submission_id: cycy233-misllm-v7-8-2-5_v2
developer_uid: shiroe40
status: inactive
model_repo: cycy233/misllm_v7.8.2.5
reward_repo: ChaiML/reward_gpt2_medium_preference_24m_e2
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': True}
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': True}
timestamp: 2024-06-17T03:20:20+00:00
model_name: base
model_eval_status: success
model_group: cycy233/misllm_v7.8.2.5
num_battles: 15261
num_wins: 7691
celo_rating: 1193.54
propriety_score: 0.6925219328784292
propriety_total_count: 7181.0
submission_type: basic
model_architecture: MistralForCausalLM
model_num_parameters: 7248023552.0
best_of: 16
max_input_tokens: 512
max_output_tokens: 64
display_name: base
ineligible_reason: None
language_model: cycy233/misllm_v7.8.2.5
model_size: 7B
reward_model: ChaiML/reward_gpt2_medium_preference_24m_e2
us_pacific_date: 2024-06-16
win_ratio: 0.5039643535810235
Resubmit model
Running pipeline stage MKMLizer
Starting job with name cycy233-misllm-v7-8-2-5-v2-mkmlizer
Waiting for job on cycy233-misllm-v7-8-2-5-v2-mkmlizer to finish
cycy233-misllm-v7-8-2-5-v2-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
cycy233-misllm-v7-8-2-5-v2-mkmlizer: ║ _____ __ __ ║
cycy233-misllm-v7-8-2-5-v2-mkmlizer: ║ / _/ /_ ___ __/ / ___ ___ / / ║
cycy233-misllm-v7-8-2-5-v2-mkmlizer: ║ / _/ / // / |/|/ / _ \/ -_) -_) / ║
cycy233-misllm-v7-8-2-5-v2-mkmlizer: ║ /_//_/\_, /|__,__/_//_/\__/\__/_/ ║
cycy233-misllm-v7-8-2-5-v2-mkmlizer: ║ /___/ ║
cycy233-misllm-v7-8-2-5-v2-mkmlizer: ║ ║
cycy233-misllm-v7-8-2-5-v2-mkmlizer: ║ Version: 0.8.14 ║
cycy233-misllm-v7-8-2-5-v2-mkmlizer: ║ Copyright 2023 MK ONE TECHNOLOGIES Inc. ║
cycy233-misllm-v7-8-2-5-v2-mkmlizer: ║ https://mk1.ai ║
cycy233-misllm-v7-8-2-5-v2-mkmlizer: ║ ║
cycy233-misllm-v7-8-2-5-v2-mkmlizer: ║ The license key for the current software has been verified as ║
cycy233-misllm-v7-8-2-5-v2-mkmlizer: ║ belonging to: ║
cycy233-misllm-v7-8-2-5-v2-mkmlizer: ║ ║
cycy233-misllm-v7-8-2-5-v2-mkmlizer: ║ Chai Research Corp. ║
cycy233-misllm-v7-8-2-5-v2-mkmlizer: ║ Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f ║
cycy233-misllm-v7-8-2-5-v2-mkmlizer: ║ Expiration: 2024-07-15 23:59:59 ║
cycy233-misllm-v7-8-2-5-v2-mkmlizer: ║ ║
cycy233-misllm-v7-8-2-5-v2-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
cycy233-misllm-v7-8-2-5-v2-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-v7-8-2-5-v2-mkmlizer: warnings.warn(warning_message, FutureWarning)
cycy233-misllm-v7-8-2-5-v2-mkmlizer: Downloaded to shared memory in 31.416s
cycy233-misllm-v7-8-2-5-v2-mkmlizer: quantizing model to /dev/shm/model_cache
cycy233-misllm-v7-8-2-5-v2-mkmlizer: Saving flywheel model at /dev/shm/model_cache
cycy233-misllm-v7-8-2-5-v2-mkmlizer: Loading 0: 0%| | 0/291 [00:00<?, ?it/s] Loading 0: 34%|███▎ | 98/291 [00:01<00:02, 71.15it/s] Loading 0: 70%|███████ | 204/291 [00:02<00:01, 83.31it/s] Loading 0: 100%|██████████| 291/291 [00:03<00:00, 71.40it/s] You set `add_prefix_space`. The tokenizer needs to be converted from the slow tokenizers
cycy233-misllm-v7-8-2-5-v2-mkmlizer: quantized model in 15.600s
cycy233-misllm-v7-8-2-5-v2-mkmlizer: Processed model cycy233/misllm_v7.8.2.5 in 49.362s
cycy233-misllm-v7-8-2-5-v2-mkmlizer: creating bucket guanaco-mkml-models
cycy233-misllm-v7-8-2-5-v2-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
cycy233-misllm-v7-8-2-5-v2-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/cycy233-misllm-v7-8-2-5-v2
cycy233-misllm-v7-8-2-5-v2-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/cycy233-misllm-v7-8-2-5-v2/config.json
cycy233-misllm-v7-8-2-5-v2-mkmlizer: cp /dev/shm/model_cache/tokenizer.model s3://guanaco-mkml-models/cycy233-misllm-v7-8-2-5-v2/tokenizer.model
cycy233-misllm-v7-8-2-5-v2-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/cycy233-misllm-v7-8-2-5-v2/tokenizer.json
cycy233-misllm-v7-8-2-5-v2-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/cycy233-misllm-v7-8-2-5-v2/tokenizer_config.json
cycy233-misllm-v7-8-2-5-v2-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/cycy233-misllm-v7-8-2-5-v2/special_tokens_map.json
