rica40325-feedback-dpo-2

submission_id: rica40325-feedback-dpo-2_v1

developer_uid: rica40325

alignment_samples: 11444

alignment_score: 1.6011314468404716

best_of: 16

celo_rating: 1067.94

display_name: rica40325-feedback-dpo-2_v1

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}

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': 16, 'max_output_tokens': 64}

gpu_counts: {'NVIDIA RTX A5000': 1}

is_internal_developer: False

language_model: rica40325/feedback_dpo_2

latencies: [{'batch_size': 1, 'throughput': 0.9095997385886189, 'latency_mean': 1.0992883145809174, 'latency_p50': 1.092139720916748, 'latency_p90': 1.2234450340270997}, {'batch_size': 4, 'throughput': 1.8420331448565834, 'latency_mean': 2.1622723400592805, 'latency_p50': 2.1470340490341187, 'latency_p90': 2.428829312324524}, {'batch_size': 5, 'throughput': 1.8873014257560319, 'latency_mean': 2.637916258573532, 'latency_p50': 2.6523847579956055, 'latency_p90': 2.9407574415206907}, {'batch_size': 8, 'throughput': 2.0065364334246216, 'latency_mean': 3.956706461906433, 'latency_p50': 3.9436265230178833, 'latency_p90': 4.4327685832977295}, {'batch_size': 10, 'throughput': 2.0415229047390806, 'latency_mean': 4.844008741378784, 'latency_p50': 4.775707244873047, 'latency_p90': 5.6778518676757805}, {'batch_size': 12, 'throughput': 2.0487447757467128, 'latency_mean': 5.772784374952316, 'latency_p50': 5.7916271686553955, 'latency_p90': 6.4897768020629885}, {'batch_size': 15, 'throughput': 2.0497700084299355, 'latency_mean': 7.1779693520069126, 'latency_p50': 7.321241140365601, 'latency_p90': 7.994797992706299}]

