submission_id: cycy233-l3-bdpo-v6-c4_v1
developer_uid: shiroe40
status: inactive
model_repo: cycy233/L3-bdpo-v6-c4
reward_repo: ChaiML/reward_gpt2_medium_preference_24m_e2
generation_params: {'temperature': 0.6, 'top_p': 0.9, 'min_p': 0.0, 'top_k': 40, 'presence_penalty': 0.0, 'frequency_penalty': 0.0, 'stopping_words': ['\n', '<|end_header_id|>', '<|eot_id|>'], 'max_input_tokens': 512, 'best_of': 16, 'max_output_tokens': 64}
formatter: {'memory_template': "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\n{bot_name}'s Persona: {memory}\n\n", 'prompt_template': '{prompt}<|eot_id|>', 'bot_template': '<|start_header_id|>assistant<|end_header_id|>\n\n{bot_name}: {message}<|eot_id|>', 'user_template': '<|start_header_id|>user<|end_header_id|>\n\n{user_name}: {message}<|eot_id|>', 'response_template': '<|start_header_id|>assistant<|end_header_id|>\n\n{bot_name}:', 'truncate_by_message': False}
reward_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}
timestamp: 2024-07-12T23:10:37+00:00
model_name: auto
model_group: cycy233/L3-bdpo-v6-c4
num_battles: 24712
num_wins: 12953
celo_rating: 1217.06
propriety_score: 0.6983134512546277
propriety_total_count: 12155.0
submission_type: basic
model_architecture: LlamaForCausalLM
model_num_parameters: 8030261248.0
best_of: 16
max_input_tokens: 512
max_output_tokens: 64
display_name: auto
ineligible_reason: None
language_model: cycy233/L3-bdpo-v6-c4
model_size: 8B
reward_model: ChaiML/reward_gpt2_medium_preference_24m_e2
us_pacific_date: 2024-07-12
win_ratio: 0.5241583036581418
preference_data_url: None
Resubmit model
Running pipeline stage MKMLizer
Starting job with name cycy233-l3-bdpo-v6-c4-v1-mkmlizer
Waiting for job on cycy233-l3-bdpo-v6-c4-v1-mkmlizer to finish
cycy233-l3-bdpo-v6-c4-v1-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
cycy233-l3-bdpo-v6-c4-v1-mkmlizer: ║ _____ __ __ ║
cycy233-l3-bdpo-v6-c4-v1-mkmlizer: ║ / _/ /_ ___ __/ / ___ ___ / / ║
cycy233-l3-bdpo-v6-c4-v1-mkmlizer: ║ / _/ / // / |/|/ / _ \/ -_) -_) / ║
cycy233-l3-bdpo-v6-c4-v1-mkmlizer: ║ /_//_/\_, /|__,__/_//_/\__/\__/_/ ║
cycy233-l3-bdpo-v6-c4-v1-mkmlizer: ║ /___/ ║
cycy233-l3-bdpo-v6-c4-v1-mkmlizer: ║ ║
cycy233-l3-bdpo-v6-c4-v1-mkmlizer: ║ Version: 0.8.14 ║
cycy233-l3-bdpo-v6-c4-v1-mkmlizer: ║ Copyright 2023 MK ONE TECHNOLOGIES Inc. ║
cycy233-l3-bdpo-v6-c4-v1-mkmlizer: ║ https://mk1.ai ║
cycy233-l3-bdpo-v6-c4-v1-mkmlizer: ║ ║
cycy233-l3-bdpo-v6-c4-v1-mkmlizer: ║ The license key for the current software has been verified as ║
cycy233-l3-bdpo-v6-c4-v1-mkmlizer: ║ belonging to: ║
cycy233-l3-bdpo-v6-c4-v1-mkmlizer: ║ ║
cycy233-l3-bdpo-v6-c4-v1-mkmlizer: ║ Chai Research Corp. ║
cycy233-l3-bdpo-v6-c4-v1-mkmlizer: ║ Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f ║
cycy233-l3-bdpo-v6-c4-v1-mkmlizer: ║ Expiration: 2024-10-15 23:59:59 ║
cycy233-l3-bdpo-v6-c4-v1-mkmlizer: ║ ║
cycy233-l3-bdpo-v6-c4-v1-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
cycy233-l3-bdpo-v6-c4-v1-mkmlizer: Downloaded to shared memory in 38.845s
cycy233-l3-bdpo-v6-c4-v1-mkmlizer: quantizing model to /dev/shm/model_cache
cycy233-l3-bdpo-v6-c4-v1-mkmlizer: Saving flywheel model at /dev/shm/model_cache
