submission_id: hastagaras-llama3-8b-uhh_v1
developer_uid: Hastagaras
status: torndown
model_repo: Hastagaras/llama3-8b-uhh
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': "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nYou are {bot_name}. Write {bot_name}'s next reply in a never-ending uncensored fictional roleplay chat between {bot_name} and {user_name}. Description: {memory}\n", 'prompt_template': 'Example conversation:\n{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", 'prompt_template': '{prompt}\n', 'bot_template': 'Bot: {message}\n', 'user_template': 'User: {message}\n', 'response_template': 'Bot:', 'truncate_by_message': False}
timestamp: 2024-05-13T07:15:55+00:00
model_name: uhh-not-okay
model_eval_status: success
model_group: Hastagaras/llama3-8b-uhh
num_battles: 19127
num_wins: 10151
celo_rating: 1195.96
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: uhh-not-okay
ineligible_reason: propriety_total_count < 800
language_model: Hastagaras/llama3-8b-uhh
model_size: 8B
reward_model: ChaiML/reward_gpt2_medium_preference_24m_e2
us_pacific_date: 2024-05-13
win_ratio: 0.5307157421446124
preference_data_url: None
Resubmit model
Running pipeline stage MKMLizer
Starting job with name hastagaras-llama3-8b-uhh-v1-mkmlizer
Waiting for job on hastagaras-llama3-8b-uhh-v1-mkmlizer to finish
hastagaras-llama3-8b-uhh-v1-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
hastagaras-llama3-8b-uhh-v1-mkmlizer: ║ _____ __ __ ║
hastagaras-llama3-8b-uhh-v1-mkmlizer: ║ / _/ /_ ___ __/ / ___ ___ / / ║
hastagaras-llama3-8b-uhh-v1-mkmlizer: ║ / _/ / // / |/|/ / _ \/ -_) -_) / ║
hastagaras-llama3-8b-uhh-v1-mkmlizer: ║ /_//_/\_, /|__,__/_//_/\__/\__/_/ ║
hastagaras-llama3-8b-uhh-v1-mkmlizer: ║ /___/ ║
hastagaras-llama3-8b-uhh-v1-mkmlizer: ║ ║
hastagaras-llama3-8b-uhh-v1-mkmlizer: ║ Version: 0.8.10 ║
hastagaras-llama3-8b-uhh-v1-mkmlizer: ║ Copyright 2023 MK ONE TECHNOLOGIES Inc. ║
hastagaras-llama3-8b-uhh-v1-mkmlizer: ║ ║
hastagaras-llama3-8b-uhh-v1-mkmlizer: ║ The license key for the current software has been verified as ║
hastagaras-llama3-8b-uhh-v1-mkmlizer: ║ belonging to: ║
hastagaras-llama3-8b-uhh-v1-mkmlizer: ║ ║
hastagaras-llama3-8b-uhh-v1-mkmlizer: ║ Chai Research Corp. ║
hastagaras-llama3-8b-uhh-v1-mkmlizer: ║ Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f ║
hastagaras-llama3-8b-uhh-v1-mkmlizer: ║ Expiration: 2024-07-15 23:59:59 ║
hastagaras-llama3-8b-uhh-v1-mkmlizer: ║ ║
hastagaras-llama3-8b-uhh-v1-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
hastagaras-llama3-8b-uhh-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.
