submission_id: undi95-redacted-e2_v1
developer_uid: Undi95
status: torndown
model_repo: Undi95/redacted-e2
reward_repo: ChaiML/reward_gpt2_medium_preference_24m_e2
generation_params: {'temperature': 0.8, 'top_p': 0.9, 'min_p': 0.0, 'top_k': 70, 'presence_penalty': 0.9, 'frequency_penalty': 0.9, 'stopping_words': ['\n', '<|eot_id|>'], '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 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{bot_name}'s persona: {memory}.<|eot_id|>", 'prompt_template': '{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_name}: {message}\n', 'user_template': '{user_name}: {message}\n', 'response_template': '{bot_name}:', 'truncate_by_message': False}
timestamp: 2024-04-21T00:03:23+00:00
model_name: undi95-redacted-e2_v1
model_eval_status: success
model_group: Undi95/redacted-e2
num_battles: 5730
num_wins: 2727
celo_rating: 1130.44
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: undi95-redacted-e2_v1
ineligible_reason: propriety_total_count < 800
language_model: Undi95/redacted-e2
model_size: 8B
reward_model: ChaiML/reward_gpt2_medium_preference_24m_e2
us_pacific_date: 2024-04-20
win_ratio: 0.47591623036649217
preference_data_url: None
Resubmit model
Running pipeline stage MKMLizer
Starting job with name undi95-redacted-e2-v1-mkmlizer
Waiting for job on undi95-redacted-e2-v1-mkmlizer to finish
undi95-redacted-e2-v1-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
undi95-redacted-e2-v1-mkmlizer: ║ _____ __ __ ║
undi95-redacted-e2-v1-mkmlizer: ║ / _/ /_ ___ __/ / ___ ___ / / ║
undi95-redacted-e2-v1-mkmlizer: ║ / _/ / // / |/|/ / _ \/ -_) -_) / ║
undi95-redacted-e2-v1-mkmlizer: ║ /_//_/\_, /|__,__/_//_/\__/\__/_/ ║
undi95-redacted-e2-v1-mkmlizer: ║ /___/ ║
undi95-redacted-e2-v1-mkmlizer: ║ ║
undi95-redacted-e2-v1-mkmlizer: ║ Version: 0.8.10 ║
undi95-redacted-e2-v1-mkmlizer: ║ Copyright 2023 MK ONE TECHNOLOGIES Inc. ║
undi95-redacted-e2-v1-mkmlizer: ║ ║
undi95-redacted-e2-v1-mkmlizer: ║ The license key for the current software has been verified as ║
undi95-redacted-e2-v1-mkmlizer: ║ belonging to: ║
undi95-redacted-e2-v1-mkmlizer: ║ ║
undi95-redacted-e2-v1-mkmlizer: ║ Chai Research Corp. ║
undi95-redacted-e2-v1-mkmlizer: ║ Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f ║
undi95-redacted-e2-v1-mkmlizer: ║ Expiration: 2024-07-15 23:59:59 ║
undi95-redacted-e2-v1-mkmlizer: ║ ║
undi95-redacted-e2-v1-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
undi95-redacted-e2-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.
undi95-redacted-e2-v1-mkmlizer: warnings.warn(warning_message, FutureWarning)
undi95-redacted-e2-v1-mkmlizer: Downloaded to shared memory in 33.687s
undi95-redacted-e2-v1-mkmlizer: quantizing model to /dev/shm/model_cache
undi95-redacted-e2-v1-mkmlizer: Saving flywheel model at /dev/shm/model_cache
undi95-redacted-e2-v1-mkmlizer: Loading 0: 0%| | 0/291 [00:00<?, ?it/s] Loading 0: 29%|██▊ | 83/291 [00:01<00:03, 66.32it/s] Loading 0: 64%|██████▍ | 187/291 [00:02<00:01, 77.37it/s] Loading 0: 97%|█████████▋| 281/291 [00:03<00:00, 82.91it/s] Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
undi95-redacted-e2-v1-mkmlizer: quantized model in 23.943s
undi95-redacted-e2-v1-mkmlizer: Processed model Undi95/redacted-e2 in 59.224s
undi95-redacted-e2-v1-mkmlizer: creating bucket guanaco-mkml-models
undi95-redacted-e2-v1-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
undi95-redacted-e2-v1-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/undi95-redacted-e2-v1
undi95-redacted-e2-v1-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/undi95-redacted-e2-v1/config.json
undi95-redacted-e2-v1-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/undi95-redacted-e2-v1/special_tokens_map.json
undi95-redacted-e2-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/undi95-redacted-e2-v1/tokenizer_config.json
undi95-redacted-e2-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/undi95-redacted-e2-v1/tokenizer.json
undi95-redacted-e2-v1-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/undi95-redacted-e2-v1/flywheel_model.0.safetensors
undi95-redacted-e2-v1-mkmlizer: loading reward model from ChaiML/reward_gpt2_medium_preference_24m_e2
undi95-redacted-e2-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.
undi95-redacted-e2-v1-mkmlizer: warnings.warn(
undi95-redacted-e2-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.
undi95-redacted-e2-v1-mkmlizer: warnings.warn(
undi95-redacted-e2-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.
