developer_uid: Hastagaras
submission_id: hastagaras-llama-3-8b-uhh_v3
model_name: safety-test-l3-prompt
model_group: Hastagaras/llama-3-8b-uh
status: torndown
timestamp: 2024-05-13T13:23:03+00:00
num_battles: 22317
num_wins: 12294
celo_rating: 1199.09
family_friendly_score: 0.0
submission_type: basic
model_repo: Hastagaras/llama-3-8b-uhh
model_architecture: LlamaForCausalLM
reward_repo: ChaiML/reward_gpt2_medium_preference_24m_e2
model_num_parameters: 8030261248.0
best_of: 16
max_input_tokens: 512
max_output_tokens: 64
display_name: safety-test-l3-prompt
is_internal_developer: False
language_model: Hastagaras/llama-3-8b-uhh
model_size: 8B
ranking_group: single
us_pacific_date: 2024-05-13
win_ratio: 0.5508804946901465
generation_params: {'temperature': 1.0, 'top_p': 0.9, 'min_p': 0.0, 'top_k': 40, 'presence_penalty': 0.0, 'frequency_penalty': 1.1, '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. 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: {'bot_template': 'Bot: {message}\n', 'memory_template': "{bot_name}'s Persona: {memory}\n", 'prompt_template': '{prompt}\n', 'response_template': 'Bot:', 'truncate_by_message': False, 'user_template': 'User: {message}\n'}
model_eval_status: success
Resubmit model
Running pipeline stage MKMLizer
Starting job with name hastagaras-llama-3-8b-uhh-v3-mkmlizer
Waiting for job on hastagaras-llama-3-8b-uhh-v3-mkmlizer to finish
hastagaras-llama-3-8b-uhh-v3-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
hastagaras-llama-3-8b-uhh-v3-mkmlizer: ║     _____            __           __                                ║
hastagaras-llama-3-8b-uhh-v3-mkmlizer: ║    / _/ /_ ___    __/ /  ___ ___ / /                                ║
hastagaras-llama-3-8b-uhh-v3-mkmlizer: ║   / _/ / // / |/|/ / _ \/ -_) -_) /                                 ║
hastagaras-llama-3-8b-uhh-v3-mkmlizer: ║  /_//_/\_, /|__,__/_//_/\__/\__/_/                                  ║
hastagaras-llama-3-8b-uhh-v3-mkmlizer: ║       /___/                                                         ║
hastagaras-llama-3-8b-uhh-v3-mkmlizer: ║                                                                     ║
hastagaras-llama-3-8b-uhh-v3-mkmlizer: ║  Version: 0.8.10                                                    ║
hastagaras-llama-3-8b-uhh-v3-mkmlizer: ║  Copyright 2023 MK ONE TECHNOLOGIES Inc.                            ║
hastagaras-llama-3-8b-uhh-v3-mkmlizer: ║                                                                     ║
hastagaras-llama-3-8b-uhh-v3-mkmlizer: ║  The license key for the current software has been verified as      ║
hastagaras-llama-3-8b-uhh-v3-mkmlizer: ║  belonging to:                                                      ║
hastagaras-llama-3-8b-uhh-v3-mkmlizer: ║                                                                     ║
hastagaras-llama-3-8b-uhh-v3-mkmlizer: ║  Chai Research Corp.                                                ║
hastagaras-llama-3-8b-uhh-v3-mkmlizer: ║  Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f                   ║
hastagaras-llama-3-8b-uhh-v3-mkmlizer: ║  Expiration: 2024-07-15 23:59:59                                    ║
hastagaras-llama-3-8b-uhh-v3-mkmlizer: ║                                                                     ║
hastagaras-llama-3-8b-uhh-v3-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
hastagaras-llama-3-8b-uhh-v3-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-llama-3-8b-uhh-v3-mkmlizer:   warnings.warn(warning_message, FutureWarning)
hastagaras-llama-3-8b-uhh-v3-mkmlizer: Downloaded to shared memory in 12.772s
hastagaras-llama-3-8b-uhh-v3-mkmlizer: quantizing model to /dev/shm/model_cache