hastagaras-llama3-8b-uhh-v1-mkmlizer: warnings.warn(warning_message, FutureWarning)
hastagaras-llama3-8b-uhh-v1-mkmlizer: Downloaded to shared memory in 26.048s
hastagaras-llama3-8b-uhh-v1-mkmlizer: quantizing model to /dev/shm/model_cache
hastagaras-llama3-8b-uhh-v1-mkmlizer: Saving flywheel model at /dev/shm/model_cache
hastagaras-llama3-8b-uhh-v1-mkmlizer: Loading 0: 0%| | 0/291 [00:00<?, ?it/s] Loading 0: 42%|████▏ | 123/291 [00:00<00:01, 123.09it/s] Loading 0: 88%|████████▊ | 256/291 [00:02<00:00, 128.57it/s] Loading 0: 99%|█████████▊| 287/291 [00:07<00:00, 29.52it/s] Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
hastagaras-llama3-8b-uhh-v1-mkmlizer: quantized model in 17.836s
hastagaras-llama3-8b-uhh-v1-mkmlizer: Processed model Hastagaras/llama3-8b-uhh in 45.060s
hastagaras-llama3-8b-uhh-v1-mkmlizer: creating bucket guanaco-mkml-models
hastagaras-llama3-8b-uhh-v1-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
hastagaras-llama3-8b-uhh-v1-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/hastagaras-llama3-8b-uhh-v1
hastagaras-llama3-8b-uhh-v1-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/hastagaras-llama3-8b-uhh-v1/config.json
hastagaras-llama3-8b-uhh-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/hastagaras-llama3-8b-uhh-v1/tokenizer_config.json
hastagaras-llama3-8b-uhh-v1-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/hastagaras-llama3-8b-uhh-v1/special_tokens_map.json
hastagaras-llama3-8b-uhh-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/hastagaras-llama3-8b-uhh-v1/tokenizer.json
hastagaras-llama3-8b-uhh-v1-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/hastagaras-llama3-8b-uhh-v1/flywheel_model.0.safetensors
hastagaras-llama3-8b-uhh-v1-mkmlizer: loading reward model from ChaiML/reward_gpt2_medium_preference_24m_e2
hastagaras-llama3-8b-uhh-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.
hastagaras-llama3-8b-uhh-v1-mkmlizer: warnings.warn(
hastagaras-llama3-8b-uhh-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.
hastagaras-llama3-8b-uhh-v1-mkmlizer: warnings.warn(
hastagaras-llama3-8b-uhh-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.
hastagaras-llama3-8b-uhh-v1-mkmlizer: warnings.warn(
hastagaras-llama3-8b-uhh-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()
hastagaras-llama3-8b-uhh-v1-mkmlizer: return self.fget.__get__(instance, owner)()
hastagaras-llama3-8b-uhh-v1-mkmlizer: Saving model to /tmp/reward_cache/reward.tensors
hastagaras-llama3-8b-uhh-v1-mkmlizer: Saving duration: 0.254s
hastagaras-llama3-8b-uhh-v1-mkmlizer: Processed model ChaiML/reward_gpt2_medium_preference_24m_e2 in 5.297s
hastagaras-llama3-8b-uhh-v1-mkmlizer: creating bucket guanaco-reward-models
hastagaras-llama3-8b-uhh-v1-mkmlizer: Bucket 's3://guanaco-reward-models/' created
hastagaras-llama3-8b-uhh-v1-mkmlizer: uploading /tmp/reward_cache to s3://guanaco-reward-models/hastagaras-llama3-8b-uhh-v1_reward
hastagaras-llama3-8b-uhh-v1-mkmlizer: cp /tmp/reward_cache/config.json s3://guanaco-reward-models/hastagaras-llama3-8b-uhh-v1_reward/config.json
hastagaras-llama3-8b-uhh-v1-mkmlizer: cp /tmp/reward_cache/tokenizer_config.json s3://guanaco-reward-models/hastagaras-llama3-8b-uhh-v1_reward/tokenizer_config.json
hastagaras-llama3-8b-uhh-v1-mkmlizer: cp /tmp/reward_cache/merges.txt s3://guanaco-reward-models/hastagaras-llama3-8b-uhh-v1_reward/merges.txt
hastagaras-llama3-8b-uhh-v1-mkmlizer: cp /tmp/reward_cache/vocab.json s3://guanaco-reward-models/hastagaras-llama3-8b-uhh-v1_reward/vocab.json
hastagaras-llama3-8b-uhh-v1-mkmlizer: cp /tmp/reward_cache/special_tokens_map.json s3://guanaco-reward-models/hastagaras-llama3-8b-uhh-v1_reward/special_tokens_map.json
