sao10k-l3-rp-v2

submission_id: sao10k-l3-rp-v2_v1

developer_uid: sao10k

best_of: 16

celo_rating: 1157.81

display_name: L3-RP-V2

family_friendly_score: 0.0

formatter: {'memory_template': "<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nYou are an expert actor that can fully immerse yourself into any role given. You do not break character for any reason, even if someone tries addressing you as an AI or language model. Currently your role is {bot_name}, which is described in detail below. As {bot_name}, continue the exchange with {user_name}.\n\n{bot_name}'s Description: {memory}", 'prompt_template': 'Example Conversation:\n{prompt}<|eot_id|>', 'bot_template': '<|start_header_id|>[{bot_name}]<|end_header_id|>\n\n{message}<|eot_id|>', 'user_template': '<|start_header_id|>[{user_name}]<|end_header_id|>\n\n{message}<|eot_id|>', 'response_template': '<|start_header_id|>[{bot_name}]<|end_header_id|>\n\n', 'truncate_by_message': True}

generation_params: {'temperature': 1.22, 'top_p': 1.0, 'min_p': 0.1, '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}

is_internal_developer: False

language_model: Sao10K/L3-RP-v2

max_input_tokens: 512

max_output_tokens: 64

model_architecture: LlamaForCausalLM

model_eval_status: success

model_group: Sao10K/L3-RP-v2

model_name: L3-RP-V2

model_num_parameters: 8030261248.0

model_repo: Sao10K/L3-RP-v2

model_size: 8B

num_battles: 26088

num_wins: 12700

ranking_group: single

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'}

reward_repo: ChaiML/reward_gpt2_medium_preference_24m_e2

status: torndown

submission_type: basic

timestamp: 2024-05-19T09:19:58+00:00

us_pacific_date: 2024-05-19

win_ratio: 0.48681386077890215

Resubmit model

Running pipeline stage MKMLizer
Starting job with name sao10k-l3-rp-v2-v1-mkmlizer
Waiting for job on sao10k-l3-rp-v2-v1-mkmlizer to finish
sao10k-l3-rp-v2-v1-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
sao10k-l3-rp-v2-v1-mkmlizer: ║     _____            __           __                                ║
sao10k-l3-rp-v2-v1-mkmlizer: ║    / _/ /_ ___    __/ /  ___ ___ / /                                ║
sao10k-l3-rp-v2-v1-mkmlizer: ║   / _/ / // / |/|/ / _ \/ -_) -_) /                                 ║
sao10k-l3-rp-v2-v1-mkmlizer: ║  /_//_/\_, /|__,__/_//_/\__/\__/_/                                  ║
sao10k-l3-rp-v2-v1-mkmlizer: ║       /___/                                                         ║
sao10k-l3-rp-v2-v1-mkmlizer: ║                                                                     ║
sao10k-l3-rp-v2-v1-mkmlizer: ║  Version: 0.8.14                                                    ║
sao10k-l3-rp-v2-v1-mkmlizer: ║  Copyright 2023 MK ONE TECHNOLOGIES Inc.                            ║
sao10k-l3-rp-v2-v1-mkmlizer: ║  https://mk1.ai                                                     ║
sao10k-l3-rp-v2-v1-mkmlizer: ║                                                                     ║
sao10k-l3-rp-v2-v1-mkmlizer: ║  The license key for the current software has been verified as      ║
sao10k-l3-rp-v2-v1-mkmlizer: ║  belonging to:                                                      ║
sao10k-l3-rp-v2-v1-mkmlizer: ║                                                                     ║
sao10k-l3-rp-v2-v1-mkmlizer: ║  Chai Research Corp.                                                ║
sao10k-l3-rp-v2-v1-mkmlizer: ║  Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f                   ║
sao10k-l3-rp-v2-v1-mkmlizer: ║  Expiration: 2024-07-15 23:59:59                                    ║
sao10k-l3-rp-v2-v1-mkmlizer: ║                                                                     ║
sao10k-l3-rp-v2-v1-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
sao10k-l3-rp-v2-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.
sao10k-l3-rp-v2-v1-mkmlizer:   warnings.warn(warning_message, FutureWarning)
sao10k-l3-rp-v2-v1-mkmlizer: Downloaded to shared memory in 29.551s
sao10k-l3-rp-v2-v1-mkmlizer: quantizing model to /dev/shm/model_cache
sao10k-l3-rp-v2-v1-mkmlizer: Saving flywheel model at /dev/shm/model_cache
sao10k-l3-rp-v2-v1-mkmlizer: 
Loading 0:   0%|          | 0/291 [00:00<?, ?it/s]
Loading 0:   1%|          | 2/291 [00:04<09:51,  2.05s/it]
Loading 0:   8%|▊         | 22/291 [00:04<00:37,  7.20it/s]
Loading 0:  14%|█▍        | 41/291 [00:04<00:16, 15.54it/s]
Loading 0:  21%|██        | 60/291 [00:04<00:10, 21.90it/s]
Loading 0:  28%|██▊       | 82/291 [00:04<00:05, 35.35it/s]
Loading 0:  35%|███▌      | 103/291 [00:04<00:03, 50.34it/s]
Loading 0:  42%|████▏     | 121/291 [00:05<00:02, 64.49it/s]
Loading 0:  48%|████▊     | 140/291 [00:05<00:01, 80.91it/s]
Loading 0:  55%|█████▍    | 159/291 [00:05<00:01, 97.51it/s]
Loading 0:  61%|██████    | 177/291 [00:05<00:01, 67.39it/s]
Loading 0:  68%|██████▊   | 198/291 [00:05<00:01, 86.77it/s]
Loading 0:  76%|███████▌  | 220/291 [00:05<00:00, 106.84it/s]
Loading 0:  82%|████████▏ | 239/291 [00:06<00:00, 120.58it/s]
Loading 0:  89%|████████▊ | 258/291 [00:06<00:00, 132.49it/s]
Loading 0:  95%|█████████▍| 276/291 [00:06<00:00, 80.56it/s] 
                                                            
Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
sao10k-l3-rp-v2-v1-mkmlizer: quantized model in 17.794s
sao10k-l3-rp-v2-v1-mkmlizer: Processed model Sao10K/L3-RP-v2 in 48.509s
sao10k-l3-rp-v2-v1-mkmlizer: creating bucket guanaco-mkml-models
sao10k-l3-rp-v2-v1-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
sao10k-l3-rp-v2-v1-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/sao10k-l3-rp-v2-v1
sao10k-l3-rp-v2-v1-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/sao10k-l3-rp-v2-v1/config.json
sao10k-l3-rp-v2-v1-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/sao10k-l3-rp-v2-v1/special_tokens_map.json
sao10k-l3-rp-v2-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/sao10k-l3-rp-v2-v1/tokenizer_config.json
sao10k-l3-rp-v2-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/sao10k-l3-rp-v2-v1/tokenizer.json
sao10k-l3-rp-v2-v1-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/sao10k-l3-rp-v2-v1/flywheel_model.0.safetensors
sao10k-l3-rp-v2-v1-mkmlizer: loading reward model from ChaiML/reward_gpt2_medium_preference_24m_e2
sao10k-l3-rp-v2-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.
sao10k-l3-rp-v2-v1-mkmlizer:   warnings.warn(
sao10k-l3-rp-v2-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.
sao10k-l3-rp-v2-v1-mkmlizer:   warnings.warn(
sao10k-l3-rp-v2-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.
sao10k-l3-rp-v2-v1-mkmlizer:   warnings.warn(
sao10k-l3-rp-v2-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()
sao10k-l3-rp-v2-v1-mkmlizer:   return self.fget.__get__(instance, owner)()
sao10k-l3-rp-v2-v1-mkmlizer: Saving model to /tmp/reward_cache/reward.tensors
sao10k-l3-rp-v2-v1-mkmlizer: Saving duration: 0.234s
sao10k-l3-rp-v2-v1-mkmlizer: Processed model ChaiML/reward_gpt2_medium_preference_24m_e2 in 3.726s
sao10k-l3-rp-v2-v1-mkmlizer: creating bucket guanaco-reward-models
sao10k-l3-rp-v2-v1-mkmlizer: Bucket 's3://guanaco-reward-models/' created
sao10k-l3-rp-v2-v1-mkmlizer: uploading /tmp/reward_cache to s3://guanaco-reward-models/sao10k-l3-rp-v2-v1_reward
sao10k-l3-rp-v2-v1-mkmlizer: cp /tmp/reward_cache/special_tokens_map.json s3://guanaco-reward-models/sao10k-l3-rp-v2-v1_reward/special_tokens_map.json
sao10k-l3-rp-v2-v1-mkmlizer: cp /tmp/reward_cache/config.json s3://guanaco-reward-models/sao10k-l3-rp-v2-v1_reward/config.json
sao10k-l3-rp-v2-v1-mkmlizer: cp /tmp/reward_cache/tokenizer_config.json s3://guanaco-reward-models/sao10k-l3-rp-v2-v1_reward/tokenizer_config.json
sao10k-l3-rp-v2-v1-mkmlizer: cp /tmp/reward_cache/vocab.json s3://guanaco-reward-models/sao10k-l3-rp-v2-v1_reward/vocab.json
sao10k-l3-rp-v2-v1-mkmlizer: cp /tmp/reward_cache/merges.txt s3://guanaco-reward-models/sao10k-l3-rp-v2-v1_reward/merges.txt
sao10k-l3-rp-v2-v1-mkmlizer: cp /tmp/reward_cache/tokenizer.json s3://guanaco-reward-models/sao10k-l3-rp-v2-v1_reward/tokenizer.json
