cycy233-llm3-1

developer_uid: shiroe40

submission_id: cycy233-llm3-1_v1

model_name: rolev1

model_group: cycy233/llm3.1

status: torndown

timestamp: 2024-05-09T09:39:18+00:00

num_battles: 12560

num_wins: 5141

celo_rating: 1105.69

family_friendly_score: 0.0

submission_type: basic

model_repo: cycy233/llm3.1

model_architecture: LlamaForCausalLM

reward_repo: ChaiML/reward_gpt2_medium_preference_24m_e2

model_num_parameters: 6738415616.0

best_of: 16

max_input_tokens: 512

max_output_tokens: 64

display_name: rolev1

is_internal_developer: False

language_model: cycy233/llm3.1

model_size: 7B

ranking_group: single

us_pacific_date: 2024-05-09

win_ratio: 0.4093152866242038

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': '"### Instruction:\nYou are a creative agent roleplaying as a character called {bot_name}. Stay true to the persona given, reply with short and descriptive sentences. Do not be repetitive.\n{bot_name}\'s Persona: {memory}\n"', 'prompt_template': '"### Input:\n# Example conversation:\n{prompt}\n# Actual conversation:\n"', 'bot_template': '{bot_name}: {message}\n', 'user_template': '{user_name}: {message}\n', 'response_template': '"### Response:\n{bot_name}:"', 'truncate_by_message': False}

model_eval_status: success

reward_formatter: {'bot_template': '{bot_name}: {message}\n', 'memory_template': "{bot_name}'s Persona: {memory}\n####\n", 'prompt_template': '{prompt}\n<START>\n', 'response_template': '{bot_name}:', 'truncate_by_message': False, 'user_template': '{user_name}: {message}\n'}

Resubmit model

Running pipeline stage MKMLizer
Starting job with name cycy233-llm3-1-v1-mkmlizer
Waiting for job on cycy233-llm3-1-v1-mkmlizer to finish
cycy233-llm3-1-v1-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
cycy233-llm3-1-v1-mkmlizer: ║     _____            __           __                                ║
cycy233-llm3-1-v1-mkmlizer: ║    / _/ /_ ___    __/ /  ___ ___ / /                                ║
cycy233-llm3-1-v1-mkmlizer: ║   / _/ / // / |/|/ / _ \/ -_) -_) /                                 ║
cycy233-llm3-1-v1-mkmlizer: ║  /_//_/\_, /|__,__/_//_/\__/\__/_/                                  ║
cycy233-llm3-1-v1-mkmlizer: ║       /___/                                                         ║
cycy233-llm3-1-v1-mkmlizer: ║                                                                     ║
cycy233-llm3-1-v1-mkmlizer: ║  Version: 0.8.10                                                    ║
cycy233-llm3-1-v1-mkmlizer: ║  Copyright 2023 MK ONE TECHNOLOGIES Inc.                            ║
cycy233-llm3-1-v1-mkmlizer: ║                                                                     ║
cycy233-llm3-1-v1-mkmlizer: ║  The license key for the current software has been verified as      ║
cycy233-llm3-1-v1-mkmlizer: ║  belonging to:                                                      ║
cycy233-llm3-1-v1-mkmlizer: ║                                                                     ║
cycy233-llm3-1-v1-mkmlizer: ║  Chai Research Corp.                                                ║
cycy233-llm3-1-v1-mkmlizer: ║  Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f                   ║
cycy233-llm3-1-v1-mkmlizer: ║  Expiration: 2024-07-15 23:59:59                                    ║
cycy233-llm3-1-v1-mkmlizer: ║                                                                     ║
cycy233-llm3-1-v1-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
cycy233-llm3-1-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.
cycy233-llm3-1-v1-mkmlizer:   warnings.warn(warning_message, FutureWarning)
cycy233-llm3-1-v1-mkmlizer: Downloaded to shared memory in 24.145s
cycy233-llm3-1-v1-mkmlizer: quantizing model to /dev/shm/model_cache
cycy233-llm3-1-v1-mkmlizer: Saving flywheel model at /dev/shm/model_cache
cycy233-llm3-1-v1-mkmlizer: 
Loading 0:   0%|          | 0/291 [00:00<?, ?it/s]
Loading 0:  36%|███▋      | 106/291 [00:01<00:01, 103.94it/s]
Loading 0:  74%|███████▍  | 215/291 [00:02<00:00, 106.37it/s]
Loading 0: 100%|██████████| 291/291 [00:03<00:00, 85.68it/s] 
                                                            
