submission_id: hastagaras-llama-3-8b-well_v2
developer_uid: Hastagaras
status: torndown
model_repo: Hastagaras/llama-3-8b-well
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': "<|begin_of_text|>Roleplay scenario of {bot_name} and {user_name}.\n\n{bot_name}'s Persona:\n{memory}\n\n", 'prompt_template': '{bot_name}: {prompt}\n', 'bot_template': '{bot_name}: {message}\n', 'user_template': '{user_name}: {message}\n', 'response_template': '{bot_name}:', 'truncate_by_message': False}
reward_formatter: {'memory_template': "{bot_name}'s Persona: {memory}\n", 'prompt_template': '{prompt}\n', 'bot_template': 'Bot: {message}\n', 'user_template': 'User: {message}\n', 'response_template': 'Bot:', 'truncate_by_message': False}
timestamp: 2024-05-14T03:52:45+00:00
model_name: reduce-safety
model_eval_status: success
model_group: Hastagaras/llama-3-8b-we
num_battles: 12175
num_wins: 6253
celo_rating: 1182.01
propriety_score: 0.0
propriety_total_count: 0.0
submission_type: basic
model_architecture: LlamaForCausalLM
model_num_parameters: 8030261248.0
best_of: 16
max_input_tokens: 512
max_output_tokens: 64
display_name: reduce-safety
ineligible_reason: propriety_total_count < 800
language_model: Hastagaras/llama-3-8b-well
model_size: 8B
reward_model: ChaiML/reward_gpt2_medium_preference_24m_e2
us_pacific_date: 2024-05-13
win_ratio: 0.5135934291581109
preference_data_url: None
Resubmit model
Running pipeline stage MKMLizer
Starting job with name hastagaras-llama-3-8b-well-v2-mkmlizer
Waiting for job on hastagaras-llama-3-8b-well-v2-mkmlizer to finish
hastagaras-llama-3-8b-well-v2-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
hastagaras-llama-3-8b-well-v2-mkmlizer: ║     _____            __           __                                ║
hastagaras-llama-3-8b-well-v2-mkmlizer: ║    / _/ /_ ___    __/ /  ___ ___ / /                                ║
hastagaras-llama-3-8b-well-v2-mkmlizer: ║   / _/ / // / |/|/ / _ \/ -_) -_) /                                 ║
hastagaras-llama-3-8b-well-v2-mkmlizer: ║  /_//_/\_, /|__,__/_//_/\__/\__/_/                                  ║
hastagaras-llama-3-8b-well-v2-mkmlizer: ║       /___/                                                         ║
hastagaras-llama-3-8b-well-v2-mkmlizer: ║                                                                     ║
hastagaras-llama-3-8b-well-v2-mkmlizer: ║  Version: 0.8.10                                                    ║
hastagaras-llama-3-8b-well-v2-mkmlizer: ║  Copyright 2023 MK ONE TECHNOLOGIES Inc.                            ║
hastagaras-llama-3-8b-well-v2-mkmlizer: ║                                                                     ║
hastagaras-llama-3-8b-well-v2-mkmlizer: ║  The license key for the current software has been verified as      ║
hastagaras-llama-3-8b-well-v2-mkmlizer: ║  belonging to:                                                      ║
hastagaras-llama-3-8b-well-v2-mkmlizer: ║                                                                     ║
hastagaras-llama-3-8b-well-v2-mkmlizer: ║  Chai Research Corp.                                                ║
hastagaras-llama-3-8b-well-v2-mkmlizer: ║  Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f                   ║
hastagaras-llama-3-8b-well-v2-mkmlizer: ║  Expiration: 2024-07-15 23:59:59                                    ║
hastagaras-llama-3-8b-well-v2-mkmlizer: ║                                                                     ║
hastagaras-llama-3-8b-well-v2-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
hastagaras-llama-3-8b-well-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.