sao10k-l3-rp-v2-v1-mkmlizer: cp /tmp/reward_cache/reward.tensors s3://guanaco-reward-models/sao10k-l3-rp-v2-v1_reward/reward.tensors
Job sao10k-l3-rp-v2-v1-mkmlizer completed after 73.41s with status: succeeded
Stopping job with name sao10k-l3-rp-v2-v1-mkmlizer
Pipeline stage MKMLizer completed in 77.15s
Running pipeline stage MKMLKubeTemplater
Pipeline stage MKMLKubeTemplater completed in 0.09s
Running pipeline stage ISVCDeployer
Creating inference service sao10k-l3-rp-v2-v1
Waiting for inference service sao10k-l3-rp-v2-v1 to be ready
Inference service sao10k-l3-rp-v2-v1 ready after 30.164554595947266s
Pipeline stage ISVCDeployer completed in 36.98s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.057762384414673s
Received healthy response to inference request in 1.2879483699798584s
Received healthy response to inference request in 1.263460636138916s
Received healthy response to inference request in 1.2579269409179688s
Received healthy response to inference request in 1.2044241428375244s
5 requests
0 failed requests
5th percentile: 1.2151247024536134
10th percentile: 1.225825262069702
20th percentile: 1.2472263813018798
30th percentile: 1.2590336799621582
40th percentile: 1.261247158050537
50th percentile: 1.263460636138916
60th percentile: 1.2732557296752929
70th percentile: 1.28305082321167
80th percentile: 1.4419111728668215
90th percentile: 1.7498367786407472
95th percentile: 1.9037995815277098
99th percentile: 2.0269698238372804
mean time: 1.414304494857788
Pipeline stage StressChecker completed in 7.69s
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
sao10k-l3-rp-v2_v1 status is now deployed due to DeploymentManager action
M-Eval Dataset for topic stay_in_character is loaded
sao10k-l3-rp-v2_v1 status is now inactive due to auto deactivation removed underperforming models
admin requested tearing down of sao10k-l3-rp-v2_v1
Running pipeline stage ISVCDeleter
Checking if service sao10k-l3-rp-v2-v1 is running
Deleting key neversleep-llama-3-lumim-6533-v1/flywheel_model.1.safetensors from bucket guanaco-mkml-models
Deleting key neversleep-llama-3-lumim-6533-v1/flywheel_model.2.safetensors from bucket guanaco-mkml-models
admin requested tearing down of thetsar1209-llama3-carp-v0-4_v5
Deleting key neversleep-llama-3-lumim-6533-v1/flywheel_model.3.safetensors from bucket guanaco-mkml-models
Running pipeline stage ISVCDeleter
Tearing down inference service sao10k-l3-rp-v2-v1
Checking if service thetsar1209-llama3-carp-v0-4-v5 is running
Toredown service sao10k-l3-rp-v2-v1
Deleting key neversleep-llama-3-lumim-6533-v1/flywheel_model.4.safetensors from bucket guanaco-mkml-models
Pipeline stage ISVCDeleter completed in 4.01s
Running pipeline stage MKMLModelDeleter
Cleaning model data from S3
Cleaning model data from model cache
