cycy233-misllm-v5-5

developer_uid: shiroe40

submission_id: cycy233-misllm-v5-5_v1

model_name: base

model_group: cycy233/misllm_v5.5

status: torndown

timestamp: 2024-05-24T01:56:57+00:00

num_battles: 13420

num_wins: 6778

celo_rating: 1168.16

family_friendly_score: 0.0

submission_type: basic

model_repo: cycy233/misllm_v5.5

model_architecture: MistralForCausalLM

reward_repo: ChaiML/reward_gpt2_medium_preference_24m_e2

model_num_parameters: 7241732096.0

best_of: 16

max_input_tokens: 512

max_output_tokens: 64

display_name: base

is_internal_developer: False

language_model: cycy233/misllm_v5.5

model_size: 7B

ranking_group: single

us_pacific_date: 2024-05-23

win_ratio: 0.5050670640834575

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': 512, 'best_of': 16, 'max_output_tokens': 64}

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}

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'}

model_eval_status: success

Resubmit model

Running pipeline stage MKMLizer
Starting job with name cycy233-misllm-v5-5-v1-mkmlizer
Waiting for job on cycy233-misllm-v5-5-v1-mkmlizer to finish
cycy233-misllm-v5-5-v1-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
cycy233-misllm-v5-5-v1-mkmlizer: ║     _____            __           __                                ║
cycy233-misllm-v5-5-v1-mkmlizer: ║    / _/ /_ ___    __/ /  ___ ___ / /                                ║
cycy233-misllm-v5-5-v1-mkmlizer: ║   / _/ / // / |/|/ / _ \/ -_) -_) /                                 ║
cycy233-misllm-v5-5-v1-mkmlizer: ║  /_//_/\_, /|__,__/_//_/\__/\__/_/                                  ║
cycy233-misllm-v5-5-v1-mkmlizer: ║       /___/                                                         ║
cycy233-misllm-v5-5-v1-mkmlizer: ║                                                                     ║
cycy233-misllm-v5-5-v1-mkmlizer: ║  Version: 0.8.14                                                    ║
cycy233-misllm-v5-5-v1-mkmlizer: ║  Copyright 2023 MK ONE TECHNOLOGIES Inc.                            ║
cycy233-misllm-v5-5-v1-mkmlizer: ║  https://mk1.ai                                                     ║
cycy233-misllm-v5-5-v1-mkmlizer: ║                                                                     ║
cycy233-misllm-v5-5-v1-mkmlizer: ║  The license key for the current software has been verified as      ║
cycy233-misllm-v5-5-v1-mkmlizer: ║  belonging to:                                                      ║
cycy233-misllm-v5-5-v1-mkmlizer: ║                                                                     ║
cycy233-misllm-v5-5-v1-mkmlizer: ║  Chai Research Corp.                                                ║
cycy233-misllm-v5-5-v1-mkmlizer: ║  Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f                   ║
cycy233-misllm-v5-5-v1-mkmlizer: ║  Expiration: 2024-07-15 23:59:59                                    ║
cycy233-misllm-v5-5-v1-mkmlizer: ║                                                                     ║
cycy233-misllm-v5-5-v1-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
cycy233-misllm-v5-5-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.
cycy233-misllm-v5-5-v1-mkmlizer:   warnings.warn(warning_message, FutureWarning)
cycy233-misllm-v5-5-v1-mkmlizer: Downloaded to shared memory in 39.716s
cycy233-misllm-v5-5-v1-mkmlizer: quantizing model to /dev/shm/model_cache
cycy233-misllm-v5-5-v1-mkmlizer: Saving flywheel model at /dev/shm/model_cache
cycy233-misllm-v5-5-v1-mkmlizer: 
Loading 0:   0%|          | 0/291 [00:00<?, ?it/s]
Loading 0:  34%|███▎      | 98/291 [00:01<00:02, 89.82it/s]
Loading 0:  70%|███████   | 204/291 [00:02<00:00, 101.79it/s]
Loading 0: 100%|██████████| 291/291 [00:03<00:00, 86.07it/s] 
                                                            