hastagaras-llama-3-8b-well-v2-mkmlizer:   warnings.warn(warning_message, FutureWarning)
hastagaras-llama-3-8b-well-v2-mkmlizer: Downloaded to shared memory in 17.741s
hastagaras-llama-3-8b-well-v2-mkmlizer: quantizing model to /dev/shm/model_cache
hastagaras-llama-3-8b-well-v2-mkmlizer: Saving flywheel model at /dev/shm/model_cache
hastagaras-llama-3-8b-well-v2-mkmlizer: 
Loading 0:   0%|          | 0/291 [00:00<?, ?it/s]
Loading 0:  38%|███▊      | 112/291 [00:01<00:01, 110.93it/s]
Loading 0:  82%|████████▏ | 239/291 [00:02<00:00, 119.90it/s]
Loading 0:  99%|█████████▊| 287/291 [00:07<00:00, 30.59it/s] 
                                                            
Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
hastagaras-llama-3-8b-well-v2-mkmlizer: quantized model in 17.605s
hastagaras-llama-3-8b-well-v2-mkmlizer: Processed model Hastagaras/llama-3-8b-well in 36.459s
hastagaras-llama-3-8b-well-v2-mkmlizer: creating bucket guanaco-mkml-models
hastagaras-llama-3-8b-well-v2-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
hastagaras-llama-3-8b-well-v2-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/hastagaras-llama-3-8b-well-v2
hastagaras-llama-3-8b-well-v2-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/hastagaras-llama-3-8b-well-v2/config.json
hastagaras-llama-3-8b-well-v2-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/hastagaras-llama-3-8b-well-v2/tokenizer_config.json
hastagaras-llama-3-8b-well-v2-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/hastagaras-llama-3-8b-well-v2/special_tokens_map.json
hastagaras-llama-3-8b-well-v2-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/hastagaras-llama-3-8b-well-v2/tokenizer.json
hastagaras-llama-3-8b-well-v2-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/hastagaras-llama-3-8b-well-v2/flywheel_model.0.safetensors
hastagaras-llama-3-8b-well-v2-mkmlizer: loading reward model from ChaiML/reward_gpt2_medium_preference_24m_e2
hastagaras-llama-3-8b-well-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.
hastagaras-llama-3-8b-well-v2-mkmlizer:   warnings.warn(
hastagaras-llama-3-8b-well-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.
hastagaras-llama-3-8b-well-v2-mkmlizer:   warnings.warn(
hastagaras-llama-3-8b-well-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()
hastagaras-llama-3-8b-well-v2-mkmlizer:   return self.fget.__get__(instance, owner)()
hastagaras-llama-3-8b-well-v2-mkmlizer: Saving model to /tmp/reward_cache/reward.tensors
hastagaras-llama-3-8b-well-v2-mkmlizer: Saving duration: 0.241s
hastagaras-llama-3-8b-well-v2-mkmlizer: Processed model ChaiML/reward_gpt2_medium_preference_24m_e2 in 3.526s
hastagaras-llama-3-8b-well-v2-mkmlizer: creating bucket guanaco-reward-models
hastagaras-llama-3-8b-well-v2-mkmlizer: Bucket 's3://guanaco-reward-models/' created
hastagaras-llama-3-8b-well-v2-mkmlizer: uploading /tmp/reward_cache to s3://guanaco-reward-models/hastagaras-llama-3-8b-well-v2_reward
hastagaras-llama-3-8b-well-v2-mkmlizer: cp /tmp/reward_cache/config.json s3://guanaco-reward-models/hastagaras-llama-3-8b-well-v2_reward/config.json
hastagaras-llama-3-8b-well-v2-mkmlizer: cp /tmp/reward_cache/tokenizer_config.json s3://guanaco-reward-models/hastagaras-llama-3-8b-well-v2_reward/tokenizer_config.json
hastagaras-llama-3-8b-well-v2-mkmlizer: cp /tmp/reward_cache/special_tokens_map.json s3://guanaco-reward-models/hastagaras-llama-3-8b-well-v2_reward/special_tokens_map.json
hastagaras-llama-3-8b-well-v2-mkmlizer: cp /tmp/reward_cache/merges.txt s3://guanaco-reward-models/hastagaras-llama-3-8b-well-v2_reward/merges.txt
hastagaras-llama-3-8b-well-v2-mkmlizer: cp /tmp/reward_cache/vocab.json s3://guanaco-reward-models/hastagaras-llama-3-8b-well-v2_reward/vocab.json
hastagaras-llama-3-8b-well-v2-mkmlizer: cp /tmp/reward_cache/tokenizer.json s3://guanaco-reward-models/hastagaras-llama-3-8b-well-v2_reward/tokenizer.json