undi95-redacted-e2-v1-mkmlizer: warnings.warn(
undi95-redacted-e2-v1-mkmlizer: Saving model to /tmp/reward_cache/reward.tensors
undi95-redacted-e2-v1-mkmlizer: Saving duration: 0.277s
undi95-redacted-e2-v1-mkmlizer: Processed model ChaiML/reward_gpt2_medium_preference_24m_e2 in 5.329s
undi95-redacted-e2-v1-mkmlizer: creating bucket guanaco-reward-models
undi95-redacted-e2-v1-mkmlizer: Bucket 's3://guanaco-reward-models/' created
undi95-redacted-e2-v1-mkmlizer: uploading /tmp/reward_cache to s3://guanaco-reward-models/undi95-redacted-e2-v1_reward
undi95-redacted-e2-v1-mkmlizer: cp /tmp/reward_cache/config.json s3://guanaco-reward-models/undi95-redacted-e2-v1_reward/config.json
undi95-redacted-e2-v1-mkmlizer: cp /tmp/reward_cache/special_tokens_map.json s3://guanaco-reward-models/undi95-redacted-e2-v1_reward/special_tokens_map.json
undi95-redacted-e2-v1-mkmlizer: cp /tmp/reward_cache/tokenizer_config.json s3://guanaco-reward-models/undi95-redacted-e2-v1_reward/tokenizer_config.json
undi95-redacted-e2-v1-mkmlizer: cp /tmp/reward_cache/vocab.json s3://guanaco-reward-models/undi95-redacted-e2-v1_reward/vocab.json
undi95-redacted-e2-v1-mkmlizer: cp /tmp/reward_cache/merges.txt s3://guanaco-reward-models/undi95-redacted-e2-v1_reward/merges.txt
undi95-redacted-e2-v1-mkmlizer: cp /tmp/reward_cache/tokenizer.json s3://guanaco-reward-models/undi95-redacted-e2-v1_reward/tokenizer.json
undi95-redacted-e2-v1-mkmlizer: cp /tmp/reward_cache/reward.tensors s3://guanaco-reward-models/undi95-redacted-e2-v1_reward/reward.tensors
Job undi95-redacted-e2-v1-mkmlizer completed after 84.04s with status: succeeded
Stopping job with name undi95-redacted-e2-v1-mkmlizer
Pipeline stage MKMLizer completed in 87.31s
Running pipeline stage MKMLKubeTemplater
Pipeline stage MKMLKubeTemplater completed in 0.29s
Running pipeline stage ISVCDeployer
Creating inference service undi95-redacted-e2-v1
Waiting for inference service undi95-redacted-e2-v1 to be ready
Inference service undi95-redacted-e2-v1 ready after 30.17183208465576s
Pipeline stage ISVCDeployer completed in 37.98s
Running pipeline stage StressChecker
Received healthy response to inference request in 1.7963440418243408s
Received healthy response to inference request in 0.9655933380126953s
Received healthy response to inference request in 0.9631044864654541s
Received healthy response to inference request in 1.2554564476013184s
Received healthy response to inference request in 1.011626958847046s
5 requests
0 failed requests
5th percentile: 0.9636022567749023
10th percentile: 0.9641000270843506
20th percentile: 0.9650955677032471
30th percentile: 0.9748000621795654
40th percentile: 0.9932135105133056
50th percentile: 1.011626958847046
60th percentile: 1.1091587543487549
70th percentile: 1.2066905498504639
80th percentile: 1.363633966445923
90th percentile: 1.5799890041351319
95th percentile: 1.6881665229797362
99th percentile: 1.77470853805542
mean time: 1.198425054550171
Pipeline stage StressChecker completed in 6.62s
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
undi95-redacted-e2_v1 status is now deployed due to DeploymentManager action
undi95-redacted-e2_v1 status is now inactive due to auto deactivation removed underperforming models
admin requested tearing down of undi95-redacted-e2_v1
Running pipeline stage ISVCDeleter
Checking if service undi95-redacted-e2-v1 is running
Tearing down inference service undi95-redacted-e2-v1
Toredown service undi95-redacted-e2-v1
Pipeline stage ISVCDeleter completed in 3.85s
Running pipeline stage MKMLModelDeleter
Cleaning model data from S3
Cleaning model data from model cache
Deleting key undi95-redacted-e2-v1/config.json from bucket guanaco-mkml-models
Deleting key undi95-redacted-e2-v1/flywheel_model.0.safetensors from bucket guanaco-mkml-models
Deleting key undi95-redacted-e2-v1/special_tokens_map.json from bucket guanaco-mkml-models
Deleting key undi95-redacted-e2-v1/tokenizer.json from bucket guanaco-mkml-models
Deleting key undi95-redacted-e2-v1/tokenizer_config.json from bucket guanaco-mkml-models
Cleaning model data from model cache
Deleting key undi95-redacted-e2-v1_reward/config.json from bucket guanaco-reward-models
Deleting key undi95-redacted-e2-v1_reward/merges.txt from bucket guanaco-reward-models
Deleting key undi95-redacted-e2-v1_reward/reward.tensors from bucket guanaco-reward-models
Deleting key undi95-redacted-e2-v1_reward/special_tokens_map.json from bucket guanaco-reward-models
Deleting key undi95-redacted-e2-v1_reward/tokenizer.json from bucket guanaco-reward-models
Deleting key undi95-redacted-e2-v1_reward/tokenizer_config.json from bucket guanaco-reward-models
Deleting key undi95-redacted-e2-v1_reward/vocab.json from bucket guanaco-reward-models
Pipeline stage MKMLModelDeleter completed in 1.98s
undi95-redacted-e2_v1 status is now torndown due to DeploymentManager action

Usage Metrics

Latency Metrics