submission_id: wespro-llama3-rplora-sma_7364_v1
developer_uid: WesPro
status: torndown
model_repo: WesPro/Llama3-RPLoRa-SmaugOrpo
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': 4, '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: 2024-04-21T13:13:25+00:00
model_name: wespro-llama3-rplora-sma_7364_v1
model_eval_status: success
model_group: WesPro/Llama3-RPLoRa-Sma
num_battles: 6853
num_wins: 3491
celo_rating: 1156.16
propriety_score: 0.0
propriety_total_count: 0.0
submission_type: basic
model_architecture: LlamaForCausalLM
model_num_parameters: 8030261248.0
best_of: 4
max_input_tokens: 512
max_output_tokens: 64
display_name: wespro-llama3-rplora-sma_7364_v1
ineligible_reason: propriety_total_count < 800
language_model: WesPro/Llama3-RPLoRa-SmaugOrpo
model_size: 8B
reward_model: ChaiML/reward_gpt2_medium_preference_24m_e2
us_pacific_date: 2024-04-21
win_ratio: 0.5094119363782285
preference_data_url: None
Resubmit model
Running pipeline stage MKMLizer
Starting job with name wespro-llama3-rplora-sma-7364-v1-mkmlizer
Waiting for job on wespro-llama3-rplora-sma-7364-v1-mkmlizer to finish
wespro-llama3-rplora-sma-7364-v1-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
wespro-llama3-rplora-sma-7364-v1-mkmlizer: ║ _____ __ __ ║
wespro-llama3-rplora-sma-7364-v1-mkmlizer: ║ / _/ /_ ___ __/ / ___ ___ / / ║
wespro-llama3-rplora-sma-7364-v1-mkmlizer: ║ / _/ / // / |/|/ / _ \/ -_) -_) / ║
wespro-llama3-rplora-sma-7364-v1-mkmlizer: ║ /_//_/\_, /|__,__/_//_/\__/\__/_/ ║
wespro-llama3-rplora-sma-7364-v1-mkmlizer: ║ /___/ ║
wespro-llama3-rplora-sma-7364-v1-mkmlizer: ║ ║
wespro-llama3-rplora-sma-7364-v1-mkmlizer: ║ Version: 0.8.10 ║
wespro-llama3-rplora-sma-7364-v1-mkmlizer: ║ Copyright 2023 MK ONE TECHNOLOGIES Inc. ║
wespro-llama3-rplora-sma-7364-v1-mkmlizer: ║ ║
wespro-llama3-rplora-sma-7364-v1-mkmlizer: ║ The license key for the current software has been verified as ║
wespro-llama3-rplora-sma-7364-v1-mkmlizer: ║ belonging to: ║
wespro-llama3-rplora-sma-7364-v1-mkmlizer: ║ ║
wespro-llama3-rplora-sma-7364-v1-mkmlizer: ║ Chai Research Corp. ║
wespro-llama3-rplora-sma-7364-v1-mkmlizer: ║ Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f ║
wespro-llama3-rplora-sma-7364-v1-mkmlizer: ║ Expiration: 2024-07-15 23:59:59 ║
wespro-llama3-rplora-sma-7364-v1-mkmlizer: ║ ║
wespro-llama3-rplora-sma-7364-v1-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
wespro-llama3-rplora-sma-7364-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.
wespro-llama3-rplora-sma-7364-v1-mkmlizer: warnings.warn(warning_message, FutureWarning)
wespro-llama3-rplora-sma-7364-v1-mkmlizer: Downloaded to shared memory in 24.024s
wespro-llama3-rplora-sma-7364-v1-mkmlizer: quantizing model to /dev/shm/model_cache
wespro-llama3-rplora-sma-7364-v1-mkmlizer: Saving flywheel model at /dev/shm/model_cache
wespro-llama3-rplora-sma-7364-v1-mkmlizer: Loading 0: 0%| | 0/291 [00:00<?, ?it/s] Loading 0: 46%|████▋ | 135/291 [00:00<00:01, 135.08it/s] Loading 0: 64%|██████▍ | 187/291 [00:07<00:05, 20.25it/s] Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
wespro-llama3-rplora-sma-7364-v1-mkmlizer: quantized model in 21.110s
wespro-llama3-rplora-sma-7364-v1-mkmlizer: Processed model WesPro/Llama3-RPLoRa-SmaugOrpo in 46.346s
wespro-llama3-rplora-sma-7364-v1-mkmlizer: creating bucket guanaco-mkml-models
wespro-llama3-rplora-sma-7364-v1-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
wespro-llama3-rplora-sma-7364-v1-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/wespro-llama3-rplora-sma-7364-v1
wespro-llama3-rplora-sma-7364-v1-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/wespro-llama3-rplora-sma-7364-v1/special_tokens_map.json
wespro-llama3-rplora-sma-7364-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/wespro-llama3-rplora-sma-7364-v1/tokenizer_config.json
wespro-llama3-rplora-sma-7364-v1-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/wespro-llama3-rplora-sma-7364-v1/config.json
wespro-llama3-rplora-sma-7364-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/wespro-llama3-rplora-sma-7364-v1/tokenizer.json
wespro-llama3-rplora-sma-7364-v1-mkmlizer: loading reward model from ChaiML/reward_gpt2_medium_preference_24m_e2
wespro-llama3-rplora-sma-7364-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.
