submission_id: hastagaras-llama3-8b-dar_7411_v4
developer_uid: Hastagaras
status: torndown
model_repo: Hastagaras/llama3-8b-darktune-KTO
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': "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nPlay the role of {bot_name}. {bot_name}'s persona: {memory}<|eot_id|>", 'prompt_template': '<|start_header_id|>assistant<|end_header_id|>\n\n{bot_name}: {prompt}<|eot_id|>', 'bot_template': '<|start_header_id|>assistant<|end_header_id|>\n\n{bot_name}: {message}<|eot_id|>', 'user_template': '<|start_header_id|>user<|end_header_id|>\n\n{user_name}: {message}<|eot_id|>', 'response_template': '<|start_header_id|>assistant<|end_header_id|>\n\n{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: 2024-05-11T03:57:12+00:00
model_name: drktune-retry-kto
model_eval_status: pending
model_group: Hastagaras/llama3-8b-dar
num_battles: 20547
num_wins: 10425
celo_rating: 1178.88
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: drktune-retry-kto
ineligible_reason: propriety_total_count < 800
language_model: Hastagaras/llama3-8b-darktune-KTO
model_size: 8B
reward_model: ChaiML/reward_gpt2_medium_preference_24m_e2
us_pacific_date: 2024-05-10
win_ratio: 0.507373339173602
preference_data_url: None
Resubmit model
Running pipeline stage MKMLizer
Starting job with name hastagaras-llama3-8b-dar-7411-v4-mkmlizer
Waiting for job on hastagaras-llama3-8b-dar-7411-v4-mkmlizer to finish
Stopping job with name hastagaras-llama3-8b-dar-7411-v4-mkmlizer
%s, retrying in %s seconds...
Starting job with name hastagaras-llama3-8b-dar-7411-v4-mkmlizer
Waiting for job on hastagaras-llama3-8b-dar-7411-v4-mkmlizer to finish
Stopping job with name hastagaras-llama3-8b-dar-7411-v4-mkmlizer
%s, retrying in %s seconds...
Starting job with name hastagaras-llama3-8b-dar-7411-v4-mkmlizer
Waiting for job on hastagaras-llama3-8b-dar-7411-v4-mkmlizer to finish
hastagaras-llama3-8b-dar-7411-v4-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
hastagaras-llama3-8b-dar-7411-v4-mkmlizer: ║ _____ __ __ ║
hastagaras-llama3-8b-dar-7411-v4-mkmlizer: ║ / _/ /_ ___ __/ / ___ ___ / / ║
hastagaras-llama3-8b-dar-7411-v4-mkmlizer: ║ / _/ / // / |/|/ / _ \/ -_) -_) / ║
hastagaras-llama3-8b-dar-7411-v4-mkmlizer: ║ /_//_/\_, /|__,__/_//_/\__/\__/_/ ║
hastagaras-llama3-8b-dar-7411-v4-mkmlizer: ║ /___/ ║
hastagaras-llama3-8b-dar-7411-v4-mkmlizer: ║ ║
hastagaras-llama3-8b-dar-7411-v4-mkmlizer: ║ Version: 0.8.10 ║
hastagaras-llama3-8b-dar-7411-v4-mkmlizer: ║ Copyright 2023 MK ONE TECHNOLOGIES Inc. ║
hastagaras-llama3-8b-dar-7411-v4-mkmlizer: ║ ║
hastagaras-llama3-8b-dar-7411-v4-mkmlizer: ║ The license key for the current software has been verified as ║
hastagaras-llama3-8b-dar-7411-v4-mkmlizer: ║ belonging to: ║
hastagaras-llama3-8b-dar-7411-v4-mkmlizer: ║ ║
hastagaras-llama3-8b-dar-7411-v4-mkmlizer: ║ Chai Research Corp. ║
hastagaras-llama3-8b-dar-7411-v4-mkmlizer: ║ Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f ║
hastagaras-llama3-8b-dar-7411-v4-mkmlizer: ║ Expiration: 2024-07-15 23:59:59 ║
hastagaras-llama3-8b-dar-7411-v4-mkmlizer: ║ ║
hastagaras-llama3-8b-dar-7411-v4-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
hastagaras-llama3-8b-dar-7411-v4-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-llama3-8b-dar-7411-v4-mkmlizer: warnings.warn(warning_message, FutureWarning)
hastagaras-llama3-8b-dar-7411-v4-mkmlizer: Downloaded to shared memory in 10.719s
hastagaras-llama3-8b-dar-7411-v4-mkmlizer: quantizing model to /dev/shm/model_cache
hastagaras-llama3-8b-dar-7411-v4-mkmlizer: Saving flywheel model at /dev/shm/model_cache
hastagaras-llama3-8b-dar-7411-v4-mkmlizer: quantized model in 17.100s
hastagaras-llama3-8b-dar-7411-v4-mkmlizer: Processed model Hastagaras/llama3-8b-darktune-KTO in 28.857s
hastagaras-llama3-8b-dar-7411-v4-mkmlizer: creating bucket guanaco-mkml-models
hastagaras-llama3-8b-dar-7411-v4-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
hastagaras-llama3-8b-dar-7411-v4-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/hastagaras-llama3-8b-dar-7411-v4
hastagaras-llama3-8b-dar-7411-v4-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/hastagaras-llama3-8b-dar-7411-v4/config.json
hastagaras-llama3-8b-dar-7411-v4-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/hastagaras-llama3-8b-dar-7411-v4/tokenizer_config.json
hastagaras-llama3-8b-dar-7411-v4-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/hastagaras-llama3-8b-dar-7411-v4/special_tokens_map.json
hastagaras-llama3-8b-dar-7411-v4-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/hastagaras-llama3-8b-dar-7411-v4/tokenizer.json
hastagaras-llama3-8b-dar-7411-v4-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/hastagaras-llama3-8b-dar-7411-v4/flywheel_model.0.safetensors
hastagaras-llama3-8b-dar-7411-v4-mkmlizer: loading reward model from ChaiML/reward_gpt2_medium_preference_24m_e2
hastagaras-llama3-8b-dar-7411-v4-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-llama3-8b-dar-7411-v4-mkmlizer: warnings.warn(
hastagaras-llama3-8b-dar-7411-v4-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-llama3-8b-dar-7411-v4-mkmlizer: warnings.warn(
hastagaras-llama3-8b-dar-7411-v4-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.
