rinen0721-llama1023

submission_id: rinen0721-llama1023_v1

developer_uid: rinen0721

best_of: 8

celo_rating: 1214.03

display_name: rinen0721-llama1023_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': 1.0, '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': 1024, 'best_of': 8, 'max_output_tokens': 64}

is_internal_developer: False

language_model: rinen0721/llama1023

max_input_tokens: 1024

max_output_tokens: 64

model_architecture: LlamaForCausalLM

model_group: rinen0721/llama1023

model_name: rinen0721-llama1023_v1

model_num_parameters: 8030261248.0

model_repo: rinen0721/llama1023

model_size: 8B

num_battles: 13490

num_wins: 6155

ranking_group: single

status: torndown

submission_type: basic

timestamp: 2024-10-23T02:12:45+00:00

us_pacific_date: 2024-10-22

win_ratio: 0.45626389918458116

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 rinen0721-llama1023-v1-mkmlizer
Waiting for job on rinen0721-llama1023-v1-mkmlizer to finish
rinen0721-llama1023-v1-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
rinen0721-llama1023-v1-mkmlizer: ║     _____            __           __                                ║
rinen0721-llama1023-v1-mkmlizer: ║    / _/ /_ ___    __/ /  ___ ___ / /                                ║
rinen0721-llama1023-v1-mkmlizer: ║   / _/ / // / |/|/ / _ \/ -_) -_) /                                 ║
rinen0721-llama1023-v1-mkmlizer: ║  /_//_/\_, /|__,__/_//_/\__/\__/_/                                  ║
rinen0721-llama1023-v1-mkmlizer: ║       /___/                                                         ║
rinen0721-llama1023-v1-mkmlizer: ║                                                                     ║
rinen0721-llama1023-v1-mkmlizer: ║  Version: 0.11.12                                                   ║
rinen0721-llama1023-v1-mkmlizer: ║  Copyright 2023 MK ONE TECHNOLOGIES Inc.                            ║
rinen0721-llama1023-v1-mkmlizer: ║  https://mk1.ai                                                     ║
rinen0721-llama1023-v1-mkmlizer: ║                                                                     ║
rinen0721-llama1023-v1-mkmlizer: ║  The license key for the current software has been verified as      ║
rinen0721-llama1023-v1-mkmlizer: ║  belonging to:                                                      ║
rinen0721-llama1023-v1-mkmlizer: ║                                                                     ║
rinen0721-llama1023-v1-mkmlizer: ║  Chai Research Corp.                                                ║
rinen0721-llama1023-v1-mkmlizer: ║  Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f                   ║
rinen0721-llama1023-v1-mkmlizer: ║  Expiration: 2025-01-15 23:59:59                                    ║
rinen0721-llama1023-v1-mkmlizer: ║                                                                     ║
rinen0721-llama1023-v1-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
rinen0721-llama1023-v1-mkmlizer: Downloaded to shared memory in 66.911s
rinen0721-llama1023-v1-mkmlizer: quantizing model to /dev/shm/model_cache, profile:s0, folder:/tmp/tmphpfexamg, device:0
rinen0721-llama1023-v1-mkmlizer: Saving flywheel model at /dev/shm/model_cache
rinen0721-llama1023-v1-mkmlizer: /opt/conda/lib/python3.10/site-packages/mk1/flywheel/functional/loader.py:55: FutureWarning: You are using `torch.load` with `weights_only=False` (the current default value), which uses the default pickle module implicitly. It is possible to construct malicious pickle data which will execute arbitrary code during unpickling (See https://github.com/pytorch/pytorch/blob/main/SECURITY.md#untrusted-models for more details). In a future release, the default value for `weights_only` will be flipped to `True`. This limits the functions that could be executed during unpickling. Arbitrary objects will no longer be allowed to be loaded via this mode unless they are explicitly allowlisted by the user via `torch.serialization.add_safe_globals`. We recommend you start setting `weights_only=True` for any use case where you don't have full control of the loaded file. Please open an issue on GitHub for any issues related to this experimental feature.
rinen0721-llama1023-v1-mkmlizer:   tensors = torch.load(model_shard_filename, map_location=torch.device(self.device), mmap=True)
rinen0721-llama1023-v1-mkmlizer: quantized model in 26.186s
rinen0721-llama1023-v1-mkmlizer: Processed model rinen0721/llama1023 in 93.097s
rinen0721-llama1023-v1-mkmlizer: creating bucket guanaco-mkml-models
rinen0721-llama1023-v1-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
rinen0721-llama1023-v1-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/rinen0721-llama1023-v1
rinen0721-llama1023-v1-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/rinen0721-llama1023-v1/special_tokens_map.json
rinen0721-llama1023-v1-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/rinen0721-llama1023-v1/config.json
rinen0721-llama1023-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/rinen0721-llama1023-v1/tokenizer_config.json
rinen0721-llama1023-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/rinen0721-llama1023-v1/tokenizer.json
rinen0721-llama1023-v1-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/rinen0721-llama1023-v1/flywheel_model.0.safetensors
rinen0721-llama1023-v1-mkmlizer: 
Loading 0:   0%|          | 0/291 [00:00<?, ?it/s]
Loading 0:   2%|▏         | 7/291 [00:00<00:05, 48.56it/s]
Loading 0:   8%|▊         | 22/291 [00:00<00:03, 80.44it/s]
Loading 0:  11%|█         | 31/291 [00:00<00:03, 82.38it/s]
Loading 0:  14%|█▍        | 42/291 [00:00<00:02, 91.70it/s]
Loading 0:  18%|█▊        | 52/291 [00:00<00:02, 81.25it/s]
Loading 0:  21%|██        | 61/291 [00:00<00:02, 81.28it/s]
Loading 0:  24%|██▍       | 70/291 [00:00<00:02, 81.41it/s]
Loading 0:  27%|██▋       | 79/291 [00:00<00:02, 82.87it/s]
Loading 0:  30%|███       | 88/291 [00:02<00:09, 20.82it/s]
Loading 0:  33%|███▎      | 97/291 [00:02<00:07, 27.01it/s]
Loading 0:  36%|███▋      | 106/291 [00:02<00:05, 34.10it/s]
Loading 0:  40%|███▉      | 116/291 [00:02<00:04, 43.31it/s]
Loading 0:  43%|████▎     | 125/291 [00:02<00:03, 50.72it/s]
Loading 0:  48%|████▊     | 139/291 [00:02<00:02, 62.68it/s]
Loading 0:  52%|█████▏    | 151/291 [00:02<00:02, 67.94it/s]
Loading 0:  55%|█████▌    | 161/291 [00:02<00:01, 74.55it/s]
Loading 0:  59%|█████▉    | 171/291 [00:03<00:01, 80.25it/s]
Loading 0:  62%|██████▏   | 181/291 [00:03<00:01, 82.46it/s]
Loading 0:  66%|██████▌   | 191/291 [00:04<00:04, 23.66it/s]
Loading 0:  69%|██████▊   | 200/291 [00:04<00:03, 29.63it/s]
Loading 0:  73%|███████▎  | 211/291 [00:04<00:02, 36.81it/s]
Loading 0:  76%|███████▌  | 220/291 [00:04<00:01, 43.84it/s]
Loading 0:  79%|███████▊  | 229/291 [00:04<00:01, 50.64it/s]
Loading 0:  82%|████████▏ | 238/291 [00:04<00:00, 57.10it/s]
Loading 0:  85%|████████▍ | 247/291 [00:04<00:00, 62.37it/s]
Loading 0:  89%|████████▉ | 259/291 [00:05<00:00, 67.96it/s]
Loading 0:  92%|█████████▏| 268/291 [00:05<00:00, 71.71it/s]
Loading 0:  96%|█████████▌| 278/291 [00:05<00:00, 78.42it/s]
Loading 0:  99%|█████████▊| 287/291 [00:05<00:00, 43.33it/s]
                                                            
