submission_id: setiaku-run1-rescaled_v1
developer_uid: sao10k
status: inactive
model_repo: Setiaku/Run1-Rescaled
reward_repo: ChaiML/reward_gpt2_medium_preference_24m_e2
generation_params: {'temperature': 1.22, 'top_p': 1.0, 'min_p': 0.075, 'top_k': 40, 'presence_penalty': 0.0, 'frequency_penalty': 0.0, 'stopping_words': ['\n,', '<|end_header_id|>,', '<|eot_id|>,', '\n\n{user_name}'], '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\nThis is an entertaining conversation. You are {bot_name} who has the persona: {memory}.\nEngage in a chat with {user_name} while staying in character. Try to flirt with {user_name}. Engage in *roleplay* actions. Describe the scene dramatically.', 'prompt_template': 'Example Conversation:\n{prompt}<|eot_id|>', 'bot_template': '<|start_header_id|>[{bot_name}]<|end_header_id|>\n\n{message}<|eot_id|>', 'user_template': '<|start_header_id|>[{user_name}]<|end_header_id|>\n\n{message}<|eot_id|>', 'response_template': '<|start_header_id|>[{bot_name}]<|end_header_id|>\n\n', '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': False}
timestamp: 2024-05-28T11:51:42+00:00
model_name: L3-Run-1-Formatted-Test
model_eval_status: success
model_group: Setiaku/Run1-Rescaled
num_battles: 7775
num_wins: 4192
celo_rating: 1202.62
safety_score: 0.84
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: L3-Run-1-Formatted-Test
ineligible_reason: propriety_total_count < 5000
language_model: Setiaku/Run1-Rescaled
model_size: 8B
reward_model: ChaiML/reward_gpt2_medium_preference_24m_e2
us_pacific_date: 2024-05-28
win_ratio: 0.5391639871382636
Resubmit model
Running pipeline stage MKMLizer
Starting job with name setiaku-run1-rescaled-v1-mkmlizer
Waiting for job on setiaku-run1-rescaled-v1-mkmlizer to finish
setiaku-run1-rescaled-v1-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
setiaku-run1-rescaled-v1-mkmlizer: ║ _____ __ __ ║
setiaku-run1-rescaled-v1-mkmlizer: ║ / _/ /_ ___ __/ / ___ ___ / / ║
setiaku-run1-rescaled-v1-mkmlizer: ║ / _/ / // / |/|/ / _ \/ -_) -_) / ║
setiaku-run1-rescaled-v1-mkmlizer: ║ /_//_/\_, /|__,__/_//_/\__/\__/_/ ║
setiaku-run1-rescaled-v1-mkmlizer: ║ /___/ ║
setiaku-run1-rescaled-v1-mkmlizer: ║ ║
setiaku-run1-rescaled-v1-mkmlizer: ║ Version: 0.8.14 ║
setiaku-run1-rescaled-v1-mkmlizer: ║ Copyright 2023 MK ONE TECHNOLOGIES Inc. ║
setiaku-run1-rescaled-v1-mkmlizer: ║ https://mk1.ai ║
setiaku-run1-rescaled-v1-mkmlizer: ║ ║
setiaku-run1-rescaled-v1-mkmlizer: ║ The license key for the current software has been verified as ║
setiaku-run1-rescaled-v1-mkmlizer: ║ belonging to: ║
setiaku-run1-rescaled-v1-mkmlizer: ║ ║
setiaku-run1-rescaled-v1-mkmlizer: ║ Chai Research Corp. ║
setiaku-run1-rescaled-v1-mkmlizer: ║ Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f ║
setiaku-run1-rescaled-v1-mkmlizer: ║ Expiration: 2024-07-15 23:59:59 ║
setiaku-run1-rescaled-v1-mkmlizer: ║ ║
setiaku-run1-rescaled-v1-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
setiaku-run1-rescaled-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.
setiaku-run1-rescaled-v1-mkmlizer: warnings.warn(warning_message, FutureWarning)
setiaku-run1-rescaled-v1-mkmlizer: Downloaded to shared memory in 26.182s
setiaku-run1-rescaled-v1-mkmlizer: quantizing model to /dev/shm/model_cache
setiaku-run1-rescaled-v1-mkmlizer: Saving flywheel model at /dev/shm/model_cache
setiaku-run1-rescaled-v1-mkmlizer: Loading 0: 0%| | 0/291 [00:00<?, ?it/s] Loading 0: 1%| | 2/291 [00:04<09:45, 2.02s/it] Loading 0: 8%|▊ | 22/291 [00:04<00:36, 7.29it/s] Loading 0: 14%|█▍ | 42/291 [00:04<00:15, 16.16it/s] Loading 0: 21%|██ | 60/291 [00:04<00:10, 22.47it/s] Loading 0: 29%|██▉ | 85/291 [00:04<00:05, 38.25it/s] Loading 0: 36%|███▌ | 104/291 [00:04<00:03, 51.89it/s] Loading 0: 42%|████▏ | 123/291 [00:04<00:02, 67.55it/s] Loading 0: 48%|████▊ | 141/291 [00:05<00:01, 82.10it/s] Loading 0: 57%|█████▋ | 166/291 [00:05<00:01, 74.54it/s] Loading 0: 64%|██████▎ | 185/291 [00:05<00:01, 90.31it/s] Loading 0: 70%|███████ | 205/291 [00:05<00:00, 108.18it/s] Loading 0: 79%|███████▊ | 229/291 [00:05<00:00, 131.52it/s] Loading 0: 86%|████████▌ | 249/291 [00:05<00:00, 143.26it/s] Loading 0: 92%|█████████▏| 268/291 [00:06<00:00, 97.87it/s] Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
setiaku-run1-rescaled-v1-mkmlizer: quantized model in 17.352s
setiaku-run1-rescaled-v1-mkmlizer: Processed model Setiaku/Run1-Rescaled in 44.481s
setiaku-run1-rescaled-v1-mkmlizer: creating bucket guanaco-mkml-models
setiaku-run1-rescaled-v1-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
setiaku-run1-rescaled-v1-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/setiaku-run1-rescaled-v1
setiaku-run1-rescaled-v1-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/setiaku-run1-rescaled-v1/special_tokens_map.json
setiaku-run1-rescaled-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/setiaku-run1-rescaled-v1/tokenizer_config.json
setiaku-run1-rescaled-v1-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/setiaku-run1-rescaled-v1/config.json
setiaku-run1-rescaled-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/setiaku-run1-rescaled-v1/tokenizer.json
setiaku-run1-rescaled-v1-mkmlizer: loading reward model from ChaiML/reward_gpt2_medium_preference_24m_e2
setiaku-run1-rescaled-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.