You set `add_prefix_space`. The tokenizer needs to be converted from the slow tokenizers
cycy233-misllm-v5-5-v1-mkmlizer: quantized model in 11.579s
cycy233-misllm-v5-5-v1-mkmlizer: Processed model cycy233/misllm_v5.5 in 52.276s
cycy233-misllm-v5-5-v1-mkmlizer: creating bucket guanaco-mkml-models
cycy233-misllm-v5-5-v1-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
cycy233-misllm-v5-5-v1-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/cycy233-misllm-v5-5-v1
cycy233-misllm-v5-5-v1-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/cycy233-misllm-v5-5-v1/config.json
cycy233-misllm-v5-5-v1-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/cycy233-misllm-v5-5-v1/special_tokens_map.json
cycy233-misllm-v5-5-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/cycy233-misllm-v5-5-v1/tokenizer_config.json
cycy233-misllm-v5-5-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer.model s3://guanaco-mkml-models/cycy233-misllm-v5-5-v1/tokenizer.model
cycy233-misllm-v5-5-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/cycy233-misllm-v5-5-v1/tokenizer.json
cycy233-misllm-v5-5-v1-mkmlizer: loading reward model from ChaiML/reward_gpt2_medium_preference_24m_e2
cycy233-misllm-v5-5-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.
cycy233-misllm-v5-5-v1-mkmlizer:   warnings.warn(
cycy233-misllm-v5-5-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.
cycy233-misllm-v5-5-v1-mkmlizer:   warnings.warn(
cycy233-misllm-v5-5-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-misllm-v5-5-v1-mkmlizer:   warnings.warn(
cycy233-misllm-v5-5-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-misllm-v5-5-v1-mkmlizer:   return self.fget.__get__(instance, owner)()
cycy233-misllm-v5-5-v1-mkmlizer: Saving model to /tmp/reward_cache/reward.tensors
cycy233-misllm-v5-5-v1-mkmlizer: Saving duration: 0.307s
cycy233-misllm-v5-5-v1-mkmlizer: Processed model ChaiML/reward_gpt2_medium_preference_24m_e2 in 6.610s
cycy233-misllm-v5-5-v1-mkmlizer: creating bucket guanaco-reward-models
cycy233-misllm-v5-5-v1-mkmlizer: Bucket 's3://guanaco-reward-models/' created
cycy233-misllm-v5-5-v1-mkmlizer: uploading /tmp/reward_cache to s3://guanaco-reward-models/cycy233-misllm-v5-5-v1_reward
cycy233-misllm-v5-5-v1-mkmlizer: cp /tmp/reward_cache/config.json s3://guanaco-reward-models/cycy233-misllm-v5-5-v1_reward/config.json
cycy233-misllm-v5-5-v1-mkmlizer: cp /tmp/reward_cache/special_tokens_map.json s3://guanaco-reward-models/cycy233-misllm-v5-5-v1_reward/special_tokens_map.json
cycy233-misllm-v5-5-v1-mkmlizer: cp /tmp/reward_cache/tokenizer_config.json s3://guanaco-reward-models/cycy233-misllm-v5-5-v1_reward/tokenizer_config.json
cycy233-misllm-v5-5-v1-mkmlizer: cp /tmp/reward_cache/merges.txt s3://guanaco-reward-models/cycy233-misllm-v5-5-v1_reward/merges.txt
cycy233-misllm-v5-5-v1-mkmlizer: cp /tmp/reward_cache/vocab.json s3://guanaco-reward-models/cycy233-misllm-v5-5-v1_reward/vocab.json
cycy233-misllm-v5-5-v1-mkmlizer: cp /tmp/reward_cache/tokenizer.json s3://guanaco-reward-models/cycy233-misllm-v5-5-v1_reward/tokenizer.json
Job cycy233-misllm-v5-5-v1-mkmlizer completed after 84.5s with status: succeeded
Stopping job with name cycy233-misllm-v5-5-v1-mkmlizer
Pipeline stage MKMLizer completed in 89.26s
Running pipeline stage MKMLKubeTemplater
Pipeline stage MKMLKubeTemplater completed in 0.08s
Running pipeline stage ISVCDeployer
