submission_id: rica40325-9-15-sftdpo_v6
developer_uid: rica40325
best_of: 16
celo_rating: 1261.58
display_name: rica40325-9-15-sftdpo_v1
family_friendly_score: 0.0
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': 1.15, 'top_p': 1.0, 'min_p': 0.0, 'top_k': 50, '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: rica40325/9-15-sftdpo
latencies: [{'batch_size': 1, 'throughput': 0.9059236508694792, 'latency_mean': 1.1037821209430694, 'latency_p50': 1.103683352470398, 'latency_p90': 1.2281744956970215}, {'batch_size': 4, 'throughput': 1.7920460600901102, 'latency_mean': 2.2255457866191866, 'latency_p50': 2.2326905727386475, 'latency_p90': 2.470873975753784}, {'batch_size': 5, 'throughput': 1.8824270268330823, 'latency_mean': 2.640229424238205, 'latency_p50': 2.6324115991592407, 'latency_p90': 2.9641838550567625}, {'batch_size': 8, 'throughput': 1.9895892398194615, 'latency_mean': 3.9906913053989412, 'latency_p50': 4.033903002738953, 'latency_p90': 4.461517119407654}, {'batch_size': 10, 'throughput': 2.0114789092635452, 'latency_mean': 4.92627485871315, 'latency_p50': 4.885356664657593, 'latency_p90': 5.754275465011596}, {'batch_size': 12, 'throughput': 2.0434252318738313, 'latency_mean': 5.792347919940949, 'latency_p50': 5.78665816783905, 'latency_p90': 6.6523994445800785}, {'batch_size': 15, 'throughput': 2.0253154337172554, 'latency_mean': 7.258609280586243, 'latency_p50': 7.349904537200928, 'latency_p90': 7.98691418170929}]
max_input_tokens: 512
max_output_tokens: 64
model_architecture: LlamaForCausalLM
model_group: rica40325/9-15-sftdpo
model_name: rica40325-9-15-sftdpo_v1
model_num_parameters: 8030261248.0
model_repo: rica40325/9-15-sftdpo
model_size: 8B
num_battles: 11973
num_wins: 6146
ranking_group: single
status: torndown
submission_type: basic
throughput_3p7s: 1.99
timestamp: 2024-09-17T03:19:36+00:00
us_pacific_date: 2024-09-16
win_ratio: 0.513321640357471
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 rica40325-9-15-sftdpo-v6-mkmlizer
Waiting for job on rica40325-9-15-sftdpo-v6-mkmlizer to finish
rica40325-9-15-sftdpo-v6-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
rica40325-9-15-sftdpo-v6-mkmlizer: ║ _____ __ __ ║
rica40325-9-15-sftdpo-v6-mkmlizer: ║ / _/ /_ ___ __/ / ___ ___ / / ║
rica40325-9-15-sftdpo-v6-mkmlizer: ║ / _/ / // / |/|/ / _ \/ -_) -_) / ║
rica40325-9-15-sftdpo-v6-mkmlizer: ║ /_//_/\_, /|__,__/_//_/\__/\__/_/ ║
rica40325-9-15-sftdpo-v6-mkmlizer: ║ /___/ ║
rica40325-9-15-sftdpo-v6-mkmlizer: ║ ║
rica40325-9-15-sftdpo-v6-mkmlizer: ║ Version: 0.10.1 ║
rica40325-9-15-sftdpo-v6-mkmlizer: ║ Copyright 2023 MK ONE TECHNOLOGIES Inc. ║
rica40325-9-15-sftdpo-v6-mkmlizer: ║ https://mk1.ai ║
rica40325-9-15-sftdpo-v6-mkmlizer: ║ ║
rica40325-9-15-sftdpo-v6-mkmlizer: ║ The license key for the current software has been verified as ║
rica40325-9-15-sftdpo-v6-mkmlizer: ║ belonging to: ║
rica40325-9-15-sftdpo-v6-mkmlizer: ║ ║
rica40325-9-15-sftdpo-v6-mkmlizer: ║ Chai Research Corp. ║
rica40325-9-15-sftdpo-v6-mkmlizer: ║ Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f ║
rica40325-9-15-sftdpo-v6-mkmlizer: ║ Expiration: 2024-10-15 23:59:59 ║
