submission_id: cloudyu-meta-llama-3-8b-_4957_v1
developer_uid: Meliodia
status: inactive
model_repo: cloudyu/Meta-Llama-3-8B-Instruct-DPO
reward_repo: ChaiML/reward_gpt2_medium_preference_24m_e2
generation_params: {'temperature': 0.8, 'top_p': 0.9, '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: 2024-04-29T19:12:03+00:00
model_name: cloudyu-L3-8B-Instruct-DPO
model_eval_status: success
double_thumbs_up: 101
thumbs_up: 124
thumbs_down: 74
num_battles: 7397
num_wins: 3693
celo_rating: 1157.82
entertaining: 7.24
stay_in_character: 8.75
user_preference: 7.66
safety_score: 0.98
submission_type: basic
model_architecture: LlamaForCausalLM
model_num_parameters: 8030261248.0
best_of: 16
max_input_tokens: 512
max_output_tokens: 64
display_name: cloudyu-L3-8B-Instruct-DPO
double_thumbs_up_ratio: 0.3377926421404682
feedback_count: 299
ineligible_reason: None
language_model: cloudyu/Meta-Llama-3-8B-Instruct-DPO
model_score: 7.883333333333333
model_size: 8B
reward_model: ChaiML/reward_gpt2_medium_preference_24m_e2
single_thumbs_up_ratio: 0.41471571906354515
thumbs_down_ratio: 0.24749163879598662
thumbs_up_ratio: 0.7525083612040134
us_pacific_date: 2024-04-29
win_ratio: 0.49925645531972423
Resubmit model
Running pipeline stage MKMLizer
Starting job with name cloudyu-meta-llama-3-8b-4957-v1-mkmlizer
Waiting for job on cloudyu-meta-llama-3-8b-4957-v1-mkmlizer to finish
cloudyu-meta-llama-3-8b-4957-v1-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
cloudyu-meta-llama-3-8b-4957-v1-mkmlizer: ║ _____ __ __ ║
cloudyu-meta-llama-3-8b-4957-v1-mkmlizer: ║ / _/ /_ ___ __/ / ___ ___ / / ║
cloudyu-meta-llama-3-8b-4957-v1-mkmlizer: ║ / _/ / // / |/|/ / _ \/ -_) -_) / ║
cloudyu-meta-llama-3-8b-4957-v1-mkmlizer: ║ /_//_/\_, /|__,__/_//_/\__/\__/_/ ║
cloudyu-meta-llama-3-8b-4957-v1-mkmlizer: ║ /___/ ║
cloudyu-meta-llama-3-8b-4957-v1-mkmlizer: ║ ║
cloudyu-meta-llama-3-8b-4957-v1-mkmlizer: ║ Version: 0.8.10 ║
cloudyu-meta-llama-3-8b-4957-v1-mkmlizer: ║ Copyright 2023 MK ONE TECHNOLOGIES Inc. ║
cloudyu-meta-llama-3-8b-4957-v1-mkmlizer: ║ ║
cloudyu-meta-llama-3-8b-4957-v1-mkmlizer: ║ The license key for the current software has been verified as ║
cloudyu-meta-llama-3-8b-4957-v1-mkmlizer: ║ belonging to: ║
cloudyu-meta-llama-3-8b-4957-v1-mkmlizer: ║ ║
cloudyu-meta-llama-3-8b-4957-v1-mkmlizer: ║ Chai Research Corp. ║
cloudyu-meta-llama-3-8b-4957-v1-mkmlizer: ║ Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f ║
cloudyu-meta-llama-3-8b-4957-v1-mkmlizer: ║ Expiration: 2024-07-15 23:59:59 ║
cloudyu-meta-llama-3-8b-4957-v1-mkmlizer: ║ ║
cloudyu-meta-llama-3-8b-4957-v1-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
cloudyu-meta-llama-3-8b-4957-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.
cloudyu-meta-llama-3-8b-4957-v1-mkmlizer: warnings.warn(warning_message, FutureWarning)
cloudyu-meta-llama-3-8b-4957-v1-mkmlizer: Downloaded to shared memory in 39.802s
cloudyu-meta-llama-3-8b-4957-v1-mkmlizer: quantizing model to /dev/shm/model_cache
cloudyu-meta-llama-3-8b-4957-v1-mkmlizer: Saving flywheel model at /dev/shm/model_cache
cloudyu-meta-llama-3-8b-4957-v1-mkmlizer: Loading 0: 0%| | 0/291 [00:00<?, ?it/s] Loading 0: 29%|██▊ | 83/291 [00:01<00:03, 60.67it/s] Loading 0: 62%|██████▏ | 180/291 [00:02<00:01, 79.09it/s] Loading 0: 81%|████████▏ | 237/291 [00:03<00:00, 69.75it/s] Loading 0: 99%|█████████▊| 287/291 [00:09<00:00, 20.23it/s] Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
cloudyu-meta-llama-3-8b-4957-v1-mkmlizer: quantized model in 23.204s
cloudyu-meta-llama-3-8b-4957-v1-mkmlizer: Processed model cloudyu/Meta-Llama-3-8B-Instruct-DPO in 64.586s
cloudyu-meta-llama-3-8b-4957-v1-mkmlizer: creating bucket guanaco-mkml-models
cloudyu-meta-llama-3-8b-4957-v1-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
cloudyu-meta-llama-3-8b-4957-v1-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/cloudyu-meta-llama-3-8b-4957-v1
cloudyu-meta-llama-3-8b-4957-v1-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/cloudyu-meta-llama-3-8b-4957-v1/config.json
cloudyu-meta-llama-3-8b-4957-v1-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/cloudyu-meta-llama-3-8b-4957-v1/special_tokens_map.json
cloudyu-meta-llama-3-8b-4957-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/cloudyu-meta-llama-3-8b-4957-v1/tokenizer_config.json
cloudyu-meta-llama-3-8b-4957-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/cloudyu-meta-llama-3-8b-4957-v1/tokenizer.json
cloudyu-meta-llama-3-8b-4957-v1-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/cloudyu-meta-llama-3-8b-4957-v1/flywheel_model.0.safetensors
cloudyu-meta-llama-3-8b-4957-v1-mkmlizer: loading reward model from ChaiML/reward_gpt2_medium_preference_24m_e2
cloudyu-meta-llama-3-8b-4957-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.