hastagaras-llama-3-8b-well-v2-mkmlizer: cp /tmp/reward_cache/reward.tensors s3://guanaco-reward-models/hastagaras-llama-3-8b-well-v2_reward/reward.tensors
Job hastagaras-llama-3-8b-well-v2-mkmlizer completed after 63.24s with status: succeeded
Stopping job with name hastagaras-llama-3-8b-well-v2-mkmlizer
Pipeline stage MKMLizer completed in 68.12s
Running pipeline stage MKMLKubeTemplater
Pipeline stage MKMLKubeTemplater completed in 0.09s
Running pipeline stage ISVCDeployer
Creating inference service hastagaras-llama-3-8b-well-v2
Waiting for inference service hastagaras-llama-3-8b-well-v2 to be ready
Inference service hastagaras-llama-3-8b-well-v2 ready after 40.56714916229248s
Pipeline stage ISVCDeployer completed in 48.41s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.114213228225708s
Received healthy response to inference request in 1.2371113300323486s
Received healthy response to inference request in 1.2581331729888916s
Received healthy response to inference request in 1.2469308376312256s
Received healthy response to inference request in 1.2340290546417236s
5 requests
0 failed requests
5th percentile: 1.2346455097198485
10th percentile: 1.2352619647979737
20th percentile: 1.2364948749542237
30th percentile: 1.239075231552124
40th percentile: 1.2430030345916747
50th percentile: 1.2469308376312256
60th percentile: 1.251411771774292
70th percentile: 1.2558927059173584
80th percentile: 1.429349184036255
90th percentile: 1.7717812061309814
95th percentile: 1.9429972171783445
99th percentile: 2.0799700260162353
mean time: 1.4180835247039796
Pipeline stage StressChecker completed in 7.83s
Running pipeline stage DaemonicModelEvalScorer
Pipeline stage DaemonicModelEvalScorer completed in 0.04s
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
hastagaras-llama-3-8b-well_v2 status is now deployed due to DeploymentManager action
hastagaras-llama-3-8b-well_v2 status is now inactive due to auto deactivation removed underperforming models
admin requested tearing down of hastagaras-llama-3-8b-well_v2
Running pipeline stage ISVCDeleter
Checking if service hastagaras-llama-3-8b-well-v2 is running
Tearing down inference service hastagaras-llama-3-8b-well-v2
Toredown service hastagaras-llama-3-8b-well-v2
Pipeline stage ISVCDeleter completed in 4.50s
Running pipeline stage MKMLModelDeleter
Cleaning model data from S3
Cleaning model data from model cache
Deleting key hastagaras-llama-3-8b-well-v2/config.json from bucket guanaco-mkml-models
Deleting key hastagaras-llama-3-8b-well-v2/flywheel_model.0.safetensors from bucket guanaco-mkml-models
Deleting key hastagaras-llama-3-8b-well-v2/special_tokens_map.json from bucket guanaco-mkml-models
Deleting key hastagaras-llama-3-8b-well-v2/tokenizer.json from bucket guanaco-mkml-models
Deleting key hastagaras-llama-3-8b-well-v2/tokenizer_config.json from bucket guanaco-mkml-models
Cleaning model data from model cache
Deleting key hastagaras-llama-3-8b-well-v2_reward/config.json from bucket guanaco-reward-models
Deleting key hastagaras-llama-3-8b-well-v2_reward/merges.txt from bucket guanaco-reward-models
Deleting key hastagaras-llama-3-8b-well-v2_reward/reward.tensors from bucket guanaco-reward-models
Deleting key hastagaras-llama-3-8b-well-v2_reward/special_tokens_map.json from bucket guanaco-reward-models
Deleting key hastagaras-llama-3-8b-well-v2_reward/tokenizer.json from bucket guanaco-reward-models
Deleting key hastagaras-llama-3-8b-well-v2_reward/tokenizer_config.json from bucket guanaco-reward-models
Deleting key hastagaras-llama-3-8b-well-v2_reward/vocab.json from bucket guanaco-reward-models
Pipeline stage MKMLModelDeleter completed in 2.65s
hastagaras-llama-3-8b-well_v2 status is now torndown due to DeploymentManager action

Usage Metrics

Latency Metrics