submission_id: yuyuclimb-yxyllama2-v1-2_v1
developer_uid: yuyuclimb
status: inactive
model_repo: yuyuclimb/yxyllama2-v1.2
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': False}
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: 2024-05-10T09:20:31+00:00
model_name: yxyllama2-v2
model_eval_status: success
double_thumbs_up: 256
thumbs_up: 402
thumbs_down: 220
num_battles: 25369
num_wins: 11610
celo_rating: 1139.72
entertaining: 6.92
stay_in_character: 8.33
user_preference: 6.92
safety_score: 0.64
submission_type: basic
model_architecture: LlamaForCausalLM
model_num_parameters: 6738415616.0
best_of: 16
max_input_tokens: 512
max_output_tokens: 64
display_name: yxyllama2-v2
double_thumbs_up_ratio: 0.29157175398633256
feedback_count: 878
ineligible_reason: None
language_model: yuyuclimb/yxyllama2-v1.2
model_score: 7.390000000000001
model_size: 7B
reward_model: ChaiML/reward_gpt2_medium_preference_24m_e2
single_thumbs_up_ratio: 0.45785876993166286
thumbs_down_ratio: 0.2505694760820046
thumbs_up_ratio: 0.7494305239179955
us_pacific_date: 2024-05-10
win_ratio: 0.45764515747565926
Resubmit model
Running pipeline stage MKMLizer
Starting job with name yuyuclimb-yxyllama2-v1-2-v1-mkmlizer
Waiting for job on yuyuclimb-yxyllama2-v1-2-v1-mkmlizer to finish
yuyuclimb-yxyllama2-v1-2-v1-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
yuyuclimb-yxyllama2-v1-2-v1-mkmlizer: ║ _____ __ __ ║
yuyuclimb-yxyllama2-v1-2-v1-mkmlizer: ║ / _/ /_ ___ __/ / ___ ___ / / ║
yuyuclimb-yxyllama2-v1-2-v1-mkmlizer: ║ / _/ / // / |/|/ / _ \/ -_) -_) / ║
yuyuclimb-yxyllama2-v1-2-v1-mkmlizer: ║ /_//_/\_, /|__,__/_//_/\__/\__/_/ ║
yuyuclimb-yxyllama2-v1-2-v1-mkmlizer: ║ /___/ ║
yuyuclimb-yxyllama2-v1-2-v1-mkmlizer: ║ ║
yuyuclimb-yxyllama2-v1-2-v1-mkmlizer: ║ Version: 0.8.10 ║
yuyuclimb-yxyllama2-v1-2-v1-mkmlizer: ║ Copyright 2023 MK ONE TECHNOLOGIES Inc. ║
yuyuclimb-yxyllama2-v1-2-v1-mkmlizer: ║ ║
yuyuclimb-yxyllama2-v1-2-v1-mkmlizer: ║ The license key for the current software has been verified as ║
yuyuclimb-yxyllama2-v1-2-v1-mkmlizer: ║ belonging to: ║
yuyuclimb-yxyllama2-v1-2-v1-mkmlizer: ║ ║
yuyuclimb-yxyllama2-v1-2-v1-mkmlizer: ║ Chai Research Corp. ║
yuyuclimb-yxyllama2-v1-2-v1-mkmlizer: ║ Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f ║
yuyuclimb-yxyllama2-v1-2-v1-mkmlizer: ║ Expiration: 2024-07-15 23:59:59 ║
yuyuclimb-yxyllama2-v1-2-v1-mkmlizer: ║ ║
yuyuclimb-yxyllama2-v1-2-v1-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
yuyuclimb-yxyllama2-v1-2-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.
yuyuclimb-yxyllama2-v1-2-v1-mkmlizer: warnings.warn(warning_message, FutureWarning)
yuyuclimb-yxyllama2-v1-2-v1-mkmlizer: Downloaded to shared memory in 25.863s
yuyuclimb-yxyllama2-v1-2-v1-mkmlizer: quantizing model to /dev/shm/model_cache
yuyuclimb-yxyllama2-v1-2-v1-mkmlizer: Saving flywheel model at /dev/shm/model_cache
yuyuclimb-yxyllama2-v1-2-v1-mkmlizer: Loading 0: 0%| | 0/291 [00:00<?, ?it/s] Loading 0: 36%|███▋ | 106/291 [00:01<00:01, 103.37it/s] Loading 0: 100%|██████████| 291/291 [00:02<00:00, 97.31it/s] You set `add_prefix_space`. The tokenizer needs to be converted from the slow tokenizers
yuyuclimb-yxyllama2-v1-2-v1-mkmlizer: quantized model in 10.491s
yuyuclimb-yxyllama2-v1-2-v1-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
yuyuclimb-yxyllama2-v1-2-v1-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/yuyuclimb-yxyllama2-v1-2-v1
yuyuclimb-yxyllama2-v1-2-v1-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/yuyuclimb-yxyllama2-v1-2-v1/config.json
yuyuclimb-yxyllama2-v1-2-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/yuyuclimb-yxyllama2-v1-2-v1/tokenizer_config.json
yuyuclimb-yxyllama2-v1-2-v1-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/yuyuclimb-yxyllama2-v1-2-v1/special_tokens_map.json
yuyuclimb-yxyllama2-v1-2-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer.model s3://guanaco-mkml-models/yuyuclimb-yxyllama2-v1-2-v1/tokenizer.model
yuyuclimb-yxyllama2-v1-2-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/yuyuclimb-yxyllama2-v1-2-v1/tokenizer.json
yuyuclimb-yxyllama2-v1-2-v1-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/yuyuclimb-yxyllama2-v1-2-v1/flywheel_model.0.safetensors
yuyuclimb-yxyllama2-v1-2-v1-mkmlizer: loading reward model from ChaiML/reward_gpt2_medium_preference_24m_e2
yuyuclimb-yxyllama2-v1-2-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.
