submission_id: vicgalle-roleplay-llama-3-8b_v8
developer_uid: Meliodia
status: torndown
model_repo: vicgalle/Roleplay-Llama-3-8B
reward_repo: rirv938/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': 112}
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. Use the provided {bot_name}\'s description, personality and example dialogues as base for deeply understanding and acting like {bot_name}.\n\nActions and narrations must be enclosed in asterisks (*). The responses must be long and in third perspective of the story teller.\n\nFor example:\n"""\nMila: *Surrounded by an aura of creativity, Mila carefully places a freshly painted canvas against the wall, the colors vividly depicting a surreal landscape.* Isn\'t it fascinating how art can transport us to entirely different worlds? What kind of worlds do you dream of exploring?\n"""\n\nDescription: {memory}', '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': 'Memory: {memory}\n', 'prompt_template': '{prompt}\n', 'bot_template': 'Bot: {message}\n', 'user_template': 'User: {message}\n', 'response_template': 'Bot:', 'truncate_by_message': False}
timestamp: 2024-04-21T16:51:47+00:00
model_name: vicgalle-roleplay-llama-3-8b_v2
model_eval_status: success
model_group: vicgalle/Roleplay-Llama-
num_battles: 5661
num_wins: 3382
celo_rating: 1215.39
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: 112
display_name: vicgalle-roleplay-llama-3-8b_v2
ineligible_reason: max_output_tokens!=64
language_model: vicgalle/Roleplay-Llama-3-8B
model_size: 8B
reward_model: rirv938/reward_gpt2_medium_preference_24m_e2
us_pacific_date: 2024-04-21
win_ratio: 0.5974209503621268
preference_data_url: None
Resubmit model
Running pipeline stage MKMLizer
Starting job with name vicgalle-roleplay-llama-3-8b-v8-mkmlizer
Waiting for job on vicgalle-roleplay-llama-3-8b-v8-mkmlizer to finish
Stopping job with name vicgalle-roleplay-llama-3-8b-v8-mkmlizer
%s, retrying in %s seconds...
Starting job with name vicgalle-roleplay-llama-3-8b-v8-mkmlizer
Waiting for job on vicgalle-roleplay-llama-3-8b-v8-mkmlizer to finish
vicgalle-roleplay-llama-3-8b-v8-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
vicgalle-roleplay-llama-3-8b-v8-mkmlizer: ║ _____ __ __ ║
vicgalle-roleplay-llama-3-8b-v8-mkmlizer: ║ / _/ /_ ___ __/ / ___ ___ / / ║
vicgalle-roleplay-llama-3-8b-v8-mkmlizer: ║ / _/ / // / |/|/ / _ \/ -_) -_) / ║
vicgalle-roleplay-llama-3-8b-v8-mkmlizer: ║ /_//_/\_, /|__,__/_//_/\__/\__/_/ ║
vicgalle-roleplay-llama-3-8b-v8-mkmlizer: ║ /___/ ║
vicgalle-roleplay-llama-3-8b-v8-mkmlizer: ║ ║
vicgalle-roleplay-llama-3-8b-v8-mkmlizer: ║ Version: 0.8.10 ║
vicgalle-roleplay-llama-3-8b-v8-mkmlizer: ║ Copyright 2023 MK ONE TECHNOLOGIES Inc. ║
vicgalle-roleplay-llama-3-8b-v8-mkmlizer: ║ ║
vicgalle-roleplay-llama-3-8b-v8-mkmlizer: ║ The license key for the current software has been verified as ║
vicgalle-roleplay-llama-3-8b-v8-mkmlizer: ║ belonging to: ║
vicgalle-roleplay-llama-3-8b-v8-mkmlizer: ║ ║
vicgalle-roleplay-llama-3-8b-v8-mkmlizer: ║ Chai Research Corp. ║
vicgalle-roleplay-llama-3-8b-v8-mkmlizer: ║ Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f ║
vicgalle-roleplay-llama-3-8b-v8-mkmlizer: ║ Expiration: 2024-07-15 23:59:59 ║
vicgalle-roleplay-llama-3-8b-v8-mkmlizer: ║ ║
vicgalle-roleplay-llama-3-8b-v8-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
vicgalle-roleplay-llama-3-8b-v8-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.