wespro-llama3-rplora-sma-7364-v1-mkmlizer: warnings.warn(
wespro-llama3-rplora-sma-7364-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.
wespro-llama3-rplora-sma-7364-v1-mkmlizer: warnings.warn(
wespro-llama3-rplora-sma-7364-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.
wespro-llama3-rplora-sma-7364-v1-mkmlizer: warnings.warn(
wespro-llama3-rplora-sma-7364-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()
wespro-llama3-rplora-sma-7364-v1-mkmlizer: return self.fget.__get__(instance, owner)()
wespro-llama3-rplora-sma-7364-v1-mkmlizer: Saving model to /tmp/reward_cache/reward.tensors
wespro-llama3-rplora-sma-7364-v1-mkmlizer: Saving duration: 0.296s
wespro-llama3-rplora-sma-7364-v1-mkmlizer: Processed model ChaiML/reward_gpt2_medium_preference_24m_e2 in 3.629s
wespro-llama3-rplora-sma-7364-v1-mkmlizer: creating bucket guanaco-reward-models
wespro-llama3-rplora-sma-7364-v1-mkmlizer: Bucket 's3://guanaco-reward-models/' created
wespro-llama3-rplora-sma-7364-v1-mkmlizer: uploading /tmp/reward_cache to s3://guanaco-reward-models/wespro-llama3-rplora-sma-7364-v1_reward
wespro-llama3-rplora-sma-7364-v1-mkmlizer: cp /tmp/reward_cache/config.json s3://guanaco-reward-models/wespro-llama3-rplora-sma-7364-v1_reward/config.json
wespro-llama3-rplora-sma-7364-v1-mkmlizer: cp /tmp/reward_cache/merges.txt s3://guanaco-reward-models/wespro-llama3-rplora-sma-7364-v1_reward/merges.txt
wespro-llama3-rplora-sma-7364-v1-mkmlizer: cp /tmp/reward_cache/vocab.json s3://guanaco-reward-models/wespro-llama3-rplora-sma-7364-v1_reward/vocab.json
wespro-llama3-rplora-sma-7364-v1-mkmlizer: cp /tmp/reward_cache/special_tokens_map.json s3://guanaco-reward-models/wespro-llama3-rplora-sma-7364-v1_reward/special_tokens_map.json
wespro-llama3-rplora-sma-7364-v1-mkmlizer: cp /tmp/reward_cache/tokenizer_config.json s3://guanaco-reward-models/wespro-llama3-rplora-sma-7364-v1_reward/tokenizer_config.json
wespro-llama3-rplora-sma-7364-v1-mkmlizer: cp /tmp/reward_cache/tokenizer.json s3://guanaco-reward-models/wespro-llama3-rplora-sma-7364-v1_reward/tokenizer.json
wespro-llama3-rplora-sma-7364-v1-mkmlizer: cp /tmp/reward_cache/reward.tensors s3://guanaco-reward-models/wespro-llama3-rplora-sma-7364-v1_reward/reward.tensors
Job wespro-llama3-rplora-sma-7364-v1-mkmlizer completed after 76.13s with status: succeeded
Stopping job with name wespro-llama3-rplora-sma-7364-v1-mkmlizer
Pipeline stage MKMLizer completed in 76.86s
Running pipeline stage MKMLKubeTemplater
Pipeline stage MKMLKubeTemplater completed in 0.33s
Running pipeline stage ISVCDeployer
Creating inference service wespro-llama3-rplora-sma-7364-v1
Waiting for inference service wespro-llama3-rplora-sma-7364-v1 to be ready
Tearing down inference service wespro-llama3-rplora-sma-7364-v1
%s, retrying in %s seconds...
