submission_id: meseca-20062024-c1-dpo_v1
developer_uid: nguyenzzz
status: inactive
model_repo: meseca/20062024-c1-dpo
reward_repo: ChaiML/reward_gpt2_medium_preference_24m_e2
generation_params: {'temperature': 0.95, 'top_p': 1.0, 'min_p': 0.1, 'top_k': 60, 'presence_penalty': 0.0, 'frequency_penalty': 0.0, 'stopping_words': ['\n', '<|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: 2024-06-27T15:16:50+00:00
model_name: meseca-20062024-c1-dpo_v1
model_group: meseca/20062024-c1-dpo
num_battles: 15605
num_wins: 7304
celo_rating: 1164.61
propriety_score: 0.7539639881752217
propriety_total_count: 7442.0
submission_type: basic
model_architecture: LlamaForCausalLM
model_num_parameters: 8030261248.0
best_of: 16
max_input_tokens: 512
max_output_tokens: 64
display_name: meseca-20062024-c1-dpo_v1
ineligible_reason: None
language_model: meseca/20062024-c1-dpo
model_size: 8B
reward_model: ChaiML/reward_gpt2_medium_preference_24m_e2
us_pacific_date: 2024-06-27
win_ratio: 0.4680551105414931
Resubmit model
Running pipeline stage MKMLizer
Starting job with name meseca-20062024-c1-dpo-v1-mkmlizer
Waiting for job on meseca-20062024-c1-dpo-v1-mkmlizer to finish
meseca-20062024-c1-dpo-v1-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
meseca-20062024-c1-dpo-v1-mkmlizer: ║ _____ __ __ ║
meseca-20062024-c1-dpo-v1-mkmlizer: ║ / _/ /_ ___ __/ / ___ ___ / / ║
meseca-20062024-c1-dpo-v1-mkmlizer: ║ / _/ / // / |/|/ / _ \/ -_) -_) / ║
meseca-20062024-c1-dpo-v1-mkmlizer: ║ /_//_/\_, /|__,__/_//_/\__/\__/_/ ║
meseca-20062024-c1-dpo-v1-mkmlizer: ║ /___/ ║
meseca-20062024-c1-dpo-v1-mkmlizer: ║ ║
meseca-20062024-c1-dpo-v1-mkmlizer: ║ Version: 0.8.14 ║
meseca-20062024-c1-dpo-v1-mkmlizer: ║ Copyright 2023 MK ONE TECHNOLOGIES Inc. ║
meseca-20062024-c1-dpo-v1-mkmlizer: ║ https://mk1.ai ║
meseca-20062024-c1-dpo-v1-mkmlizer: ║ ║
meseca-20062024-c1-dpo-v1-mkmlizer: ║ The license key for the current software has been verified as ║
meseca-20062024-c1-dpo-v1-mkmlizer: ║ belonging to: ║
meseca-20062024-c1-dpo-v1-mkmlizer: ║ ║
meseca-20062024-c1-dpo-v1-mkmlizer: ║ Chai Research Corp. ║
meseca-20062024-c1-dpo-v1-mkmlizer: ║ Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f ║
meseca-20062024-c1-dpo-v1-mkmlizer: ║ Expiration: 2024-07-15 23:59:59 ║
meseca-20062024-c1-dpo-v1-mkmlizer: ║ ║
meseca-20062024-c1-dpo-v1-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
meseca-20062024-c1-dpo-v1-mkmlizer: /opt/conda/lib/python3.10/site-packages/huggingface_hub/utils/_deprecation.py:131: FutureWarning: 'list_files_info' (from 'huggingface_hub.hf_api') is deprecated and will be removed from version '0.23'. Use `list_repo_tree` and `get_paths_info` instead.
meseca-20062024-c1-dpo-v1-mkmlizer: warnings.warn(warning_message, FutureWarning)
Connection pool is full, discarding connection: %s
Connection pool is full, discarding connection: %s
Connection pool is full, discarding connection: %s
meseca-20062024-c1-dpo-v1-mkmlizer: Downloaded to shared memory in 41.751s
meseca-20062024-c1-dpo-v1-mkmlizer: quantizing model to /dev/shm/model_cache
meseca-20062024-c1-dpo-v1-mkmlizer: Saving flywheel model at /dev/shm/model_cache
meseca-20062024-c1-dpo-v1-mkmlizer: Loading 0: 0%| | 0/291 [00:00<?, ?it/s] Loading 0: 4%|▍ | 13/291 [00:00<00:02, 118.12it/s] Loading 0: 9%|▊ | 25/291 [00:00<00:02, 110.13it/s] Loading 0: 14%|█▎ | 40/291 [00:00<00:02, 118.30it/s] Loading 0: 18%|█▊ | 53/291 [00:00<00:01, 120.30it/s] Loading 0: 23%|██▎ | 67/291 [00:00<00:01, 121.29it/s] Loading 0: 27%|██▋ | 80/291 [00:00<00:01, 122.40it/s] Loading 0: 32%|███▏ | 93/291 [00:01<00:03, 63.56it/s] Loading 0: 35%|███▌ | 103/291 [00:01<00:02, 69.25it/s] Loading 0: 39%|███▉ | 113/291 [00:01<00:02, 74.55it/s] Loading 0: 44%|████▍ | 129/291 [00:01<00:01, 92.00it/s] Loading 0: 48%|████▊ | 140/291 [00:01<00:01, 92.98it/s] Loading 0: 54%|█████▎ | 156/291 [00:01<00:01, 106.33it/s] Loading 0: 58%|█████▊ | 168/291 [00:01<00:01, 106.92it/s] Loading 0: 63%|██████▎ | 184/291 [00:01<00:00, 120.02it/s] Loading 0: 68%|██████▊ | 197/291 [00:02<00:01, 72.71it/s] Loading 0: 72%|███████▏ | 210/291 [00:02<00:00, 82.53it/s] Loading 0: 76%|███████▌ | 221/291 [00:02<00:00, 87.62it/s] Loading 0: 82%|████████▏ | 238/291 [00:02<00:00, 102.88it/s] Loading 0: 86%|████████▌ | 250/291 [00:02<00:00, 104.91it/s] Loading 0: 91%|█████████ | 265/291 [00:02<00:00, 112.97it/s] Loading 0: 97%|█████████▋| 281/291 [00:02<00:00, 121.26it/s] Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