Job rinen0721-llama1023-v1-mkmlizer completed after 115.17s with status: succeeded
Stopping job with name rinen0721-llama1023-v1-mkmlizer
Pipeline stage MKMLizer completed in 115.76s
run pipeline stage %s
Running pipeline stage MKMLTemplater
Pipeline stage MKMLTemplater completed in 0.19s
run pipeline stage %s
Running pipeline stage MKMLDeployer
Creating inference service rinen0721-llama1023-v1
Waiting for inference service rinen0721-llama1023-v1 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
Inference service rinen0721-llama1023-v1 ready after 140.50694513320923s
Pipeline stage MKMLDeployer completed in 141.16s
run pipeline stage %s
Running pipeline stage StressChecker
Received healthy response to inference request in 1.8641269207000732s
Received healthy response to inference request in 1.2977864742279053s
Received healthy response to inference request in 1.4947025775909424s
Received healthy response to inference request in 1.2685153484344482s
Received healthy response to inference request in 1.827622890472412s
5 requests
0 failed requests
5th percentile: 1.2743695735931397
10th percentile: 1.280223798751831
20th percentile: 1.2919322490692138
30th percentile: 1.3371696949005127
40th percentile: 1.4159361362457275
50th percentile: 1.4947025775909424
60th percentile: 1.6278707027435302
70th percentile: 1.7610388278961182
80th percentile: 1.8349236965179443
90th percentile: 1.8495253086090089
95th percentile: 1.856826114654541
99th percentile: 1.8626667594909667
mean time: 1.5505508422851562
Pipeline stage StressChecker completed in 9.42s
run pipeline stage %s
Running pipeline stage OfflineFamilyFriendlyTriggerPipeline
run_pipeline:run_in_cloud %s
starting trigger_guanaco_pipeline args=%s
Pipeline stage OfflineFamilyFriendlyTriggerPipeline completed in 1.64s
Shutdown handler de-registered
rinen0721-llama1023_v1 status is now deployed due to DeploymentManager action
rinen0721-llama1023_v1 status is now inactive due to auto deactivation removed underperforming models
rinen0721-llama1023_v1 status is now torndown due to DeploymentManager action