Creating inference service wespro-llama3-rplora-sma-7364-v1
Waiting for inference service wespro-llama3-rplora-sma-7364-v1 to be ready
Tearing down inference service wespro-llama3-rplora-sma-7364-v1
%s, retrying in %s seconds...
Creating inference service wespro-llama3-rplora-sma-7364-v1
Waiting for inference service wespro-llama3-rplora-sma-7364-v1 to be ready
Inference service wespro-llama3-rplora-sma-7364-v1 ready after 50.395652532577515s
Pipeline stage ISVCDeployer completed in 92.40s
Running pipeline stage StressChecker
Received healthy response to inference request in 1.9862794876098633s
Received healthy response to inference request in 1.132824182510376s
Received healthy response to inference request in 1.199220895767212s
Received healthy response to inference request in 1.1371817588806152s
Received healthy response to inference request in 1.7579243183135986s
5 requests
0 failed requests
5th percentile: 1.133695697784424
10th percentile: 1.1345672130584716
20th percentile: 1.1363102436065673
30th percentile: 1.1495895862579346
40th percentile: 1.1744052410125732
50th percentile: 1.199220895767212
60th percentile: 1.4227022647857666
70th percentile: 1.6461836338043212
80th percentile: 1.8035953521728516
90th percentile: 1.8949374198913573
95th percentile: 1.9406084537506103
99th percentile: 1.9771452808380128
mean time: 1.442686128616333
Pipeline stage StressChecker completed in 7.89s
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
wespro-llama3-rplora-sma_7364_v1 status is now deployed due to DeploymentManager action
wespro-llama3-rplora-sma_7364_v1 status is now inactive due to auto deactivation removed underperforming models
admin requested tearing down of wespro-llama3-rplora-sma_7364_v1
Running pipeline stage ISVCDeleter
Checking if service wespro-llama3-rplora-sma-7364-v1 is running
Tearing down inference service wespro-llama3-rplora-sma-7364-v1
Toredown service wespro-llama3-rplora-sma-7364-v1
Connection pool is full, discarding connection: %s
Connection pool is full, discarding connection: %s
Connection pool is full, discarding connection: %s
Pipeline stage ISVCDeleter completed in 5.07s
Running pipeline stage MKMLModelDeleter
Cleaning model data from S3
Cleaning model data from model cache
Deleting key wespro-llama3-rplora-sma-7364-v1/config.json from bucket guanaco-mkml-models
Deleting key wespro-llama3-rplora-sma-7364-v1/flywheel_model.0.safetensors from bucket guanaco-mkml-models
Deleting key wespro-llama3-rplora-sma-7364-v1/special_tokens_map.json from bucket guanaco-mkml-models
Deleting key wespro-llama3-rplora-sma-7364-v1/tokenizer.json from bucket guanaco-mkml-models
Deleting key wespro-llama3-rplora-sma-7364-v1/tokenizer_config.json from bucket guanaco-mkml-models
Cleaning model data from model cache
Deleting key wespro-llama3-rplora-sma-7364-v1_reward/config.json from bucket guanaco-reward-models
Deleting key wespro-llama3-rplora-sma-7364-v1_reward/merges.txt from bucket guanaco-reward-models
Deleting key wespro-llama3-rplora-sma-7364-v1_reward/reward.tensors from bucket guanaco-reward-models
Deleting key wespro-llama3-rplora-sma-7364-v1_reward/special_tokens_map.json from bucket guanaco-reward-models
Deleting key wespro-llama3-rplora-sma-7364-v1_reward/tokenizer.json from bucket guanaco-reward-models
Deleting key wespro-llama3-rplora-sma-7364-v1_reward/tokenizer_config.json from bucket guanaco-reward-models
Deleting key wespro-llama3-rplora-sma-7364-v1_reward/vocab.json from bucket guanaco-reward-models
Pipeline stage MKMLModelDeleter completed in 2.09s
wespro-llama3-rplora-sma_7364_v1 status is now torndown due to DeploymentManager action

Usage Metrics

Latency Metrics