max_input_tokens: 512

max_output_tokens: 64

model_architecture: LlamaForCausalLM

model_group: rica40325/feedback_dpo_2

model_name: rica40325-feedback-dpo-2_v1

model_num_parameters: 8030261248.0

model_repo: rica40325/feedback_dpo_2

model_size: 8B

num_battles: 11444

num_wins: 3294

propriety_score: 0.7

propriety_total_count: 1040.0

ranking_group: single

status: inactive

submission_type: basic

throughput_3p7s: 2.0

timestamp: 2024-09-10T05:19:25+00:00

us_pacific_date: 2024-09-09

win_ratio: 0.287836420831877

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 rica40325-feedback-dpo-2-v1-mkmlizer
Waiting for job on rica40325-feedback-dpo-2-v1-mkmlizer to finish
rica40325-feedback-dpo-2-v1-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
rica40325-feedback-dpo-2-v1-mkmlizer: ║     _____            __           __                                ║
rica40325-feedback-dpo-2-v1-mkmlizer: ║    / _/ /_ ___    __/ /  ___ ___ / /                                ║
rica40325-feedback-dpo-2-v1-mkmlizer: ║   / _/ / // / |/|/ / _ \/ -_) -_) /                                 ║
rica40325-feedback-dpo-2-v1-mkmlizer: ║  /_//_/\_, /|__,__/_//_/\__/\__/_/                                  ║
rica40325-feedback-dpo-2-v1-mkmlizer: ║       /___/                                                         ║
rica40325-feedback-dpo-2-v1-mkmlizer: ║                                                                     ║
rica40325-feedback-dpo-2-v1-mkmlizer: ║  Version: 0.10.1                                                    ║
rica40325-feedback-dpo-2-v1-mkmlizer: ║  Copyright 2023 MK ONE TECHNOLOGIES Inc.                            ║
rica40325-feedback-dpo-2-v1-mkmlizer: ║  https://mk1.ai                                                     ║
rica40325-feedback-dpo-2-v1-mkmlizer: ║                                                                     ║
rica40325-feedback-dpo-2-v1-mkmlizer: ║  The license key for the current software has been verified as      ║
rica40325-feedback-dpo-2-v1-mkmlizer: ║  belonging to:                                                      ║
rica40325-feedback-dpo-2-v1-mkmlizer: ║                                                                     ║
rica40325-feedback-dpo-2-v1-mkmlizer: ║  Chai Research Corp.                                                ║
rica40325-feedback-dpo-2-v1-mkmlizer: ║  Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f                   ║
rica40325-feedback-dpo-2-v1-mkmlizer: ║  Expiration: 2024-10-15 23:59:59                                    ║
rica40325-feedback-dpo-2-v1-mkmlizer: ║                                                                     ║
rica40325-feedback-dpo-2-v1-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
rica40325-feedback-dpo-2-v1-mkmlizer: Downloaded to shared memory in 70.243s
rica40325-feedback-dpo-2-v1-mkmlizer: quantizing model to /dev/shm/model_cache, profile:s0, folder:/tmp/tmp02pqq1eb, device:0
rica40325-feedback-dpo-2-v1-mkmlizer: Saving flywheel model at /dev/shm/model_cache
rica40325-feedback-dpo-2-v1-mkmlizer: quantized model in 29.335s
rica40325-feedback-dpo-2-v1-mkmlizer: Processed model rica40325/feedback_dpo_2 in 99.578s
rica40325-feedback-dpo-2-v1-mkmlizer: creating bucket guanaco-mkml-models
rica40325-feedback-dpo-2-v1-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
rica40325-feedback-dpo-2-v1-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/rica40325-feedback-dpo-2-v1
rica40325-feedback-dpo-2-v1-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/rica40325-feedback-dpo-2-v1/config.json
rica40325-feedback-dpo-2-v1-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/rica40325-feedback-dpo-2-v1/special_tokens_map.json
rica40325-feedback-dpo-2-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/rica40325-feedback-dpo-2-v1/tokenizer_config.json
rica40325-feedback-dpo-2-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/rica40325-feedback-dpo-2-v1/tokenizer.json
rica40325-feedback-dpo-2-v1-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/rica40325-feedback-dpo-2-v1/flywheel_model.0.safetensors
rica40325-feedback-dpo-2-v1-mkmlizer: 
Loading 0:   0%|          | 0/291 [00:00<?, ?it/s]
Loading 0:   2%|▏         | 5/291 [00:00<00:11, 24.94it/s]
Loading 0:   4%|▍         | 12/291 [00:00<00:08, 34.04it/s]
Loading 0:   5%|▌         | 16/291 [00:00<00:08, 32.45it/s]
Loading 0:   7%|▋         | 21/291 [00:00<00:07, 34.86it/s]
Loading 0:   9%|▊         | 25/291 [00:00<00:08, 31.50it/s]
Loading 0:  10%|▉         | 29/291 [00:00<00:07, 33.12it/s]
Loading 0:  11%|█▏        | 33/291 [00:01<00:10, 23.55it/s]
Loading 0:  12%|█▏        | 36/291 [00:01<00:11, 21.31it/s]
Loading 0:  14%|█▍        | 41/291 [00:01<00:10, 23.01it/s]
Loading 0:  16%|█▋        | 48/291 [00:01<00:08, 30.19it/s]
Loading 0:  18%|█▊        | 52/291 [00:01<00:07, 30.36it/s]
Loading 0:  20%|█▉        | 57/291 [00:01<00:06, 33.64it/s]
Loading 0:  21%|██        | 61/291 [00:02<00:07, 32.25it/s]
Loading 0:  23%|██▎       | 66/291 [00:02<00:06, 35.47it/s]
Loading 0:  24%|██▍       | 70/291 [00:02<00:06, 32.71it/s]
Loading 0:  25%|██▌       | 74/291 [00:02<00:06, 32.65it/s]
Loading 0:  27%|██▋       | 78/291 [00:02<00:06, 32.56it/s]
Loading 0:  28%|██▊       | 82/291 [00:02<00:09, 22.77it/s]
Loading 0:  29%|██▉       | 85/291 [00:02<00:08, 23.46it/s]
Loading 0:  31%|███       | 90/291 [00:03<00:07, 27.92it/s]
Loading 0:  32%|███▏      | 94/291 [00:03<00:07, 27.64it/s]
Loading 0:  34%|███▍      | 99/291 [00:03<00:06, 31.46it/s]
Loading 0:  35%|███▌      | 103/291 [00:03<00:06, 31.12it/s]