yuyuclimb-yxyllama2-v1-2-v1-mkmlizer: warnings.warn(
yuyuclimb-yxyllama2-v1-2-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.
yuyuclimb-yxyllama2-v1-2-v1-mkmlizer: warnings.warn(
yuyuclimb-yxyllama2-v1-2-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.
yuyuclimb-yxyllama2-v1-2-v1-mkmlizer: warnings.warn(
yuyuclimb-yxyllama2-v1-2-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()
yuyuclimb-yxyllama2-v1-2-v1-mkmlizer: return self.fget.__get__(instance, owner)()
yuyuclimb-yxyllama2-v1-2-v1-mkmlizer: Saving model to /tmp/reward_cache/reward.tensors
yuyuclimb-yxyllama2-v1-2-v1-mkmlizer: Saving duration: 0.318s
yuyuclimb-yxyllama2-v1-2-v1-mkmlizer: Processed model ChaiML/reward_gpt2_medium_preference_24m_e2 in 7.347s
yuyuclimb-yxyllama2-v1-2-v1-mkmlizer: creating bucket guanaco-reward-models
yuyuclimb-yxyllama2-v1-2-v1-mkmlizer: Bucket 's3://guanaco-reward-models/' created
yuyuclimb-yxyllama2-v1-2-v1-mkmlizer: uploading /tmp/reward_cache to s3://guanaco-reward-models/yuyuclimb-yxyllama2-v1-2-v1_reward
yuyuclimb-yxyllama2-v1-2-v1-mkmlizer: cp /tmp/reward_cache/config.json s3://guanaco-reward-models/yuyuclimb-yxyllama2-v1-2-v1_reward/config.json
yuyuclimb-yxyllama2-v1-2-v1-mkmlizer: cp /tmp/reward_cache/special_tokens_map.json s3://guanaco-reward-models/yuyuclimb-yxyllama2-v1-2-v1_reward/special_tokens_map.json
yuyuclimb-yxyllama2-v1-2-v1-mkmlizer: cp /tmp/reward_cache/tokenizer_config.json s3://guanaco-reward-models/yuyuclimb-yxyllama2-v1-2-v1_reward/tokenizer_config.json
yuyuclimb-yxyllama2-v1-2-v1-mkmlizer: cp /tmp/reward_cache/merges.txt s3://guanaco-reward-models/yuyuclimb-yxyllama2-v1-2-v1_reward/merges.txt
yuyuclimb-yxyllama2-v1-2-v1-mkmlizer: cp /tmp/reward_cache/vocab.json s3://guanaco-reward-models/yuyuclimb-yxyllama2-v1-2-v1_reward/vocab.json
yuyuclimb-yxyllama2-v1-2-v1-mkmlizer: cp /tmp/reward_cache/tokenizer.json s3://guanaco-reward-models/yuyuclimb-yxyllama2-v1-2-v1_reward/tokenizer.json
yuyuclimb-yxyllama2-v1-2-v1-mkmlizer: cp /tmp/reward_cache/reward.tensors s3://guanaco-reward-models/yuyuclimb-yxyllama2-v1-2-v1_reward/reward.tensors
Job yuyuclimb-yxyllama2-v1-2-v1-mkmlizer completed after 64.1s with status: succeeded
Stopping job with name yuyuclimb-yxyllama2-v1-2-v1-mkmlizer
Pipeline stage MKMLizer completed in 69.61s
Running pipeline stage MKMLKubeTemplater
Pipeline stage MKMLKubeTemplater completed in 0.10s
Running pipeline stage ISVCDeployer
Creating inference service yuyuclimb-yxyllama2-v1-2-v1
Waiting for inference service yuyuclimb-yxyllama2-v1-2-v1 to be ready
Inference service yuyuclimb-yxyllama2-v1-2-v1 ready after 30.20477271080017s
Pipeline stage ISVCDeployer completed in 38.34s
Running pipeline stage StressChecker
Received healthy response to inference request in 1.8577823638916016s
Received healthy response to inference request in 1.1219539642333984s
Received healthy response to inference request in 1.131270408630371s
Received healthy response to inference request in 1.151160717010498s
Received healthy response to inference request in 0.8666036128997803s
5 requests
0 failed requests
5th percentile: 0.9176736831665039
10th percentile: 0.9687437534332275
20th percentile: 1.0708838939666747
30th percentile: 1.123817253112793
40th percentile: 1.127543830871582
50th percentile: 1.131270408630371
60th percentile: 1.139226531982422
70th percentile: 1.1471826553344726
80th percentile: 1.2924850463867188
90th percentile: 1.5751337051391603
95th percentile: 1.7164580345153808
99th percentile: 1.8295174980163573
mean time: 1.2257542133331298
Pipeline stage StressChecker completed in 6.82s
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
yuyuclimb-yxyllama2-v1-2_v1 status is now deployed due to DeploymentManager action
yuyuclimb-yxyllama2-v1-2_v1 status is now inactive due to auto deactivation removed underperforming models

Usage Metrics

Latency Metrics