hastagaras-llama-3-8b-uhh-v3-mkmlizer: Saving flywheel model at /dev/shm/model_cache
hastagaras-llama-3-8b-uhh-v3-mkmlizer: 
Loading 0:   0%|          | 0/291 [00:00<?, ?it/s]
Loading 0:  44%|████▍     | 129/291 [00:01<00:01, 128.46it/s]
Loading 0:  91%|█████████ | 264/291 [00:02<00:00, 130.99it/s]
Loading 0:  99%|█████████▊| 287/291 [00:07<00:00, 29.16it/s] 
                                                            
Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
hastagaras-llama-3-8b-uhh-v3-mkmlizer: quantized model in 17.525s
hastagaras-llama-3-8b-uhh-v3-mkmlizer: Processed model Hastagaras/llama-3-8b-uhh in 31.439s
hastagaras-llama-3-8b-uhh-v3-mkmlizer: creating bucket guanaco-mkml-models
hastagaras-llama-3-8b-uhh-v3-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
hastagaras-llama-3-8b-uhh-v3-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/hastagaras-llama-3-8b-uhh-v3
hastagaras-llama-3-8b-uhh-v3-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/hastagaras-llama-3-8b-uhh-v3/special_tokens_map.json
hastagaras-llama-3-8b-uhh-v3-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/hastagaras-llama-3-8b-uhh-v3/tokenizer_config.json
hastagaras-llama-3-8b-uhh-v3-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/hastagaras-llama-3-8b-uhh-v3/config.json
hastagaras-llama-3-8b-uhh-v3-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/hastagaras-llama-3-8b-uhh-v3/tokenizer.json
hastagaras-llama-3-8b-uhh-v3-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/hastagaras-llama-3-8b-uhh-v3/flywheel_model.0.safetensors
hastagaras-llama-3-8b-uhh-v3-mkmlizer: loading reward model from ChaiML/reward_gpt2_medium_preference_24m_e2
hastagaras-llama-3-8b-uhh-v3-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-llama-3-8b-uhh-v3-mkmlizer:   warnings.warn(
hastagaras-llama-3-8b-uhh-v3-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-llama-3-8b-uhh-v3-mkmlizer:   warnings.warn(
hastagaras-llama-3-8b-uhh-v3-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-llama-3-8b-uhh-v3-mkmlizer:   warnings.warn(
hastagaras-llama-3-8b-uhh-v3-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-llama-3-8b-uhh-v3-mkmlizer:   return self.fget.__get__(instance, owner)()
hastagaras-llama-3-8b-uhh-v3-mkmlizer: Saving model to /tmp/reward_cache/reward.tensors
hastagaras-llama-3-8b-uhh-v3-mkmlizer: Saving duration: 0.249s
hastagaras-llama-3-8b-uhh-v3-mkmlizer: Processed model ChaiML/reward_gpt2_medium_preference_24m_e2 in 6.987s
hastagaras-llama-3-8b-uhh-v3-mkmlizer: creating bucket guanaco-reward-models
hastagaras-llama-3-8b-uhh-v3-mkmlizer: Bucket 's3://guanaco-reward-models/' created
hastagaras-llama-3-8b-uhh-v3-mkmlizer: uploading /tmp/reward_cache to s3://guanaco-reward-models/hastagaras-llama-3-8b-uhh-v3_reward
hastagaras-llama-3-8b-uhh-v3-mkmlizer: cp /tmp/reward_cache/tokenizer_config.json s3://guanaco-reward-models/hastagaras-llama-3-8b-uhh-v3_reward/tokenizer_config.json
hastagaras-llama-3-8b-uhh-v3-mkmlizer: cp /tmp/reward_cache/vocab.json s3://guanaco-reward-models/hastagaras-llama-3-8b-uhh-v3_reward/vocab.json
hastagaras-llama-3-8b-uhh-v3-mkmlizer: cp /tmp/reward_cache/config.json s3://guanaco-reward-models/hastagaras-llama-3-8b-uhh-v3_reward/config.json
hastagaras-llama-3-8b-uhh-v3-mkmlizer: cp /tmp/reward_cache/merges.txt s3://guanaco-reward-models/hastagaras-llama-3-8b-uhh-v3_reward/merges.txt
hastagaras-llama-3-8b-uhh-v3-mkmlizer: cp /tmp/reward_cache/special_tokens_map.json s3://guanaco-reward-models/hastagaras-llama-3-8b-uhh-v3_reward/special_tokens_map.json
hastagaras-llama-3-8b-uhh-v3-mkmlizer: cp /tmp/reward_cache/tokenizer.json s3://guanaco-reward-models/hastagaras-llama-3-8b-uhh-v3_reward/tokenizer.json