Loading 0:  37%|███▋      | 108/291 [00:03<00:05, 34.30it/s]
Loading 0:  38%|███▊      | 112/291 [00:03<00:05, 31.33it/s]
Loading 0:  40%|███▉      | 116/291 [00:03<00:05, 30.71it/s]
Loading 0:  42%|████▏     | 122/291 [00:04<00:04, 35.29it/s]
Loading 0:  44%|████▎     | 127/291 [00:04<00:04, 33.81it/s]
Loading 0:  46%|████▌     | 133/291 [00:04<00:05, 30.83it/s]
Loading 0:  47%|████▋     | 137/291 [00:04<00:05, 30.20it/s]
Loading 0:  48%|████▊     | 141/291 [00:04<00:05, 27.87it/s]
Loading 0:  51%|█████     | 147/291 [00:04<00:04, 32.73it/s]
Loading 0:  52%|█████▏    | 151/291 [00:05<00:04, 31.11it/s]
Loading 0:  54%|█████▎    | 156/291 [00:05<00:03, 34.01it/s]
Loading 0:  55%|█████▍    | 160/291 [00:05<00:03, 32.79it/s]
Loading 0:  57%|█████▋    | 165/291 [00:05<00:03, 35.70it/s]
Loading 0:  58%|█████▊    | 169/291 [00:05<00:03, 32.94it/s]
Loading 0:  60%|█████▉    | 174/291 [00:05<00:03, 35.42it/s]
Loading 0:  61%|██████    | 178/291 [00:05<00:03, 33.84it/s]
Loading 0:  64%|██████▎   | 185/291 [00:05<00:02, 38.14it/s]
Loading 0:  65%|██████▍   | 189/291 [00:06<00:04, 25.01it/s]
Loading 0:  67%|██████▋   | 194/291 [00:06<00:03, 26.04it/s]
Loading 0:  69%|██████▉   | 201/291 [00:06<00:02, 32.80it/s]
Loading 0:  70%|███████   | 205/291 [00:06<00:02, 32.13it/s]
Loading 0:  72%|███████▏  | 210/291 [00:06<00:02, 34.53it/s]
Loading 0:  74%|███████▎  | 214/291 [00:06<00:02, 33.12it/s]
Loading 0:  75%|███████▌  | 219/291 [00:07<00:01, 36.44it/s]
Loading 0:  77%|███████▋  | 223/291 [00:07<00:02, 33.86it/s]
Loading 0:  78%|███████▊  | 227/291 [00:07<00:01, 34.61it/s]
Loading 0:  79%|███████▉  | 231/291 [00:07<00:01, 34.50it/s]
Loading 0:  81%|████████  | 235/291 [00:07<00:02, 26.05it/s]
Loading 0:  82%|████████▏ | 239/291 [00:07<00:02, 25.22it/s]
Loading 0:  85%|████████▍ | 246/291 [00:07<00:01, 33.18it/s]
Loading 0:  86%|████████▌ | 250/291 [00:08<00:01, 32.61it/s]
Loading 0:  88%|████████▊ | 255/291 [00:08<00:01, 35.80it/s]
Loading 0:  89%|████████▉ | 259/291 [00:08<00:00, 33.35it/s]
Loading 0:  91%|█████████ | 264/291 [00:08<00:00, 34.37it/s]
Loading 0:  92%|█████████▏| 268/291 [00:08<00:00, 33.16it/s]
Loading 0:  94%|█████████▍| 273/291 [00:08<00:00, 36.24it/s]
Loading 0:  95%|█████████▌| 277/291 [00:08<00:00, 33.62it/s]
Loading 0:  97%|█████████▋| 281/291 [00:08<00:00, 33.41it/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-2-v1-mkmlizer completed after 125.83s with status: succeeded
Stopping job with name rica40325-feedback-dpo-2-v1-mkmlizer
Pipeline stage MKMLizer completed in 127.12s
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-2-v1
Waiting for inference service rica40325-feedback-dpo-2-v1 to be ready
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', '{"error":"ValueError : [TypeError(\\"\'numpy.int64\' object is not iterable\\"), TypeError(\'vars() argument must have __dict__ attribute\')]"}')
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
Inference service rica40325-feedback-dpo-2-v1 ready after 161.60350728034973s
Pipeline stage MKMLDeployer completed in 162.23s
run pipeline stage %s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.4899091720581055s
Received healthy response to inference request in 1.6093761920928955s
Received healthy response to inference request in 1.707568645477295s
Received healthy response to inference request in 1.588245153427124s
Received healthy response to inference request in 1.6531052589416504s
5 requests
0 failed requests
5th percentile: 1.5924713611602783
10th percentile: 1.5966975688934326
20th percentile: 1.6051499843597412
30th percentile: 1.6181220054626464
40th percentile: 1.6356136322021484
50th percentile: 1.6531052589416504
60th percentile: 1.6748906135559083
70th percentile: 1.696675968170166
80th percentile: 1.8640367507934572
90th percentile: 2.1769729614257813
95th percentile: 2.333441066741943
99th percentile: 2.458615550994873
mean time: 1.809640884399414
Pipeline stage StressChecker completed in 10.26s
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
Pipeline stage TriggerMKMLProfilingPipeline completed in 8.80s
Shutdown handler de-registered
rica40325-feedback-dpo-2_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 rica40325-feedback-dpo-2-v1-profiler
Waiting for inference service rica40325-feedback-dpo-2-v1-profiler to be ready
Inference service rica40325-feedback-dpo-2-v1-profiler ready after 160.41207766532898s
Pipeline stage MKMLProfilerDeployer completed in 160.76s
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-v1-profiler-predictor-00001-deplo22zmj:/code/chaiverse_profiler_1725946075 --namespace tenant-chaiml-guanaco
kubectl exec -it rica40325-feedback-dpo-2-v1-profiler-predictor-00001-deplo22zmj --namespace tenant-chaiml-guanaco -- sh -c 'cd /code/chaiverse_profiler_1725946075 && 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_1725946075/summary.json'
kubectl exec -it rica40325-feedback-dpo-2-v1-profiler-predictor-00001-deplo22zmj --namespace tenant-chaiml-guanaco -- bash -c 'cat /code/chaiverse_profiler_1725946075/summary.json'
Pipeline stage MKMLProfilerRunner completed in 833.00s
run pipeline stage %s
Running pipeline stage MKMLProfilerDeleter
Checking if service rica40325-feedback-dpo-2-v1-profiler is running
Tearing down inference service rica40325-feedback-dpo-2-v1-profiler
Service rica40325-feedback-dpo-2-v1-profiler has been torndown
Pipeline stage MKMLProfilerDeleter completed in 1.74s
Shutdown handler de-registered
rica40325-feedback-dpo-2_v1 status is now inactive due to auto deactivation removed underperforming models