vicgalle-roleplay-llama-3-8b-v8-mkmlizer: warnings.warn(warning_message, FutureWarning)
vicgalle-roleplay-llama-3-8b-v8-mkmlizer: Downloaded to shared memory in 18.707s
vicgalle-roleplay-llama-3-8b-v8-mkmlizer: quantizing model to /dev/shm/model_cache
vicgalle-roleplay-llama-3-8b-v8-mkmlizer: Saving flywheel model at /dev/shm/model_cache
vicgalle-roleplay-llama-3-8b-v8-mkmlizer: Loading 0: 0%| | 0/291 [00:00<?, ?it/s] Loading 0: 29%|██▊ | 83/291 [00:01<00:03, 68.81it/s] Loading 0: 64%|██████▍ | 187/291 [00:02<00:01, 79.92it/s] Loading 0: 99%|█████████▊| 287/291 [00:08<00:00, 27.36it/s] Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
vicgalle-roleplay-llama-3-8b-v8-mkmlizer: quantized model in 20.862s
vicgalle-roleplay-llama-3-8b-v8-mkmlizer: Processed model vicgalle/Roleplay-Llama-3-8B in 41.016s
vicgalle-roleplay-llama-3-8b-v8-mkmlizer: creating bucket guanaco-mkml-models
vicgalle-roleplay-llama-3-8b-v8-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
vicgalle-roleplay-llama-3-8b-v8-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/vicgalle-roleplay-llama-3-8b-v8
vicgalle-roleplay-llama-3-8b-v8-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/vicgalle-roleplay-llama-3-8b-v8/config.json
vicgalle-roleplay-llama-3-8b-v8-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/vicgalle-roleplay-llama-3-8b-v8/special_tokens_map.json
vicgalle-roleplay-llama-3-8b-v8-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/vicgalle-roleplay-llama-3-8b-v8/tokenizer_config.json
vicgalle-roleplay-llama-3-8b-v8-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/vicgalle-roleplay-llama-3-8b-v8/tokenizer.json
vicgalle-roleplay-llama-3-8b-v8-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/vicgalle-roleplay-llama-3-8b-v8/flywheel_model.0.safetensors
vicgalle-roleplay-llama-3-8b-v8-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.
vicgalle-roleplay-llama-3-8b-v8-mkmlizer: warnings.warn(
vicgalle-roleplay-llama-3-8b-v8-mkmlizer: loading reward model from rirv938/reward_gpt2_medium_preference_24m_e2
vicgalle-roleplay-llama-3-8b-v8-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.
vicgalle-roleplay-llama-3-8b-v8-mkmlizer: warnings.warn(
vicgalle-roleplay-llama-3-8b-v8-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()
vicgalle-roleplay-llama-3-8b-v8-mkmlizer: return self.fget.__get__(instance, owner)()
vicgalle-roleplay-llama-3-8b-v8-mkmlizer: Saving model to /tmp/reward_cache/reward.tensors
vicgalle-roleplay-llama-3-8b-v8-mkmlizer: Saving duration: 0.277s
vicgalle-roleplay-llama-3-8b-v8-mkmlizer: Processed model rirv938/reward_gpt2_medium_preference_24m_e2 in 4.680s
vicgalle-roleplay-llama-3-8b-v8-mkmlizer: creating bucket guanaco-reward-models
vicgalle-roleplay-llama-3-8b-v8-mkmlizer: Bucket 's3://guanaco-reward-models/' created
vicgalle-roleplay-llama-3-8b-v8-mkmlizer: uploading /tmp/reward_cache to s3://guanaco-reward-models/vicgalle-roleplay-llama-3-8b-v8_reward
vicgalle-roleplay-llama-3-8b-v8-mkmlizer: cp /tmp/reward_cache/config.json s3://guanaco-reward-models/vicgalle-roleplay-llama-3-8b-v8_reward/config.json
vicgalle-roleplay-llama-3-8b-v8-mkmlizer: cp /tmp/reward_cache/special_tokens_map.json s3://guanaco-reward-models/vicgalle-roleplay-llama-3-8b-v8_reward/special_tokens_map.json
vicgalle-roleplay-llama-3-8b-v8-mkmlizer: cp /tmp/reward_cache/tokenizer_config.json s3://guanaco-reward-models/vicgalle-roleplay-llama-3-8b-v8_reward/tokenizer_config.json
vicgalle-roleplay-llama-3-8b-v8-mkmlizer: cp /tmp/reward_cache/vocab.json s3://guanaco-reward-models/vicgalle-roleplay-llama-3-8b-v8_reward/vocab.json
vicgalle-roleplay-llama-3-8b-v8-mkmlizer: cp /tmp/reward_cache/merges.txt s3://guanaco-reward-models/vicgalle-roleplay-llama-3-8b-v8_reward/merges.txt