You set `add_prefix_space`. The tokenizer needs to be converted from the slow tokenizers
cycy233-llm3-1-v1-mkmlizer: quantized model in 10.271s
cycy233-llm3-1-v1-mkmlizer: Processed model cycy233/llm3.1 in 35.349s
cycy233-llm3-1-v1-mkmlizer: creating bucket guanaco-mkml-models
cycy233-llm3-1-v1-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
cycy233-llm3-1-v1-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/cycy233-llm3-1-v1
cycy233-llm3-1-v1-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/cycy233-llm3-1-v1/config.json
cycy233-llm3-1-v1-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/cycy233-llm3-1-v1/special_tokens_map.json
cycy233-llm3-1-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/cycy233-llm3-1-v1/tokenizer_config.json
cycy233-llm3-1-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer.model s3://guanaco-mkml-models/cycy233-llm3-1-v1/tokenizer.model
cycy233-llm3-1-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/cycy233-llm3-1-v1/tokenizer.json
cycy233-llm3-1-v1-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/cycy233-llm3-1-v1/flywheel_model.0.safetensors
cycy233-llm3-1-v1-mkmlizer: loading reward model from ChaiML/reward_gpt2_medium_preference_24m_e2
cycy233-llm3-1-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.
cycy233-llm3-1-v1-mkmlizer:   warnings.warn(
cycy233-llm3-1-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.
cycy233-llm3-1-v1-mkmlizer:   warnings.warn(
cycy233-llm3-1-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.
cycy233-llm3-1-v1-mkmlizer:   warnings.warn(
cycy233-llm3-1-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()
cycy233-llm3-1-v1-mkmlizer:   return self.fget.__get__(instance, owner)()
cycy233-llm3-1-v1-mkmlizer: Saving model to /tmp/reward_cache/reward.tensors
cycy233-llm3-1-v1-mkmlizer: Saving duration: 0.247s
cycy233-llm3-1-v1-mkmlizer: Processed model ChaiML/reward_gpt2_medium_preference_24m_e2 in 7.247s
cycy233-llm3-1-v1-mkmlizer: creating bucket guanaco-reward-models
cycy233-llm3-1-v1-mkmlizer: Bucket 's3://guanaco-reward-models/' created
cycy233-llm3-1-v1-mkmlizer: uploading /tmp/reward_cache to s3://guanaco-reward-models/cycy233-llm3-1-v1_reward
cycy233-llm3-1-v1-mkmlizer: cp /tmp/reward_cache/special_tokens_map.json s3://guanaco-reward-models/cycy233-llm3-1-v1_reward/special_tokens_map.json
cycy233-llm3-1-v1-mkmlizer: cp /tmp/reward_cache/tokenizer_config.json s3://guanaco-reward-models/cycy233-llm3-1-v1_reward/tokenizer_config.json
cycy233-llm3-1-v1-mkmlizer: cp /tmp/reward_cache/merges.txt s3://guanaco-reward-models/cycy233-llm3-1-v1_reward/merges.txt
cycy233-llm3-1-v1-mkmlizer: cp /tmp/reward_cache/config.json s3://guanaco-reward-models/cycy233-llm3-1-v1_reward/config.json
cycy233-llm3-1-v1-mkmlizer: cp /tmp/reward_cache/vocab.json s3://guanaco-reward-models/cycy233-llm3-1-v1_reward/vocab.json
cycy233-llm3-1-v1-mkmlizer: cp /tmp/reward_cache/tokenizer.json s3://guanaco-reward-models/cycy233-llm3-1-v1_reward/tokenizer.json
cycy233-llm3-1-v1-mkmlizer: cp /tmp/reward_cache/reward.tensors s3://guanaco-reward-models/cycy233-llm3-1-v1_reward/reward.tensors
Job cycy233-llm3-1-v1-mkmlizer completed after 63.09s with status: succeeded
Stopping job with name cycy233-llm3-1-v1-mkmlizer
Pipeline stage MKMLizer completed in 66.17s
Running pipeline stage MKMLKubeTemplater
