developer_uid: Meliodia
submission_id: nousresearch-meta-llama_4941_v51
model_name: NousL3-8B-Instruct
model_group: NousResearch/Meta-Llama-
status: torndown
timestamp: 2024-04-29T20:33:47+00:00
num_battles: 6742
num_wins: 3466
celo_rating: 1166.95
family_friendly_score: 0.0
submission_type: basic
model_repo: NousResearch/Meta-Llama-3-8B-Instruct
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: NousL3-8B-Instruct
is_internal_developer: True
language_model: NousResearch/Meta-Llama-3-8B-Instruct
model_size: 8B
ranking_group: single
us_pacific_date: 2024-04-29
win_ratio: 0.514090774250964
generation_params: {'temperature': 1.0, 'top_p': 1.0, 'min_p': 0.0, 'top_k': 100, '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': "{bot_name}'s Persona: {memory}\n####\n", 'prompt_template': '{prompt}\n<START>\n', 'bot_template': '{bot_name}: {message}\n', 'user_template': '{user_name}: {message}\n', 'response_template': '{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 nousresearch-meta-llama-4941-v51-mkmlizer
Waiting for job on nousresearch-meta-llama-4941-v51-mkmlizer to finish
Stopping job with name nousresearch-meta-llama-4941-v51-mkmlizer
%s, retrying in %s seconds...
Starting job with name nousresearch-meta-llama-4941-v51-mkmlizer
Waiting for job on nousresearch-meta-llama-4941-v51-mkmlizer to finish
nousresearch-meta-llama-4941-v51-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
nousresearch-meta-llama-4941-v51-mkmlizer: ║     _____            __           __                                ║
nousresearch-meta-llama-4941-v51-mkmlizer: ║    / _/ /_ ___    __/ /  ___ ___ / /                                ║
nousresearch-meta-llama-4941-v51-mkmlizer: ║   / _/ / // / |/|/ / _ \/ -_) -_) /                                 ║
nousresearch-meta-llama-4941-v51-mkmlizer: ║  /_//_/\_, /|__,__/_//_/\__/\__/_/                                  ║
nousresearch-meta-llama-4941-v51-mkmlizer: ║       /___/                                                         ║
nousresearch-meta-llama-4941-v51-mkmlizer: ║                                                                     ║
nousresearch-meta-llama-4941-v51-mkmlizer: ║  Version: 0.8.10                                                    ║
nousresearch-meta-llama-4941-v51-mkmlizer: ║  Copyright 2023 MK ONE TECHNOLOGIES Inc.                            ║
nousresearch-meta-llama-4941-v51-mkmlizer: ║                                                                     ║
nousresearch-meta-llama-4941-v51-mkmlizer: ║  The license key for the current software has been verified as      ║
nousresearch-meta-llama-4941-v51-mkmlizer: ║  belonging to:                                                      ║
nousresearch-meta-llama-4941-v51-mkmlizer: ║                                                                     ║
nousresearch-meta-llama-4941-v51-mkmlizer: ║  Chai Research Corp.                                                ║
nousresearch-meta-llama-4941-v51-mkmlizer: ║  Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f                   ║
nousresearch-meta-llama-4941-v51-mkmlizer: ║  Expiration: 2024-07-15 23:59:59                                    ║
nousresearch-meta-llama-4941-v51-mkmlizer: ║                                                                     ║
nousresearch-meta-llama-4941-v51-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
nousresearch-meta-llama-4941-v51-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.
nousresearch-meta-llama-4941-v51-mkmlizer:   warnings.warn(warning_message, FutureWarning)
nousresearch-meta-llama-4941-v51-mkmlizer: Downloaded to shared memory in 17.804s
nousresearch-meta-llama-4941-v51-mkmlizer: quantizing model to /dev/shm/model_cache
nousresearch-meta-llama-4941-v51-mkmlizer: Saving flywheel model at /dev/shm/model_cache
nousresearch-meta-llama-4941-v51-mkmlizer: 
Loading 0:   0%|          | 0/291 [00:00<?, ?it/s]
Loading 0:  42%|████▏     | 121/291 [00:01<00:01, 117.55it/s]
Loading 0:  81%|████████▏ | 237/291 [00:02<00:00, 115.93it/s]
Loading 0:  99%|█████████▊| 287/291 [00:07<00:00, 29.64it/s] 
                                                            
Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
nousresearch-meta-llama-4941-v51-mkmlizer: quantized model in 17.918s
nousresearch-meta-llama-4941-v51-mkmlizer: Processed model NousResearch/Meta-Llama-3-8B-Instruct in 36.716s
nousresearch-meta-llama-4941-v51-mkmlizer: creating bucket guanaco-mkml-models
nousresearch-meta-llama-4941-v51-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
nousresearch-meta-llama-4941-v51-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/nousresearch-meta-llama-4941-v51
nousresearch-meta-llama-4941-v51-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/nousresearch-meta-llama-4941-v51/config.json
nousresearch-meta-llama-4941-v51-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/nousresearch-meta-llama-4941-v51/special_tokens_map.json
nousresearch-meta-llama-4941-v51-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/nousresearch-meta-llama-4941-v51/tokenizer_config.json
nousresearch-meta-llama-4941-v51-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/nousresearch-meta-llama-4941-v51/tokenizer.json
nousresearch-meta-llama-4941-v51-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/nousresearch-meta-llama-4941-v51/flywheel_model.0.safetensors
nousresearch-meta-llama-4941-v51-mkmlizer: loading reward model from ChaiML/reward_gpt2_medium_preference_24m_e2
nousresearch-meta-llama-4941-v51-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.