rica40325-9-15-sftdpo-v6-mkmlizer: ║ ║
rica40325-9-15-sftdpo-v6-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
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', '')
rica40325-9-15-sftdpo-v6-mkmlizer: Downloaded to shared memory in 38.181s
rica40325-9-15-sftdpo-v6-mkmlizer: quantizing model to /dev/shm/model_cache, profile:s0, folder:/tmp/tmptniukvz6, device:0
rica40325-9-15-sftdpo-v6-mkmlizer: Saving flywheel model at /dev/shm/model_cache
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
rica40325-9-15-sftdpo-v6-mkmlizer: quantized model in 25.574s
rica40325-9-15-sftdpo-v6-mkmlizer: Processed model rica40325/9-15-sftdpo in 63.755s
rica40325-9-15-sftdpo-v6-mkmlizer: creating bucket guanaco-mkml-models
rica40325-9-15-sftdpo-v6-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
rica40325-9-15-sftdpo-v6-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/rica40325-9-15-sftdpo-v6
rica40325-9-15-sftdpo-v6-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/rica40325-9-15-sftdpo-v6/config.json
rica40325-9-15-sftdpo-v6-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/rica40325-9-15-sftdpo-v6/special_tokens_map.json
rica40325-9-15-sftdpo-v6-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/rica40325-9-15-sftdpo-v6/tokenizer_config.json
rica40325-9-15-sftdpo-v6-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/rica40325-9-15-sftdpo-v6/tokenizer.json
rica40325-9-15-sftdpo-v6-mkmlizer: Loading 0: 0%| | 0/291 [00:00<?, ?it/s] Loading 0: 1%|▏ | 4/291 [00:00<00:08, 35.44it/s] Loading 0: 5%|▌ | 16/291 [00:00<00:03, 71.12it/s] Loading 0: 11%|█ | 31/291 [00:00<00:02, 92.53it/s] Loading 0: 15%|█▍ | 43/291 [00:00<00:02, 94.10it/s] Loading 0: 18%|█▊ | 53/291 [00:00<00:02, 90.76it/s] Loading 0: 23%|██▎ | 67/291 [00:00<00:02, 95.47it/s] Loading 0: 27%|██▋ | 79/291 [00:00<00:02, 92.40it/s] Loading 0: 31%|███ | 89/291 [00:02<00:08, 24.70it/s] Loading 0: 35%|███▌ | 103/291 [00:02<00:05, 33.25it/s] Loading 0: 40%|███▉ | 115/291 [00:02<00:04, 40.73it/s] Loading 0: 45%|████▍ | 130/291 [00:02<00:03, 52.46it/s] Loading 0: 49%|████▉ | 142/291 [00:02<00:02, 58.22it/s] Loading 0: 52%|█████▏ | 151/291 [00:02<00:02, 62.57it/s] Loading 0: 57%|█████▋ | 166/291 [00:02<00:01, 75.62it/s] Loading 0: 61%|██████ | 178/291 [00:03<00:01, 80.03it/s] Loading 0: 65%|██████▍ | 188/291 [00:04<00:04, 25.69it/s] Loading 0: 67%|██████▋ | 196/291 [00:04<00:03, 29.41it/s] Loading 0: 73%|███████▎ | 211/291 [00:04<00:01, 40.57it/s] Loading 0: 77%|███████▋ | 223/291 [00:04<00:01, 49.00it/s] Loading 0: 82%|████████▏ | 238/291 [00:04<00:00, 60.40it/s] Loading 0: 85%|████████▍ | 247/291 [00:04<00:00, 64.14it/s] Loading 0: 89%|████████▉ | 259/291 [00:04<00:00, 69.78it/s] Loading 0: 92%|█████████▏| 268/291 [00:05<00:00, 73.85it/s] Loading 0: 95%|█████████▌| 277/291 [00:05<00:00, 74.38it/s] Loading 0: 99%|█████████▊| 287/291 [00:05<00:00, 45.17it/s]
Job rica40325-9-15-sftdpo-v6-mkmlizer completed after 94.73s with status: succeeded
Stopping job with name rica40325-9-15-sftdpo-v6-mkmlizer
Pipeline stage MKMLizer completed in 95.93s
run pipeline stage %s
Running pipeline stage MKMLTemplater