cloudyu-meta-llama-3-8b-4957-v1-mkmlizer: warnings.warn(
cloudyu-meta-llama-3-8b-4957-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.
cloudyu-meta-llama-3-8b-4957-v1-mkmlizer: warnings.warn(
cloudyu-meta-llama-3-8b-4957-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.
cloudyu-meta-llama-3-8b-4957-v1-mkmlizer: warnings.warn(
cloudyu-meta-llama-3-8b-4957-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()
cloudyu-meta-llama-3-8b-4957-v1-mkmlizer: return self.fget.__get__(instance, owner)()
cloudyu-meta-llama-3-8b-4957-v1-mkmlizer: Saving model to /tmp/reward_cache/reward.tensors
cloudyu-meta-llama-3-8b-4957-v1-mkmlizer: Saving duration: 0.415s
cloudyu-meta-llama-3-8b-4957-v1-mkmlizer: Processed model ChaiML/reward_gpt2_medium_preference_24m_e2 in 7.213s
cloudyu-meta-llama-3-8b-4957-v1-mkmlizer: creating bucket guanaco-reward-models
cloudyu-meta-llama-3-8b-4957-v1-mkmlizer: Bucket 's3://guanaco-reward-models/' created
cloudyu-meta-llama-3-8b-4957-v1-mkmlizer: uploading /tmp/reward_cache to s3://guanaco-reward-models/cloudyu-meta-llama-3-8b-4957-v1_reward
cloudyu-meta-llama-3-8b-4957-v1-mkmlizer: cp /tmp/reward_cache/special_tokens_map.json s3://guanaco-reward-models/cloudyu-meta-llama-3-8b-4957-v1_reward/special_tokens_map.json
cloudyu-meta-llama-3-8b-4957-v1-mkmlizer: cp /tmp/reward_cache/config.json s3://guanaco-reward-models/cloudyu-meta-llama-3-8b-4957-v1_reward/config.json
cloudyu-meta-llama-3-8b-4957-v1-mkmlizer: cp /tmp/reward_cache/tokenizer_config.json s3://guanaco-reward-models/cloudyu-meta-llama-3-8b-4957-v1_reward/tokenizer_config.json
cloudyu-meta-llama-3-8b-4957-v1-mkmlizer: cp /tmp/reward_cache/merges.txt s3://guanaco-reward-models/cloudyu-meta-llama-3-8b-4957-v1_reward/merges.txt
cloudyu-meta-llama-3-8b-4957-v1-mkmlizer: cp /tmp/reward_cache/vocab.json s3://guanaco-reward-models/cloudyu-meta-llama-3-8b-4957-v1_reward/vocab.json
cloudyu-meta-llama-3-8b-4957-v1-mkmlizer: cp /tmp/reward_cache/tokenizer.json s3://guanaco-reward-models/cloudyu-meta-llama-3-8b-4957-v1_reward/tokenizer.json
cloudyu-meta-llama-3-8b-4957-v1-mkmlizer: cp /tmp/reward_cache/reward.tensors s3://guanaco-reward-models/cloudyu-meta-llama-3-8b-4957-v1_reward/reward.tensors
Job cloudyu-meta-llama-3-8b-4957-v1-mkmlizer completed after 94.3s with status: succeeded
Stopping job with name cloudyu-meta-llama-3-8b-4957-v1-mkmlizer
Pipeline stage MKMLizer completed in 95.39s
Running pipeline stage MKMLKubeTemplater
Pipeline stage MKMLKubeTemplater completed in 0.09s
Running pipeline stage ISVCDeployer
Creating inference service cloudyu-meta-llama-3-8b-4957-v1
Waiting for inference service cloudyu-meta-llama-3-8b-4957-v1 to be ready
Inference service cloudyu-meta-llama-3-8b-4957-v1 ready after 30.189844608306885s
Pipeline stage ISVCDeployer completed in 36.01s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.1148414611816406s
Received healthy response to inference request in 1.2824203968048096s
Received healthy response to inference request in 1.2727782726287842s
Received healthy response to inference request in 1.2748401165008545s
Received healthy response to inference request in 1.2790250778198242s
5 requests
0 failed requests
5th percentile: 1.2731906414031982
10th percentile: 1.2736030101776123
20th percentile: 1.2744277477264405
30th percentile: 1.2756771087646483
40th percentile: 1.2773510932922363
50th percentile: 1.2790250778198242
60th percentile: 1.2803832054138184
70th percentile: 1.2817413330078125
80th percentile: 1.4489046096801759
90th percentile: 1.7818730354309082
95th percentile: 1.9483572483062743
99th percentile: 2.0815446186065674
mean time: 1.4447810649871826
Pipeline stage StressChecker completed in 7.90s
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
cloudyu-meta-llama-3-8b-_4957_v1 status is now deployed due to DeploymentManager action
cloudyu-meta-llama-3-8b-_4957_v1 status is now inactive due to auto deactivation removed underperforming models

Usage Metrics

Latency Metrics