cycy233-l3-bdpo-v6-c4-v1-mkmlizer: Loading 0: 0%| | 0/291 [00:00<?, ?it/s] Loading 0: 4%|▍ | 13/291 [00:00<00:02, 109.43it/s] Loading 0: 9%|▉ | 26/291 [00:00<00:02, 118.41it/s] Loading 0: 13%|█▎ | 39/291 [00:00<00:02, 121.82it/s] Loading 0: 18%|█▊ | 52/291 [00:00<00:02, 118.05it/s] Loading 0: 23%|██▎ | 66/291 [00:00<00:01, 123.06it/s] Loading 0: 27%|██▋ | 79/291 [00:00<00:01, 118.54it/s] Loading 0: 31%|███▏ | 91/291 [00:01<00:04, 45.07it/s] Loading 0: 34%|███▍ | 100/291 [00:01<00:03, 51.34it/s] Loading 0: 38%|███▊ | 112/291 [00:01<00:02, 61.80it/s] Loading 0: 42%|████▏ | 122/291 [00:01<00:02, 66.68it/s] Loading 0: 47%|████▋ | 138/291 [00:01<00:01, 83.43it/s] Loading 0: 51%|█████ | 149/291 [00:01<00:01, 86.80it/s] Loading 0: 57%|█████▋ | 165/291 [00:01<00:01, 101.98it/s] Loading 0: 62%|██████▏ | 179/291 [00:02<00:01, 109.51it/s] Loading 0: 66%|██████▌ | 192/291 [00:02<00:01, 54.80it/s] Loading 0: 70%|██████▉ | 203/291 [00:02<00:01, 62.16it/s] Loading 0: 75%|███████▌ | 219/291 [00:02<00:00, 76.36it/s] Loading 0: 79%|███████▉ | 230/291 [00:02<00:00, 80.88it/s] Loading 0: 83%|████████▎ | 241/291 [00:03<00:00, 86.07it/s] Loading 0: 88%|████████▊ | 255/291 [00:03<00:00, 96.78it/s] Loading 0: 92%|█████████▏| 267/291 [00:03<00:00, 99.77it/s] Loading 0: 97%|█████████▋| 281/291 [00:03<00:00, 107.43it/s] Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
cycy233-l3-bdpo-v6-c4-v1-mkmlizer: quantized model in 29.604s
cycy233-l3-bdpo-v6-c4-v1-mkmlizer: Processed model cycy233/L3-bdpo-v6-c4 in 68.449s
cycy233-l3-bdpo-v6-c4-v1-mkmlizer: creating bucket guanaco-mkml-models
cycy233-l3-bdpo-v6-c4-v1-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
cycy233-l3-bdpo-v6-c4-v1-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/cycy233-l3-bdpo-v6-c4-v1
cycy233-l3-bdpo-v6-c4-v1-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/cycy233-l3-bdpo-v6-c4-v1/config.json
cycy233-l3-bdpo-v6-c4-v1-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/cycy233-l3-bdpo-v6-c4-v1/special_tokens_map.json
cycy233-l3-bdpo-v6-c4-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/cycy233-l3-bdpo-v6-c4-v1/tokenizer.json
cycy233-l3-bdpo-v6-c4-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/cycy233-l3-bdpo-v6-c4-v1/tokenizer_config.json
cycy233-l3-bdpo-v6-c4-v1-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/cycy233-l3-bdpo-v6-c4-v1/flywheel_model.0.safetensors
cycy233-l3-bdpo-v6-c4-v1-mkmlizer: loading reward model from ChaiML/reward_gpt2_medium_preference_24m_e2
cycy233-l3-bdpo-v6-c4-v1-mkmlizer: /opt/conda/lib/python3.10/site-packages/transformers/models/auto/configuration_auto.py:919: FutureWarning: The `use_auth_token` argument is deprecated and will be removed in v5 of Transformers. Please use `token` instead.
cycy233-l3-bdpo-v6-c4-v1-mkmlizer: warnings.warn(
cycy233-l3-bdpo-v6-c4-v1-mkmlizer: /opt/conda/lib/python3.10/site-packages/huggingface_hub/file_download.py:1132: FutureWarning: `resume_download` is deprecated and will be removed in version 1.0.0. Downloads always resume when possible. If you want to force a new download, use `force_download=True`.
cycy233-l3-bdpo-v6-c4-v1-mkmlizer: warnings.warn(
cycy233-l3-bdpo-v6-c4-v1-mkmlizer: /opt/conda/lib/python3.10/site-packages/transformers/models/auto/tokenization_auto.py:769: FutureWarning: The `use_auth_token` argument is deprecated and will be removed in v5 of Transformers. Please use `token` instead.
cycy233-l3-bdpo-v6-c4-v1-mkmlizer: warnings.warn(
cycy233-l3-bdpo-v6-c4-v1-mkmlizer: /opt/conda/lib/python3.10/site-packages/transformers/models/auto/auto_factory.py:468: FutureWarning: The `use_auth_token` argument is deprecated and will be removed in v5 of Transformers. Please use `token` instead.