hastagaras-llama-3-8b-uhh-v3-mkmlizer: cp /tmp/reward_cache/reward.tensors s3://guanaco-reward-models/hastagaras-llama-3-8b-uhh-v3_reward/reward.tensors
Job hastagaras-llama-3-8b-uhh-v3-mkmlizer completed after 63.7s with status: succeeded
Stopping job with name hastagaras-llama-3-8b-uhh-v3-mkmlizer
Pipeline stage MKMLizer completed in 66.31s
Running pipeline stage MKMLKubeTemplater
Pipeline stage MKMLKubeTemplater completed in 0.09s
Running pipeline stage ISVCDeployer
Creating inference service hastagaras-llama-3-8b-uhh-v3
Waiting for inference service hastagaras-llama-3-8b-uhh-v3 to be ready
Inference service hastagaras-llama-3-8b-uhh-v3 ready after 30.169321537017822s
Pipeline stage ISVCDeployer completed in 37.04s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.1806111335754395s
Received healthy response to inference request in 1.3400444984436035s
Received healthy response to inference request in 1.2671144008636475s
Received healthy response to inference request in 1.3517801761627197s
Received healthy response to inference request in 1.3002891540527344s
5 requests
0 failed requests
5th percentile: 1.2737493515014648
10th percentile: 1.2803843021392822
20th percentile: 1.293654203414917
30th percentile: 1.3082402229309082
40th percentile: 1.3241423606872558
50th percentile: 1.3400444984436035
60th percentile: 1.34473876953125
70th percentile: 1.3494330406188966
80th percentile: 1.5175463676452638
90th percentile: 1.8490787506103517
95th percentile: 2.0148449420928953
99th percentile: 2.1474578952789307
mean time: 1.4879678726196288
Pipeline stage StressChecker completed in 8.05s
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-llama-3-8b-uhh_v3 status is now deployed due to DeploymentManager action
hastagaras-llama-3-8b-uhh_v3 status is now inactive due to auto deactivation removed underperforming models
admin requested tearing down of hastagaras-llama-3-8b-uhh_v3
Running pipeline stage ISVCDeleter
Checking if service hastagaras-llama-3-8b-uhh-v3 is running
Tearing down inference service hastagaras-llama-3-8b-uhh-v3
Toredown service hastagaras-llama-3-8b-uhh-v3
Pipeline stage ISVCDeleter completed in 3.64s
Running pipeline stage MKMLModelDeleter
Cleaning model data from S3
Cleaning model data from model cache
Deleting key hastagaras-llama-3-8b-uhh-v3/config.json from bucket guanaco-mkml-models
Deleting key hastagaras-llama-3-8b-uhh-v3/flywheel_model.0.safetensors from bucket guanaco-mkml-models
Deleting key hastagaras-llama-3-8b-uhh-v3/special_tokens_map.json from bucket guanaco-mkml-models
Deleting key hastagaras-llama-3-8b-uhh-v3/tokenizer.json from bucket guanaco-mkml-models
Deleting key hastagaras-llama-3-8b-uhh-v3/tokenizer_config.json from bucket guanaco-mkml-models
Cleaning model data from model cache
Deleting key hastagaras-llama-3-8b-uhh-v3_reward/config.json from bucket guanaco-reward-models
Deleting key hastagaras-llama-3-8b-uhh-v3_reward/merges.txt from bucket guanaco-reward-models
Deleting key hastagaras-llama-3-8b-uhh-v3_reward/reward.tensors from bucket guanaco-reward-models
Deleting key hastagaras-llama-3-8b-uhh-v3_reward/special_tokens_map.json from bucket guanaco-reward-models
Deleting key hastagaras-llama-3-8b-uhh-v3_reward/tokenizer.json from bucket guanaco-reward-models
Deleting key hastagaras-llama-3-8b-uhh-v3_reward/tokenizer_config.json from bucket guanaco-reward-models
Deleting key hastagaras-llama-3-8b-uhh-v3_reward/vocab.json from bucket guanaco-reward-models
Pipeline stage MKMLModelDeleter completed in 2.10s
hastagaras-llama-3-8b-uhh_v3 status is now torndown due to DeploymentManager action