meseca-05062024-v2

submission_id: meseca-05062024-v2_v1

developer_uid: nguyenzzz

best_of: 16

celo_rating: 1186.25

display_name: meseca-05062024-v2-2e_v1

family_friendly_score: 0.0

formatter: {'memory_template': "<|start_header_id|>system<|end_header_id|>\n\nYou're {bot_name} in this fictional never-ending roleplay with {user_name}.\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}

generation_params: {'temperature': 1.13, 'top_p': 1.0, 'min_p': 0.08, 'top_k': 100, 'presence_penalty': 0.0, 'frequency_penalty': 0.0, 'stopping_words': ['\n'], 'max_input_tokens': 512, 'best_of': 16, 'max_output_tokens': 64}

is_internal_developer: False

language_model: meseca/05062024-v2

max_input_tokens: 512

max_output_tokens: 64

model_architecture: LlamaForCausalLM

model_eval_status: success

model_group: meseca/05062024-v2

model_name: meseca-05062024-v2-2e_v1

model_num_parameters: 8030261248.0

model_repo: meseca/05062024-v2

model_size: 8B

num_battles: 10921

num_wins: 5492

ranking_group: single

reward_formatter: {'bot_template': '{bot_name}: {message}\n', 'memory_template': "{bot_name}'s Persona: {memory}\n####\n", 'prompt_template': '{prompt}\n<START>\n', 'response_template': '{bot_name}:', 'truncate_by_message': False, 'user_template': '{user_name}: {message}\n'}

reward_repo: ChaiML/reward_gpt2_medium_preference_24m_e2

status: torndown

submission_type: basic

timestamp: 2024-06-07T01:24:25+00:00

us_pacific_date: 2024-06-06

win_ratio: 0.5028843512498855

Resubmit model

Running pipeline stage MKMLizer
Starting job with name meseca-05062024-v2-v1-mkmlizer
Waiting for job on meseca-05062024-v2-v1-mkmlizer to finish
meseca-05062024-v2-v1-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
meseca-05062024-v2-v1-mkmlizer: ║     _____            __           __                                ║
meseca-05062024-v2-v1-mkmlizer: ║    / _/ /_ ___    __/ /  ___ ___ / /                                ║
meseca-05062024-v2-v1-mkmlizer: ║   / _/ / // / |/|/ / _ \/ -_) -_) /                                 ║
meseca-05062024-v2-v1-mkmlizer: ║  /_//_/\_, /|__,__/_//_/\__/\__/_/                                  ║
meseca-05062024-v2-v1-mkmlizer: ║       /___/                                                         ║
meseca-05062024-v2-v1-mkmlizer: ║                                                                     ║
meseca-05062024-v2-v1-mkmlizer: ║  Version: 0.8.14                                                    ║
meseca-05062024-v2-v1-mkmlizer: ║  Copyright 2023 MK ONE TECHNOLOGIES Inc.                            ║
meseca-05062024-v2-v1-mkmlizer: ║  https://mk1.ai                                                     ║
meseca-05062024-v2-v1-mkmlizer: ║                                                                     ║
meseca-05062024-v2-v1-mkmlizer: ║  The license key for the current software has been verified as      ║
meseca-05062024-v2-v1-mkmlizer: ║  belonging to:                                                      ║
meseca-05062024-v2-v1-mkmlizer: ║                                                                     ║
meseca-05062024-v2-v1-mkmlizer: ║  Chai Research Corp.                                                ║
meseca-05062024-v2-v1-mkmlizer: ║  Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f                   ║
meseca-05062024-v2-v1-mkmlizer: ║  Expiration: 2024-07-15 23:59:59                                    ║
meseca-05062024-v2-v1-mkmlizer: ║                                                                     ║
meseca-05062024-v2-v1-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
meseca-05062024-v2-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-05062024-v2-v1-mkmlizer:   warnings.warn(warning_message, FutureWarning)
meseca-05062024-v2-v1-mkmlizer: Downloaded to shared memory in 51.588s
meseca-05062024-v2-v1-mkmlizer: quantizing model to /dev/shm/model_cache
meseca-05062024-v2-v1-mkmlizer: Saving flywheel model at /dev/shm/model_cache
meseca-05062024-v2-v1-mkmlizer: 
Loading 0:   0%|          | 0/291 [00:00<?, ?it/s]
Loading 0:   4%|▍         | 13/291 [00:00<00:02, 119.65it/s]
Loading 0:  10%|▉         | 29/291 [00:00<00:01, 142.55it/s]
Loading 0:  15%|█▌        | 44/291 [00:00<00:01, 128.15it/s]
Loading 0:  20%|█▉        | 58/291 [00:00<00:01, 126.99it/s]
Loading 0:  25%|██▍       | 72/291 [00:00<00:01, 129.81it/s]
Loading 0:  30%|██▉       | 86/291 [00:00<00:02, 68.81it/s] 
Loading 0:  35%|███▌      | 102/291 [00:01<00:02, 85.23it/s]
Loading 0:  39%|███▉      | 114/291 [00:01<00:01, 90.06it/s]
Loading 0:  44%|████▍     | 129/291 [00:01<00:01, 102.64it/s]
Loading 0:  49%|████▉     | 142/291 [00:01<00:01, 105.71it/s]
Loading 0:  54%|█████▎    | 156/291 [00:01<00:01, 112.29it/s]
Loading 0:  58%|█████▊    | 169/291 [00:01<00:01, 113.12it/s]
Loading 0:  63%|██████▎   | 182/291 [00:01<00:00, 115.97it/s]
Loading 0:  67%|██████▋   | 195/291 [00:02<00:01, 72.27it/s] 
Loading 0:  72%|███████▏  | 210/291 [00:02<00:00, 86.60it/s]
Loading 0:  76%|███████▋  | 222/291 [00:02<00:00, 92.97it/s]
Loading 0:  82%|████████▏ | 238/291 [00:02<00:00, 105.28it/s]
Loading 0:  88%|████████▊ | 255/291 [00:02<00:00, 116.85it/s]
Loading 0:  92%|█████████▏| 268/291 [00:02<00:00, 118.72it/s]
Loading 0:  97%|█████████▋| 282/291 [00:02<00:00, 119.01it/s]
                                                             
Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
meseca-05062024-v2-v1-mkmlizer: quantized model in 23.477s
meseca-05062024-v2-v1-mkmlizer: Processed model meseca/05062024-v2 in 77.533s
meseca-05062024-v2-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/meseca-05062024-v2-v1/tokenizer.json
meseca-05062024-v2-v1-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/meseca-05062024-v2-v1/flywheel_model.0.safetensors
meseca-05062024-v2-v1-mkmlizer: loading reward model from ChaiML/reward_gpt2_medium_preference_24m_e2
meseca-05062024-v2-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-05062024-v2-v1-mkmlizer:   warnings.warn(
meseca-05062024-v2-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-05062024-v2-v1-mkmlizer:   warnings.warn(
meseca-05062024-v2-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-05062024-v2-v1-mkmlizer:   warnings.warn(
meseca-05062024-v2-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-05062024-v2-v1-mkmlizer:   return self.fget.__get__(instance, owner)()
meseca-05062024-v2-v1-mkmlizer: Saving model to /tmp/reward_cache/reward.tensors
meseca-05062024-v2-v1-mkmlizer: Saving duration: 0.383s
meseca-05062024-v2-v1-mkmlizer: Processed model ChaiML/reward_gpt2_medium_preference_24m_e2 in 12.395s
meseca-05062024-v2-v1-mkmlizer: creating bucket guanaco-reward-models
meseca-05062024-v2-v1-mkmlizer: Bucket 's3://guanaco-reward-models/' created
meseca-05062024-v2-v1-mkmlizer: uploading /tmp/reward_cache to s3://guanaco-reward-models/meseca-05062024-v2-v1_reward
meseca-05062024-v2-v1-mkmlizer: cp /tmp/reward_cache/config.json s3://guanaco-reward-models/meseca-05062024-v2-v1_reward/config.json
meseca-05062024-v2-v1-mkmlizer: cp /tmp/reward_cache/special_tokens_map.json s3://guanaco-reward-models/meseca-05062024-v2-v1_reward/special_tokens_map.json
meseca-05062024-v2-v1-mkmlizer: cp /tmp/reward_cache/tokenizer_config.json s3://guanaco-reward-models/meseca-05062024-v2-v1_reward/tokenizer_config.json
meseca-05062024-v2-v1-mkmlizer: cp /tmp/reward_cache/vocab.json s3://guanaco-reward-models/meseca-05062024-v2-v1_reward/vocab.json
meseca-05062024-v2-v1-mkmlizer: cp /tmp/reward_cache/merges.txt s3://guanaco-reward-models/meseca-05062024-v2-v1_reward/merges.txt
meseca-05062024-v2-v1-mkmlizer: cp /tmp/reward_cache/tokenizer.json s3://guanaco-reward-models/meseca-05062024-v2-v1_reward/tokenizer.json
meseca-05062024-v2-v1-mkmlizer: cp /tmp/reward_cache/reward.tensors s3://guanaco-reward-models/meseca-05062024-v2-v1_reward/reward.tensors
Job meseca-05062024-v2-v1-mkmlizer completed after 123.96s with status: succeeded
Stopping job with name meseca-05062024-v2-v1-mkmlizer
Pipeline stage MKMLizer completed in 126.35s
Running pipeline stage MKMLKubeTemplater
Pipeline stage MKMLKubeTemplater completed in 0.10s
Running pipeline stage ISVCDeployer
Creating inference service meseca-05062024-v2-v1
Waiting for inference service meseca-05062024-v2-v1 to be ready
Inference service meseca-05062024-v2-v1 ready after 110.56210970878601s