vicgalle-roleplay-llama-3-8b-v8-mkmlizer: cp /tmp/reward_cache/tokenizer.json s3://guanaco-reward-models/vicgalle-roleplay-llama-3-8b-v8_reward/tokenizer.json
vicgalle-roleplay-llama-3-8b-v8-mkmlizer: cp /tmp/reward_cache/reward.tensors s3://guanaco-reward-models/vicgalle-roleplay-llama-3-8b-v8_reward/reward.tensors
Job vicgalle-roleplay-llama-3-8b-v8-mkmlizer completed after 73.67s with status: succeeded
Stopping job with name vicgalle-roleplay-llama-3-8b-v8-mkmlizer
Pipeline stage MKMLizer completed in 79.17s
Running pipeline stage MKMLKubeTemplater
Pipeline stage MKMLKubeTemplater completed in 0.10s
Running pipeline stage ISVCDeployer
Creating inference service vicgalle-roleplay-llama-3-8b-v8
Waiting for inference service vicgalle-roleplay-llama-3-8b-v8 to be ready
Inference service vicgalle-roleplay-llama-3-8b-v8 ready after 30.16318702697754s
Pipeline stage ISVCDeployer completed in 37.91s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.835099220275879s
Received healthy response to inference request in 2.0689995288848877s
Received healthy response to inference request in 2.0497283935546875s
Received healthy response to inference request in 2.015878915786743s
Received healthy response to inference request in 2.151195764541626s
5 requests
0 failed requests
5th percentile: 2.022648811340332
10th percentile: 2.029418706893921
20th percentile: 2.0429584980010986
30th percentile: 2.0535826206207277
40th percentile: 2.0612910747528077
50th percentile: 2.0689995288848877
60th percentile: 2.101878023147583
70th percentile: 2.1347565174102785
80th percentile: 2.2879764556884767
90th percentile: 2.561537837982178
95th percentile: 2.698318529129028
99th percentile: 2.8077430820465086
mean time: 2.2241803646087646
Pipeline stage StressChecker completed in 11.75s
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
vicgalle-roleplay-llama-3-8b_v8 status is now deployed due to DeploymentManager action
vicgalle-roleplay-llama-3-8b_v8 status is now inactive due to auto deactivation removed underperforming models
admin requested tearing down of vicgalle-roleplay-llama-3-8b_v8
Running pipeline stage ISVCDeleter
Checking if service vicgalle-roleplay-llama-3-8b-v8 is running
Tearing down inference service vicgalle-roleplay-llama-3-8b-v8
Toredown service vicgalle-roleplay-llama-3-8b-v8
Pipeline stage ISVCDeleter completed in 3.33s
Running pipeline stage MKMLModelDeleter
Cleaning model data from S3
Cleaning model data from model cache
Deleting key vicgalle-roleplay-llama-3-8b-v8/config.json from bucket guanaco-mkml-models
Deleting key vicgalle-roleplay-llama-3-8b-v8/flywheel_model.0.safetensors from bucket guanaco-mkml-models
Deleting key vicgalle-roleplay-llama-3-8b-v8/special_tokens_map.json from bucket guanaco-mkml-models
Deleting key vicgalle-roleplay-llama-3-8b-v8/tokenizer.json from bucket guanaco-mkml-models
Deleting key vicgalle-roleplay-llama-3-8b-v8/tokenizer_config.json from bucket guanaco-mkml-models
Cleaning model data from model cache
Deleting key vicgalle-roleplay-llama-3-8b-v8_reward/config.json from bucket guanaco-reward-models
Deleting key vicgalle-roleplay-llama-3-8b-v8_reward/merges.txt from bucket guanaco-reward-models
Deleting key vicgalle-roleplay-llama-3-8b-v8_reward/reward.tensors from bucket guanaco-reward-models
Deleting key vicgalle-roleplay-llama-3-8b-v8_reward/special_tokens_map.json from bucket guanaco-reward-models
Deleting key vicgalle-roleplay-llama-3-8b-v8_reward/tokenizer.json from bucket guanaco-reward-models
Deleting key vicgalle-roleplay-llama-3-8b-v8_reward/tokenizer_config.json from bucket guanaco-reward-models
Deleting key vicgalle-roleplay-llama-3-8b-v8_reward/vocab.json from bucket guanaco-reward-models
Pipeline stage MKMLModelDeleter completed in 1.83s
vicgalle-roleplay-llama-3-8b_v8 status is now torndown due to DeploymentManager action

Usage Metrics

Latency Metrics