Creating inference service cycy233-misllm-v5-5-v1
Waiting for inference service cycy233-misllm-v5-5-v1 to be ready
Inference service cycy233-misllm-v5-5-v1 ready after 30.208329439163208s
Pipeline stage ISVCDeployer completed in 37.86s
Running pipeline stage StressChecker
Received healthy response to inference request in 1.9645094871520996s
Received healthy response to inference request in 1.0595743656158447s
Received healthy response to inference request in 1.1816167831420898s
Received healthy response to inference request in 1.1811730861663818s
Received healthy response to inference request in 1.1902744770050049s
5 requests
0 failed requests
5th percentile: 1.0838941097259522
10th percentile: 1.1082138538360595
20th percentile: 1.1568533420562743
30th percentile: 1.1812618255615235
40th percentile: 1.1814393043518066
50th percentile: 1.1816167831420898
60th percentile: 1.185079860687256
70th percentile: 1.188542938232422
80th percentile: 1.345121479034424
90th percentile: 1.6548154830932618
95th percentile: 1.8096624851226806
99th percentile: 1.9335400867462158
mean time: 1.3154296398162841
Pipeline stage StressChecker completed in 7.27s
Running pipeline stage DaemonicModelEvalScorer
Pipeline stage DaemonicModelEvalScorer completed in 0.03s
Running M-Eval for topic stay_in_character
Running pipeline stage DaemonicSafetyScorer
M-Eval Dataset for topic stay_in_character is loaded
Pipeline stage DaemonicSafetyScorer completed in 0.06s
cycy233-misllm-v5-5_v1 status is now deployed due to DeploymentManager action
cycy233-misllm-v5-5_v1 status is now inactive due to auto deactivation removed underperforming models
admin requested tearing down of cycy233-misllm-v5-5_v1
Running pipeline stage ISVCDeleter
Checking if service cycy233-misllm-v5-5-v1 is running
Tearing down inference service cycy233-misllm-v5-5-v1
Toredown service cycy233-misllm-v5-5-v1
Pipeline stage ISVCDeleter completed in 4.60s
Running pipeline stage MKMLModelDeleter
Cleaning model data from S3
Cleaning model data from model cache
Deleting key cycy233-misllm-v5-5-v1/config.json from bucket guanaco-mkml-models
Deleting key cycy233-misllm-v5-5-v1/flywheel_model.0.safetensors from bucket guanaco-mkml-models
Deleting key cycy233-misllm-v5-5-v1/special_tokens_map.json from bucket guanaco-mkml-models
Deleting key cycy233-misllm-v5-5-v1/tokenizer.json from bucket guanaco-mkml-models
Deleting key cycy233-misllm-v5-5-v1/tokenizer.model from bucket guanaco-mkml-models
Deleting key cycy233-misllm-v5-5-v1/tokenizer_config.json from bucket guanaco-mkml-models
Cleaning model data from model cache
Deleting key cycy233-misllm-v5-5-v1_reward/config.json from bucket guanaco-reward-models
Deleting key cycy233-misllm-v5-5-v1_reward/merges.txt from bucket guanaco-reward-models
Deleting key cycy233-misllm-v5-5-v1_reward/reward.tensors from bucket guanaco-reward-models
Deleting key cycy233-misllm-v5-5-v1_reward/special_tokens_map.json from bucket guanaco-reward-models
Deleting key cycy233-misllm-v5-5-v1_reward/tokenizer.json from bucket guanaco-reward-models
Deleting key cycy233-misllm-v5-5-v1_reward/tokenizer_config.json from bucket guanaco-reward-models
Deleting key cycy233-misllm-v5-5-v1_reward/vocab.json from bucket guanaco-reward-models
Pipeline stage MKMLModelDeleter completed in 2.58s
cycy233-misllm-v5-5_v1 status is now torndown due to DeploymentManager action