setiaku-run1-rescaled-v1-mkmlizer: warnings.warn(
setiaku-run1-rescaled-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.
setiaku-run1-rescaled-v1-mkmlizer: warnings.warn(
setiaku-run1-rescaled-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.
setiaku-run1-rescaled-v1-mkmlizer: warnings.warn(
setiaku-run1-rescaled-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()
setiaku-run1-rescaled-v1-mkmlizer: return self.fget.__get__(instance, owner)()
setiaku-run1-rescaled-v1-mkmlizer: Saving model to /tmp/reward_cache/reward.tensors
setiaku-run1-rescaled-v1-mkmlizer: Saving duration: 0.210s
setiaku-run1-rescaled-v1-mkmlizer: Processed model ChaiML/reward_gpt2_medium_preference_24m_e2 in 6.888s
setiaku-run1-rescaled-v1-mkmlizer: creating bucket guanaco-reward-models
setiaku-run1-rescaled-v1-mkmlizer: Bucket 's3://guanaco-reward-models/' created
setiaku-run1-rescaled-v1-mkmlizer: uploading /tmp/reward_cache to s3://guanaco-reward-models/setiaku-run1-rescaled-v1_reward
setiaku-run1-rescaled-v1-mkmlizer: cp /tmp/reward_cache/special_tokens_map.json s3://guanaco-reward-models/setiaku-run1-rescaled-v1_reward/special_tokens_map.json
setiaku-run1-rescaled-v1-mkmlizer: cp /tmp/reward_cache/config.json s3://guanaco-reward-models/setiaku-run1-rescaled-v1_reward/config.json
setiaku-run1-rescaled-v1-mkmlizer: cp /tmp/reward_cache/vocab.json s3://guanaco-reward-models/setiaku-run1-rescaled-v1_reward/vocab.json
setiaku-run1-rescaled-v1-mkmlizer: cp /tmp/reward_cache/merges.txt s3://guanaco-reward-models/setiaku-run1-rescaled-v1_reward/merges.txt
setiaku-run1-rescaled-v1-mkmlizer: cp /tmp/reward_cache/tokenizer_config.json s3://guanaco-reward-models/setiaku-run1-rescaled-v1_reward/tokenizer_config.json
setiaku-run1-rescaled-v1-mkmlizer: cp /tmp/reward_cache/tokenizer.json s3://guanaco-reward-models/setiaku-run1-rescaled-v1_reward/tokenizer.json
setiaku-run1-rescaled-v1-mkmlizer: cp /tmp/reward_cache/reward.tensors s3://guanaco-reward-models/setiaku-run1-rescaled-v1_reward/reward.tensors
Job setiaku-run1-rescaled-v1-mkmlizer completed after 72.72s with status: succeeded
Stopping job with name setiaku-run1-rescaled-v1-mkmlizer
Pipeline stage MKMLizer completed in 76.94s
Running pipeline stage MKMLKubeTemplater
Pipeline stage MKMLKubeTemplater completed in 0.10s
Running pipeline stage ISVCDeployer
Creating inference service setiaku-run1-rescaled-v1
Waiting for inference service setiaku-run1-rescaled-v1 to be ready
Inference service setiaku-run1-rescaled-v1 ready after 30.292577743530273s
Pipeline stage ISVCDeployer completed in 37.54s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.1943838596343994s
Received healthy response to inference request in 1.2583906650543213s
Received healthy response to inference request in 1.3429677486419678s
Received healthy response to inference request in 1.3307592868804932s
Received healthy response to inference request in 1.293008804321289s
5 requests
0 failed requests
5th percentile: 1.2653142929077148
10th percentile: 1.2722379207611083
20th percentile: 1.2860851764678956
30th percentile: 1.30055890083313
40th percentile: 1.3156590938568116
50th percentile: 1.3307592868804932
60th percentile: 1.335642671585083
70th percentile: 1.3405260562896728
80th percentile: 1.5132509708404542
90th percentile: 1.853817415237427
95th percentile: 2.0241006374359127
99th percentile: 2.160327215194702
mean time: 1.4839020729064942
Pipeline stage StressChecker completed in 8.04s
Running pipeline stage DaemonicModelEvalScorer
Pipeline stage DaemonicModelEvalScorer completed in 0.03s
Running pipeline stage DaemonicSafetyScorer
Pipeline stage DaemonicSafetyScorer completed in 0.03s
Running M-Eval for topic stay_in_character
setiaku-run1-rescaled_v1 status is now deployed due to DeploymentManager action
M-Eval Dataset for topic stay_in_character is loaded
setiaku-run1-rescaled_v1 status is now inactive due to auto deactivation removed underperforming models

Usage Metrics

Latency Metrics