Pipeline stage MKMLKubeTemplater completed in 0.09s
Running pipeline stage ISVCDeployer
Creating inference service cycy233-llm3-1-v1
Waiting for inference service cycy233-llm3-1-v1 to be ready
Inference service cycy233-llm3-1-v1 ready after 40.31077480316162s
Pipeline stage ISVCDeployer completed in 47.25s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.1631362438201904s
Received healthy response to inference request in 1.2309980392456055s
Received healthy response to inference request in 1.193830966949463s
Received healthy response to inference request in 1.2180352210998535s
Received healthy response to inference request in 1.1919898986816406s
5 requests
0 failed requests
5th percentile: 1.1923581123352052
10th percentile: 1.1927263259887695
20th percentile: 1.1934627532958983
30th percentile: 1.198671817779541
40th percentile: 1.2083535194396973
50th percentile: 1.2180352210998535
60th percentile: 1.2232203483581543
70th percentile: 1.228405475616455
80th percentile: 1.4174256801605227
90th percentile: 1.7902809619903566
95th percentile: 1.9767086029052732
99th percentile: 2.125850715637207
mean time: 1.3995980739593505
Pipeline stage StressChecker completed in 7.75s
Running pipeline stage DaemonicModelEvalScorer
Pipeline stage DaemonicModelEvalScorer completed in 0.03s
Running pipeline stage DaemonicSafetyScorer
Pipeline stage DaemonicSafetyScorer completed in 0.03s
Running M-Eval for topic stay_in_character
cycy233-llm3-1_v1 status is now deployed due to DeploymentManager action
M-Eval Dataset for topic stay_in_character is loaded
cycy233-llm3-1_v1 status is now inactive due to auto deactivation removed underperforming models
admin requested tearing down of cycy233-llm3-1_v1
Running pipeline stage ISVCDeleter
Checking if service cycy233-llm3-1-v1 is running
Tearing down inference service cycy233-llm3-1-v1
Toredown service cycy233-llm3-1-v1
Pipeline stage ISVCDeleter completed in 4.53s
Running pipeline stage MKMLModelDeleter
Cleaning model data from S3
Cleaning model data from model cache
Deleting key cycy233-llm3-1-v1/config.json from bucket guanaco-mkml-models
Deleting key cycy233-llm3-1-v1/flywheel_model.0.safetensors from bucket guanaco-mkml-models
Deleting key cycy233-llm3-1-v1/special_tokens_map.json from bucket guanaco-mkml-models
Deleting key cycy233-llm3-1-v1/tokenizer.json from bucket guanaco-mkml-models
Deleting key cycy233-llm3-1-v1/tokenizer.model from bucket guanaco-mkml-models
Deleting key cycy233-llm3-1-v1/tokenizer_config.json from bucket guanaco-mkml-models
Cleaning model data from model cache
Deleting key cycy233-llm3-1-v1_reward/config.json from bucket guanaco-reward-models
Deleting key cycy233-llm3-1-v1_reward/merges.txt from bucket guanaco-reward-models
Deleting key cycy233-llm3-1-v1_reward/reward.tensors from bucket guanaco-reward-models
Deleting key cycy233-llm3-1-v1_reward/special_tokens_map.json from bucket guanaco-reward-models
Deleting key cycy233-llm3-1-v1_reward/tokenizer.json from bucket guanaco-reward-models
Deleting key cycy233-llm3-1-v1_reward/tokenizer_config.json from bucket guanaco-reward-models
Deleting key cycy233-llm3-1-v1_reward/vocab.json from bucket guanaco-reward-models
Pipeline stage MKMLModelDeleter completed in 2.97s
cycy233-llm3-1_v1 status is now torndown due to DeploymentManager action