cycy233-misllm-v7-8-2-5-v2-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/cycy233-misllm-v7-8-2-5-v2/flywheel_model.0.safetensors
cycy233-misllm-v7-8-2-5-v2-mkmlizer: loading reward model from ChaiML/reward_gpt2_medium_preference_24m_e2
cycy233-misllm-v7-8-2-5-v2-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-v7-8-2-5-v2-mkmlizer: warnings.warn(
cycy233-misllm-v7-8-2-5-v2-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-v7-8-2-5-v2-mkmlizer: warnings.warn(
cycy233-misllm-v7-8-2-5-v2-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-v7-8-2-5-v2-mkmlizer: warnings.warn(
cycy233-misllm-v7-8-2-5-v2-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-v7-8-2-5-v2-mkmlizer: return self.fget.__get__(instance, owner)()
cycy233-misllm-v7-8-2-5-v2-mkmlizer: Saving model to /tmp/reward_cache/reward.tensors
cycy233-misllm-v7-8-2-5-v2-mkmlizer: Saving duration: 0.437s
cycy233-misllm-v7-8-2-5-v2-mkmlizer: Processed model ChaiML/reward_gpt2_medium_preference_24m_e2 in 165.952s
cycy233-misllm-v7-8-2-5-v2-mkmlizer: Bucket 's3://guanaco-reward-models/' created
cycy233-misllm-v7-8-2-5-v2-mkmlizer: uploading /tmp/reward_cache to s3://guanaco-reward-models/cycy233-misllm-v7-8-2-5-v2_reward
cycy233-misllm-v7-8-2-5-v2-mkmlizer: cp /tmp/reward_cache/special_tokens_map.json s3://guanaco-reward-models/cycy233-misllm-v7-8-2-5-v2_reward/special_tokens_map.json
cycy233-misllm-v7-8-2-5-v2-mkmlizer: cp /tmp/reward_cache/vocab.json s3://guanaco-reward-models/cycy233-misllm-v7-8-2-5-v2_reward/vocab.json
cycy233-misllm-v7-8-2-5-v2-mkmlizer: cp /tmp/reward_cache/tokenizer_config.json s3://guanaco-reward-models/cycy233-misllm-v7-8-2-5-v2_reward/tokenizer_config.json
cycy233-misllm-v7-8-2-5-v2-mkmlizer: cp /tmp/reward_cache/config.json s3://guanaco-reward-models/cycy233-misllm-v7-8-2-5-v2_reward/config.json
cycy233-misllm-v7-8-2-5-v2-mkmlizer: cp /tmp/reward_cache/merges.txt s3://guanaco-reward-models/cycy233-misllm-v7-8-2-5-v2_reward/merges.txt
cycy233-misllm-v7-8-2-5-v2-mkmlizer: cp /tmp/reward_cache/tokenizer.json s3://guanaco-reward-models/cycy233-misllm-v7-8-2-5-v2_reward/tokenizer.json
cycy233-misllm-v7-8-2-5-v2-mkmlizer: cp /tmp/reward_cache/reward.tensors s3://guanaco-reward-models/cycy233-misllm-v7-8-2-5-v2_reward/reward.tensors
Job cycy233-misllm-v7-8-2-5-v2-mkmlizer completed after 236.87s with status: succeeded
Stopping job with name cycy233-misllm-v7-8-2-5-v2-mkmlizer
Pipeline stage MKMLizer completed in 240.61s
Running pipeline stage MKMLKubeTemplater
Pipeline stage MKMLKubeTemplater completed in 0.10s
Running pipeline stage ISVCDeployer
Creating inference service cycy233-misllm-v7-8-2-5-v2
Waiting for inference service cycy233-misllm-v7-8-2-5-v2 to be ready
Inference service cycy233-misllm-v7-8-2-5-v2 ready after 40.24504327774048s
Pipeline stage ISVCDeployer completed in 47.80s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.1096105575561523s
Received healthy response to inference request in 1.1045444011688232s
Received healthy response to inference request in 1.0528521537780762s
Received healthy response to inference request in 1.047264575958252s
Received healthy response to inference request in 1.162161111831665s
5 requests
0 failed requests
5th percentile: 1.0483820915222168
10th percentile: 1.0494996070861817
20th percentile: 1.0517346382141113
30th percentile: 1.0631906032562255
40th percentile: 1.0838675022125244
50th percentile: 1.1045444011688232
60th percentile: 1.12759108543396
70th percentile: 1.1506377696990966
80th percentile: 1.3516510009765628
90th percentile: 1.7306307792663576
95th percentile: 1.9201206684112546
99th percentile: 2.0717125797271727
mean time: 1.2952865600585937
Pipeline stage StressChecker completed in 7.10s
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 0.04s
M-Eval Dataset for topic stay_in_character is loaded
cycy233-misllm-v7-8-2-5_v2 status is now deployed due to DeploymentManager action
cycy233-misllm-v7-8-2-5_v2 status is now inactive due to auto deactivation removed underperforming models

Usage Metrics

Latency Metrics