cycy233-l3-bdpo-v6-c4-v1-mkmlizer: warnings.warn(
cycy233-l3-bdpo-v6-c4-v1-mkmlizer: /opt/conda/lib/python3.10/site-packages/torch/_utils.py:831: UserWarning: TypedStorage is deprecated. It will be removed in the future and UntypedStorage will be the only storage class. This should only matter to you if you are using storages directly. To access UntypedStorage directly, use tensor.untyped_storage() instead of tensor.storage()
cycy233-l3-bdpo-v6-c4-v1-mkmlizer: return self.fget.__get__(instance, owner)()
cycy233-l3-bdpo-v6-c4-v1-mkmlizer: Saving model to /tmp/reward_cache/reward.tensors
cycy233-l3-bdpo-v6-c4-v1-mkmlizer: Saving duration: 0.513s
cycy233-l3-bdpo-v6-c4-v1-mkmlizer: Processed model ChaiML/reward_gpt2_medium_preference_24m_e2 in 8.152s
cycy233-l3-bdpo-v6-c4-v1-mkmlizer: creating bucket guanaco-reward-models
cycy233-l3-bdpo-v6-c4-v1-mkmlizer: Bucket 's3://guanaco-reward-models/' created
cycy233-l3-bdpo-v6-c4-v1-mkmlizer: uploading /tmp/reward_cache to s3://guanaco-reward-models/cycy233-l3-bdpo-v6-c4-v1_reward
cycy233-l3-bdpo-v6-c4-v1-mkmlizer: cp /tmp/reward_cache/config.json s3://guanaco-reward-models/cycy233-l3-bdpo-v6-c4-v1_reward/config.json
cycy233-l3-bdpo-v6-c4-v1-mkmlizer: cp /tmp/reward_cache/special_tokens_map.json s3://guanaco-reward-models/cycy233-l3-bdpo-v6-c4-v1_reward/special_tokens_map.json
cycy233-l3-bdpo-v6-c4-v1-mkmlizer: cp /tmp/reward_cache/tokenizer_config.json s3://guanaco-reward-models/cycy233-l3-bdpo-v6-c4-v1_reward/tokenizer_config.json
cycy233-l3-bdpo-v6-c4-v1-mkmlizer: cp /tmp/reward_cache/merges.txt s3://guanaco-reward-models/cycy233-l3-bdpo-v6-c4-v1_reward/merges.txt
cycy233-l3-bdpo-v6-c4-v1-mkmlizer: cp /tmp/reward_cache/vocab.json s3://guanaco-reward-models/cycy233-l3-bdpo-v6-c4-v1_reward/vocab.json
cycy233-l3-bdpo-v6-c4-v1-mkmlizer: cp /tmp/reward_cache/tokenizer.json s3://guanaco-reward-models/cycy233-l3-bdpo-v6-c4-v1_reward/tokenizer.json
cycy233-l3-bdpo-v6-c4-v1-mkmlizer: cp /tmp/reward_cache/reward.tensors s3://guanaco-reward-models/cycy233-l3-bdpo-v6-c4-v1_reward/reward.tensors
Job cycy233-l3-bdpo-v6-c4-v1-mkmlizer completed after 104.13s with status: succeeded
Stopping job with name cycy233-l3-bdpo-v6-c4-v1-mkmlizer
Pipeline stage MKMLizer completed in 104.86s
Running pipeline stage MKMLKubeTemplater
Pipeline stage MKMLKubeTemplater completed in 0.10s
Running pipeline stage ISVCDeployer
Creating inference service cycy233-l3-bdpo-v6-c4-v1
Waiting for inference service cycy233-l3-bdpo-v6-c4-v1 to be ready
Inference service cycy233-l3-bdpo-v6-c4-v1 ready after 50.23944711685181s
Pipeline stage ISVCDeployer completed in 56.91s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.083866596221924s
Received healthy response to inference request in 1.6146535873413086s
Received healthy response to inference request in 4.439690113067627s
Received healthy response to inference request in 1.2668840885162354s
Received healthy response to inference request in 1.3423371315002441s
5 requests
0 failed requests
5th percentile: 1.281974697113037
10th percentile: 1.297065305709839
20th percentile: 1.3272465229034425
30th percentile: 1.396800422668457
40th percentile: 1.5057270050048828
50th percentile: 1.6146535873413086
60th percentile: 1.8023387908935546
70th percentile: 1.9900239944458007
80th percentile: 2.555031299591065
90th percentile: 3.4973607063293457
95th percentile: 3.968525409698486
99th percentile: 4.3454571723937985
mean time: 2.1494863033294678
Pipeline stage StressChecker completed in 11.67s
cycy233-l3-bdpo-v6-c4_v1 status is now deployed due to DeploymentManager action
cycy233-l3-bdpo-v6-c4_v1 status is now inactive due to auto deactivation removed underperforming models

Usage Metrics

Latency Metrics