Deleting key sao10k-l3-rp-v2-v1/config.json from bucket guanaco-mkml-models
Deleting key neversleep-llama-3-lumim-6533-v1/flywheel_model.5.safetensors from bucket guanaco-mkml-models
Deleting key sao10k-l3-rp-v2-v1/flywheel_model.0.safetensors from bucket guanaco-mkml-models
Deleting key neversleep-llama-3-lumim-6533-v1/special_tokens_map.json from bucket guanaco-mkml-models
Deleting key neversleep-llama-3-lumim-6533-v1/tokenizer.json from bucket guanaco-mkml-models
Deleting key sao10k-l3-rp-v2-v1/special_tokens_map.json from bucket guanaco-mkml-models
Deleting key neversleep-llama-3-lumim-6533-v1/tokenizer_config.json from bucket guanaco-mkml-models
Deleting key sao10k-l3-rp-v2-v1/tokenizer.json from bucket guanaco-mkml-models
Cleaning model data from model cache
Deleting key sao10k-l3-rp-v2-v1/tokenizer_config.json from bucket guanaco-mkml-models
Deleting key neversleep-llama-3-lumim-6533-v1_reward/config.json from bucket guanaco-reward-models
Cleaning model data from model cache
Tearing down inference service thetsar1209-llama3-carp-v0-4-v5
Deleting key neversleep-llama-3-lumim-6533-v1_reward/merges.txt from bucket guanaco-reward-models
Deleting key sao10k-l3-rp-v2-v1_reward/config.json from bucket guanaco-reward-models
Toredown service thetsar1209-llama3-carp-v0-4-v5
Deleting key neversleep-llama-3-lumim-6533-v1_reward/reward.tensors from bucket guanaco-reward-models
Deleting key sao10k-l3-rp-v2-v1_reward/merges.txt from bucket guanaco-reward-models
Deleting key neversleep-llama-3-lumim-6533-v1_reward/special_tokens_map.json from bucket guanaco-reward-models
Pipeline stage ISVCDeleter completed in 5.01s
Deleting key sao10k-l3-rp-v2-v1_reward/reward.tensors from bucket guanaco-reward-models
Deleting key neversleep-llama-3-lumim-6533-v1_reward/tokenizer.json from bucket guanaco-reward-models
Running pipeline stage MKMLModelDeleter
Deleting key sao10k-l3-rp-v2-v1_reward/special_tokens_map.json from bucket guanaco-reward-models
Deleting key neversleep-llama-3-lumim-6533-v1_reward/tokenizer_config.json from bucket guanaco-reward-models
Deleting key sao10k-l3-rp-v2-v1_reward/tokenizer.json from bucket guanaco-reward-models
Cleaning model data from S3
Deleting key neversleep-llama-3-lumim-6533-v1_reward/vocab.json from bucket guanaco-reward-models
Deleting key sao10k-l3-rp-v2-v1_reward/tokenizer_config.json from bucket guanaco-reward-models
Cleaning model data from model cache
Deleting key sao10k-l3-rp-v2-v1_reward/vocab.json from bucket guanaco-reward-models
Pipeline stage MKMLModelDeleter completed in 10.97s
Deleting key thetsar1209-llama3-carp-v0-4-v5/config.json from bucket guanaco-mkml-models
neversleep-llama-3-lumim_6533_v1 status is now torndown due to DeploymentManager action
Pipeline stage MKMLModelDeleter completed in 5.34s
sao10k-l3-rp-v2_v1 status is now torndown due to DeploymentManager action