meseca-20062024-c1-dpo-v1-mkmlizer: quantized model in 24.254s
meseca-20062024-c1-dpo-v1-mkmlizer: Processed model meseca/20062024-c1-dpo in 68.553s
meseca-20062024-c1-dpo-v1-mkmlizer: creating bucket guanaco-mkml-models
meseca-20062024-c1-dpo-v1-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
meseca-20062024-c1-dpo-v1-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/meseca-20062024-c1-dpo-v1
meseca-20062024-c1-dpo-v1-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/meseca-20062024-c1-dpo-v1/config.json
meseca-20062024-c1-dpo-v1-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/meseca-20062024-c1-dpo-v1/special_tokens_map.json
meseca-20062024-c1-dpo-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/meseca-20062024-c1-dpo-v1/tokenizer_config.json
meseca-20062024-c1-dpo-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/meseca-20062024-c1-dpo-v1/tokenizer.json
meseca-20062024-c1-dpo-v1-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/meseca-20062024-c1-dpo-v1/flywheel_model.0.safetensors
meseca-20062024-c1-dpo-v1-mkmlizer: loading reward model from ChaiML/reward_gpt2_medium_preference_24m_e2
meseca-20062024-c1-dpo-v1-mkmlizer: /opt/conda/lib/python3.10/site-packages/transformers/models/auto/configuration_auto.py:913: FutureWarning: The `use_auth_token` argument is deprecated and will be removed in v5 of Transformers. Please use `token` instead.
meseca-20062024-c1-dpo-v1-mkmlizer: warnings.warn(
meseca-20062024-c1-dpo-v1-mkmlizer: /opt/conda/lib/python3.10/site-packages/transformers/models/auto/tokenization_auto.py:757: FutureWarning: The `use_auth_token` argument is deprecated and will be removed in v5 of Transformers. Please use `token` instead.
meseca-20062024-c1-dpo-v1-mkmlizer: warnings.warn(
meseca-20062024-c1-dpo-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.
meseca-20062024-c1-dpo-v1-mkmlizer: warnings.warn(
meseca-20062024-c1-dpo-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()
meseca-20062024-c1-dpo-v1-mkmlizer: return self.fget.__get__(instance, owner)()
meseca-20062024-c1-dpo-v1-mkmlizer: Saving model to /tmp/reward_cache/reward.tensors
meseca-20062024-c1-dpo-v1-mkmlizer: Saving duration: 0.427s
meseca-20062024-c1-dpo-v1-mkmlizer: Processed model ChaiML/reward_gpt2_medium_preference_24m_e2 in 4.801s
meseca-20062024-c1-dpo-v1-mkmlizer: cp /tmp/reward_cache/reward.tensors s3://guanaco-reward-models/meseca-20062024-c1-dpo-v1_reward/reward.tensors
Job meseca-20062024-c1-dpo-v1-mkmlizer completed after 95.18s with status: succeeded
Stopping job with name meseca-20062024-c1-dpo-v1-mkmlizer
Pipeline stage MKMLizer completed in 96.42s
Running pipeline stage MKMLKubeTemplater
Pipeline stage MKMLKubeTemplater completed in 0.11s
Running pipeline stage ISVCDeployer
Creating inference service meseca-20062024-c1-dpo-v1
Waiting for inference service meseca-20062024-c1-dpo-v1 to be ready
Connection pool is full, discarding connection: %s
Connection pool is full, discarding connection: %s
Connection pool is full, discarding connection: %s
Inference service meseca-20062024-c1-dpo-v1 ready after 40.44720482826233s
Pipeline stage ISVCDeployer completed in 47.30s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.160301685333252s
Received healthy response to inference request in 1.380366563796997s
Received healthy response to inference request in 1.354210376739502s
Received healthy response to inference request in 1.304527997970581s
Received healthy response to inference request in 1.3672959804534912s
5 requests
0 failed requests
5th percentile: 1.3144644737243651
10th percentile: 1.3244009494781495
20th percentile: 1.3442739009857179
30th percentile: 1.3568274974822998
40th percentile: 1.3620617389678955
50th percentile: 1.3672959804534912
60th percentile: 1.3725242137908935
70th percentile: 1.3777524471282958
80th percentile: 1.5363535881042483
90th percentile: 1.84832763671875
95th percentile: 2.004314661026001
99th percentile: 2.1291042804718017
mean time: 1.5133405208587647
Pipeline stage StressChecker completed in 8.29s
meseca-20062024-c1-dpo_v1 status is now deployed due to DeploymentManager action
meseca-20062024-c1-dpo_v1 status is now inactive due to auto deactivation removed underperforming models

Usage Metrics

Latency Metrics