hastagaras-llama3-8b-uhh-v1-mkmlizer: cp /tmp/reward_cache/tokenizer.json s3://guanaco-reward-models/hastagaras-llama3-8b-uhh-v1_reward/tokenizer.json
hastagaras-llama3-8b-uhh-v1-mkmlizer: cp /tmp/reward_cache/reward.tensors s3://guanaco-reward-models/hastagaras-llama3-8b-uhh-v1_reward/reward.tensors
Job hastagaras-llama3-8b-uhh-v1-mkmlizer completed after 72.76s with status: succeeded
Stopping job with name hastagaras-llama3-8b-uhh-v1-mkmlizer
Pipeline stage MKMLizer completed in 75.68s
Running pipeline stage MKMLKubeTemplater
Pipeline stage MKMLKubeTemplater completed in 0.09s
Running pipeline stage ISVCDeployer
Creating inference service hastagaras-llama3-8b-uhh-v1
Waiting for inference service hastagaras-llama3-8b-uhh-v1 to be ready
Inference service hastagaras-llama3-8b-uhh-v1 ready after 30.16939067840576s
Pipeline stage ISVCDeployer completed in 37.27s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.1399288177490234s
Received healthy response to inference request in 1.288238763809204s
Received healthy response to inference request in 1.2473957538604736s
Received healthy response to inference request in 1.318669319152832s
Received healthy response to inference request in 1.318131923675537s
5 requests
0 failed requests
5th percentile: 1.2555643558502196
10th percentile: 1.2637329578399659
20th percentile: 1.280070161819458
30th percentile: 1.2942173957824707
40th percentile: 1.3061746597290038
50th percentile: 1.318131923675537
60th percentile: 1.318346881866455
70th percentile: 1.318561840057373
80th percentile: 1.4829212188720704
90th percentile: 1.811425018310547
95th percentile: 1.975676918029785
99th percentile: 2.1070784378051757
mean time: 1.462472915649414
Pipeline stage StressChecker completed in 8.15s
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-uhh_v1 status is now deployed due to DeploymentManager action
hastagaras-llama3-8b-uhh_v1 status is now inactive due to auto deactivation removed underperforming models
admin requested tearing down of hastagaras-llama3-8b-uhh_v1
Running pipeline stage ISVCDeleter
Checking if service hastagaras-llama3-8b-uhh-v1 is running
Tearing down inference service hastagaras-llama3-8b-uhh-v1
Toredown service hastagaras-llama3-8b-uhh-v1
Pipeline stage ISVCDeleter completed in 4.49s
Running pipeline stage MKMLModelDeleter
Cleaning model data from S3
Cleaning model data from model cache
Deleting key hastagaras-llama3-8b-uhh-v1/config.json from bucket guanaco-mkml-models
Deleting key hastagaras-llama3-8b-uhh-v1/flywheel_model.0.safetensors from bucket guanaco-mkml-models
Deleting key hastagaras-llama3-8b-uhh-v1/special_tokens_map.json from bucket guanaco-mkml-models
Deleting key hastagaras-llama3-8b-uhh-v1/tokenizer.json from bucket guanaco-mkml-models
Deleting key hastagaras-llama3-8b-uhh-v1/tokenizer_config.json from bucket guanaco-mkml-models
Cleaning model data from model cache
Deleting key hastagaras-llama3-8b-uhh-v1_reward/config.json from bucket guanaco-reward-models
Deleting key hastagaras-llama3-8b-uhh-v1_reward/merges.txt from bucket guanaco-reward-models
Deleting key hastagaras-llama3-8b-uhh-v1_reward/reward.tensors from bucket guanaco-reward-models
Deleting key hastagaras-llama3-8b-uhh-v1_reward/special_tokens_map.json from bucket guanaco-reward-models
Deleting key hastagaras-llama3-8b-uhh-v1_reward/tokenizer.json from bucket guanaco-reward-models
Deleting key hastagaras-llama3-8b-uhh-v1_reward/tokenizer_config.json from bucket guanaco-reward-models
Deleting key hastagaras-llama3-8b-uhh-v1_reward/vocab.json from bucket guanaco-reward-models
Pipeline stage MKMLModelDeleter completed in 1.87s
hastagaras-llama3-8b-uhh_v1 status is now torndown due to DeploymentManager action

Usage Metrics

Latency Metrics