Running pipeline stage MKMLProfilerTemplater
Pipeline stage MKMLProfilerTemplater completed in 0.13s
run pipeline stage %s
Running pipeline stage MKMLProfilerDeployer
Creating inference service rica40325-feedback-dpo-5-v2-profiler
Waiting for inference service rica40325-feedback-dpo-5-v2-profiler to be ready
Inference service rica40325-feedback-dpo-5-v2-profiler ready after 160.415931224823s
Pipeline stage MKMLProfilerDeployer completed in 160.81s
run pipeline stage %s
Running pipeline stage MKMLProfilerRunner
kubectl cp /code/guanaco/guanaco_inference_services/src/inference_scripts tenant-chaiml-guanaco/rica40325-feedback-dpo-5-v2-profiler-predictor-00001-deplomcftr:/code/chaiverse_profiler_1725957858 --namespace tenant-chaiml-guanaco
kubectl exec -it rica40325-feedback-dpo-5-v2-profiler-predictor-00001-deplomcftr --namespace tenant-chaiml-guanaco -- sh -c 'cd /code/chaiverse_profiler_1725957858 && 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_1725957858/summary.json'
kubectl exec -it rica40325-feedback-dpo-5-v2-profiler-predictor-00001-deplomcftr --namespace tenant-chaiml-guanaco -- bash -c 'cat /code/chaiverse_profiler_1725957858/summary.json'
Pipeline stage MKMLProfilerRunner completed in 461.48s
run pipeline stage %s
Running pipeline stage MKMLProfilerDeleter
Checking if service rica40325-feedback-dpo-5-v2-profiler is running
Tearing down inference service rica40325-feedback-dpo-5-v2-profiler
Service rica40325-feedback-dpo-5-v2-profiler has been torndown
Pipeline stage MKMLProfilerDeleter completed in 1.93s
Shutdown handler de-registered
rica40325-feedback-dpo-5_v2 status is now inactive due to auto deactivation removed underperforming models
rica40325-feedback-dpo-5_v2 status is now torndown due to DeploymentManager action