hastagaras-llama3-8b-dar-7411-v4-mkmlizer: warnings.warn(
hastagaras-llama3-8b-dar-7411-v4-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-llama3-8b-dar-7411-v4-mkmlizer: return self.fget.__get__(instance, owner)()
hastagaras-llama3-8b-dar-7411-v4-mkmlizer: Saving model to /tmp/reward_cache/reward.tensors
hastagaras-llama3-8b-dar-7411-v4-mkmlizer: Saving duration: 0.232s
hastagaras-llama3-8b-dar-7411-v4-mkmlizer: Processed model ChaiML/reward_gpt2_medium_preference_24m_e2 in 2.991s
Job hastagaras-llama3-8b-dar-7411-v4-mkmlizer completed after 93.91s with status: succeeded
Stopping job with name hastagaras-llama3-8b-dar-7411-v4-mkmlizer
Pipeline stage MKMLizer completed in 95.34s
Running pipeline stage MKMLKubeTemplater
Pipeline stage MKMLKubeTemplater completed in 0.09s
Running pipeline stage ISVCDeployer
Creating inference service hastagaras-llama3-8b-dar-7411-v4
Waiting for inference service hastagaras-llama3-8b-dar-7411-v4 to be ready
Inference service hastagaras-llama3-8b-dar-7411-v4 ready after 40.2229437828064s
Pipeline stage ISVCDeployer completed in 45.77s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.211894989013672s
Received healthy response to inference request in 1.3499815464019775s
Received healthy response to inference request in 1.3084509372711182s
Received healthy response to inference request in 1.3635764122009277s
Received healthy response to inference request in 1.325953483581543s
5 requests
0 failed requests
5th percentile: 1.3119514465332032
10th percentile: 1.3154519557952882
20th percentile: 1.322452974319458
30th percentile: 1.33075909614563
40th percentile: 1.3403703212738036
50th percentile: 1.3499815464019775
60th percentile: 1.3554194927215577
70th percentile: 1.3608574390411377
80th percentile: 1.5332401275634768
90th percentile: 1.8725675582885744
95th percentile: 2.042231273651123
99th percentile: 2.177962245941162
mean time: 1.5119714736938477
Pipeline stage StressChecker completed in 8.21s
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
hastagaras-llama3-8b-dar_7411_v4 status is now deployed due to DeploymentManager action
hastagaras-llama3-8b-dar_7411_v4 status is now inactive due to auto deactivation removed underperforming models
admin requested tearing down of hastagaras-llama3-8b-dar_7411_v4
Running pipeline stage ISVCDeleter
Checking if service hastagaras-llama3-8b-dar-7411-v4 is running
Tearing down inference service hastagaras-llama3-8b-dar-7411-v4
Toredown service hastagaras-llama3-8b-dar-7411-v4
Pipeline stage ISVCDeleter completed in 6.25s
Running pipeline stage MKMLModelDeleter
Cleaning model data from S3
Cleaning model data from model cache
Deleting key hastagaras-llama3-8b-dar-7411-v4/config.json from bucket guanaco-mkml-models
Deleting key hastagaras-llama3-8b-dar-7411-v4/flywheel_model.0.safetensors from bucket guanaco-mkml-models
Deleting key hastagaras-llama3-8b-dar-7411-v4/special_tokens_map.json from bucket guanaco-mkml-models
Deleting key hastagaras-llama3-8b-dar-7411-v4/tokenizer.json from bucket guanaco-mkml-models
Deleting key hastagaras-llama3-8b-dar-7411-v4/tokenizer_config.json from bucket guanaco-mkml-models
Cleaning model data from model cache
Deleting key hastagaras-llama3-8b-dar-7411-v4_reward/config.json from bucket guanaco-reward-models
Deleting key hastagaras-llama3-8b-dar-7411-v4_reward/merges.txt from bucket guanaco-reward-models
Deleting key hastagaras-llama3-8b-dar-7411-v4_reward/reward.tensors from bucket guanaco-reward-models
Deleting key hastagaras-llama3-8b-dar-7411-v4_reward/special_tokens_map.json from bucket guanaco-reward-models
Deleting key hastagaras-llama3-8b-dar-7411-v4_reward/tokenizer.json from bucket guanaco-reward-models
Deleting key hastagaras-llama3-8b-dar-7411-v4_reward/tokenizer_config.json from bucket guanaco-reward-models
Deleting key hastagaras-llama3-8b-dar-7411-v4_reward/vocab.json from bucket guanaco-reward-models
Pipeline stage MKMLModelDeleter completed in 1.79s
hastagaras-llama3-8b-dar_7411_v4 status is now torndown due to DeploymentManager action

Usage Metrics

Latency Metrics