Pipeline stage ISVCDeployer completed in 117.65s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.4892659187316895s
Received healthy response to inference request in 1.3421669006347656s
Received healthy response to inference request in 1.3036208152770996s
Received healthy response to inference request in 1.2554023265838623s
Received healthy response to inference request in 10.760197639465332s
5 requests
0 failed requests
5th percentile: 1.2650460243225097
10th percentile: 1.2746897220611573
20th percentile: 1.2939771175384522
30th percentile: 1.3113300323486328
40th percentile: 1.3267484664916993
50th percentile: 1.3421669006347656
60th percentile: 1.8010065078735351
70th percentile: 2.2598461151123046
80th percentile: 4.143452262878419
90th percentile: 7.451824951171876
95th percentile: 9.106011295318602
99th percentile: 10.429360370635987
mean time: 3.43013072013855
Pipeline stage StressChecker completed in 17.73s
Running pipeline stage DaemonicModelEvalScorer
Pipeline stage DaemonicModelEvalScorer completed in 0.03s
Running pipeline stage DaemonicSafetyScorer
Running M-Eval for topic stay_in_character
Pipeline stage DaemonicSafetyScorer completed in 0.03s
M-Eval Dataset for topic stay_in_character is loaded
meseca-05062024-v2_v1 status is now deployed due to DeploymentManager action
meseca-05062024-v2_v1 status is now inactive due to auto deactivation removed underperforming models
admin requested tearing down of meseca-05062024-v2_v1
Running pipeline stage ISVCDeleter
Checking if service meseca-05062024-v2-v1 is running
Skipping teardown as no inference service was found
Pipeline stage ISVCDeleter completed in 2.41s
Running pipeline stage MKMLModelDeleter
Cleaning model data from S3
Cleaning model data from model cache
Deleting key meseca-05062024-v2-v1/config.json from bucket guanaco-mkml-models
Deleting key meseca-05062024-v2-v1/flywheel_model.0.safetensors from bucket guanaco-mkml-models
Deleting key meseca-05062024-v2-v1/special_tokens_map.json from bucket guanaco-mkml-models
Deleting key meseca-05062024-v2-v1/tokenizer.json from bucket guanaco-mkml-models
Deleting key meseca-05062024-v2-v1/tokenizer_config.json from bucket guanaco-mkml-models
Cleaning model data from model cache
Deleting key meseca-05062024-v2-v1_reward/config.json from bucket guanaco-reward-models
Deleting key meseca-05062024-v2-v1_reward/merges.txt from bucket guanaco-reward-models
Deleting key meseca-05062024-v2-v1_reward/reward.tensors from bucket guanaco-reward-models
Deleting key meseca-05062024-v2-v1_reward/special_tokens_map.json from bucket guanaco-reward-models
Deleting key meseca-05062024-v2-v1_reward/tokenizer.json from bucket guanaco-reward-models
Deleting key meseca-05062024-v2-v1_reward/tokenizer_config.json from bucket guanaco-reward-models
Deleting key meseca-05062024-v2-v1_reward/vocab.json from bucket guanaco-reward-models
Pipeline stage MKMLModelDeleter completed in 3.47s
meseca-05062024-v2_v1 status is now torndown due to DeploymentManager action