Pipeline stage MKMLTemplater completed in 0.09s
run pipeline stage %s
Running pipeline stage MKMLDeployer
Creating inference service rica40325-9-15-sftdpo-v6
Waiting for inference service rica40325-9-15-sftdpo-v6 to be ready
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', '')
Inference service rica40325-9-15-sftdpo-v6 ready after 171.17106556892395s
Pipeline stage MKMLDeployer completed in 171.84s
run pipeline stage %s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.4002997875213623s
Received healthy response to inference request in 1.5165464878082275s
Received healthy response to inference request in 1.809330940246582s
Received healthy response to inference request in 1.9754047393798828s
Received healthy response to inference request in 1.6879773139953613s
5 requests
0 failed requests
5th percentile: 1.5508326530456542
10th percentile: 1.585118818283081
20th percentile: 1.6536911487579347
30th percentile: 1.7122480392456054
40th percentile: 1.7607894897460938
50th percentile: 1.809330940246582
60th percentile: 1.8757604598999023
70th percentile: 1.9421899795532227
80th percentile: 2.060383749008179
90th percentile: 2.2303417682647706
95th percentile: 2.3153207778930662
99th percentile: 2.383303985595703
mean time: 1.8779118537902832
Pipeline stage StressChecker completed in 11.04s
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 4.50s
Shutdown handler de-registered
rica40325-9-15-sftdpo_v6 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 0.12s
run pipeline stage %s
Running pipeline stage MKMLProfilerDeployer
Creating inference service rica40325-9-15-sftdpo-v6-profiler
Waiting for inference service rica40325-9-15-sftdpo-v6-profiler to be ready
Inference service rica40325-9-15-sftdpo-v6-profiler ready after 170.42860460281372s
Pipeline stage MKMLProfilerDeployer completed in 170.83s
run pipeline stage %s
Running pipeline stage MKMLProfilerRunner
kubectl cp /code/guanaco/guanaco_inference_services/src/inference_scripts tenant-chaiml-guanaco/rica40325-9-15-sftdpo-v6-profiler-predictor-00001-deployme7mfrx:/code/chaiverse_profiler_1726543678 --namespace tenant-chaiml-guanaco
kubectl exec -it rica40325-9-15-sftdpo-v6-profiler-predictor-00001-deployme7mfrx --namespace tenant-chaiml-guanaco -- sh -c 'cd /code/chaiverse_profiler_1726543678 && 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_1726543678/summary.json'
kubectl exec -it rica40325-9-15-sftdpo-v6-profiler-predictor-00001-deployme7mfrx --namespace tenant-chaiml-guanaco -- bash -c 'cat /code/chaiverse_profiler_1726543678/summary.json'
Pipeline stage MKMLProfilerRunner completed in 840.84s
run pipeline stage %s
Running pipeline stage MKMLProfilerDeleter
Checking if service rica40325-9-15-sftdpo-v6-profiler is running
Tearing down inference service rica40325-9-15-sftdpo-v6-profiler
Service rica40325-9-15-sftdpo-v6-profiler has been torndown
Pipeline stage MKMLProfilerDeleter completed in 1.81s
Shutdown handler de-registered
rica40325-9-15-sftdpo_v6 status is now inactive due to auto deactivation removed underperforming models
rica40325-9-15-sftdpo_v6 status is now torndown due to DeploymentManager action