nousresearch-meta-llama-4941-v51-mkmlizer:   warnings.warn(
nousresearch-meta-llama-4941-v51-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.
nousresearch-meta-llama-4941-v51-mkmlizer:   warnings.warn(
nousresearch-meta-llama-4941-v51-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.
nousresearch-meta-llama-4941-v51-mkmlizer:   warnings.warn(
nousresearch-meta-llama-4941-v51-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()
nousresearch-meta-llama-4941-v51-mkmlizer:   return self.fget.__get__(instance, owner)()
nousresearch-meta-llama-4941-v51-mkmlizer: creating bucket guanaco-reward-models
nousresearch-meta-llama-4941-v51-mkmlizer: Bucket 's3://guanaco-reward-models/' created
nousresearch-meta-llama-4941-v51-mkmlizer: uploading /tmp/reward_cache to s3://guanaco-reward-models/nousresearch-meta-llama-4941-v51_reward
nousresearch-meta-llama-4941-v51-mkmlizer: cp /tmp/reward_cache/tokenizer_config.json s3://guanaco-reward-models/nousresearch-meta-llama-4941-v51_reward/tokenizer_config.json
nousresearch-meta-llama-4941-v51-mkmlizer: cp /tmp/reward_cache/config.json s3://guanaco-reward-models/nousresearch-meta-llama-4941-v51_reward/config.json
nousresearch-meta-llama-4941-v51-mkmlizer: cp /tmp/reward_cache/special_tokens_map.json s3://guanaco-reward-models/nousresearch-meta-llama-4941-v51_reward/special_tokens_map.json
nousresearch-meta-llama-4941-v51-mkmlizer: cp /tmp/reward_cache/vocab.json s3://guanaco-reward-models/nousresearch-meta-llama-4941-v51_reward/vocab.json
nousresearch-meta-llama-4941-v51-mkmlizer: cp /tmp/reward_cache/merges.txt s3://guanaco-reward-models/nousresearch-meta-llama-4941-v51_reward/merges.txt
nousresearch-meta-llama-4941-v51-mkmlizer: cp /tmp/reward_cache/tokenizer.json s3://guanaco-reward-models/nousresearch-meta-llama-4941-v51_reward/tokenizer.json
nousresearch-meta-llama-4941-v51-mkmlizer: cp /tmp/reward_cache/reward.tensors s3://guanaco-reward-models/nousresearch-meta-llama-4941-v51_reward/reward.tensors
Job nousresearch-meta-llama-4941-v51-mkmlizer completed after 63.63s with status: succeeded
Stopping job with name nousresearch-meta-llama-4941-v51-mkmlizer
Pipeline stage MKMLizer completed in 67.14s
Running pipeline stage MKMLKubeTemplater
Pipeline stage MKMLKubeTemplater completed in 0.09s
Running pipeline stage ISVCDeployer
Creating inference service nousresearch-meta-llama-4941-v51
Waiting for inference service nousresearch-meta-llama-4941-v51 to be ready
Inference service nousresearch-meta-llama-4941-v51 ready after 30.194157123565674s
Pipeline stage ISVCDeployer completed in 37.85s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.1124494075775146s
Received healthy response to inference request in 1.27797269821167s
Received healthy response to inference request in 1.35054349899292s
Received healthy response to inference request in 1.2375454902648926s
Received healthy response to inference request in 1.2918245792388916s
5 requests
0 failed requests
5th percentile: 1.245630931854248
10th percentile: 1.2537163734436034
20th percentile: 1.2698872566223145
30th percentile: 1.2807430744171142
40th percentile: 1.286283826828003
50th percentile: 1.2918245792388916
60th percentile: 1.315312147140503
70th percentile: 1.3387997150421143
80th percentile: 1.502924680709839
90th percentile: 1.8076870441436768
95th percentile: 1.9600682258605955
99th percentile: 2.081973171234131
mean time: 1.4540671348571776
Pipeline stage StressChecker completed in 7.91s
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
nousresearch-meta-llama_4941_v51 status is now deployed due to DeploymentManager action
nousresearch-meta-llama_4941_v51 status is now inactive due to auto deactivation removed underperforming models
admin requested tearing down of nousresearch-meta-llama_4941_v51
Running pipeline stage ISVCDeleter
Checking if service nousresearch-meta-llama-4941-v51 is running
Tearing down inference service nousresearch-meta-llama-4941-v49
Toredown service nousresearch-meta-llama-4941-v49
Pipeline stage ISVCDeleter completed in 6.46s
Running pipeline stage MKMLModelDeleter
Cleaning model data from S3
Tearing down inference service nousresearch-meta-llama-4941-v51
Cleaning model data from model cache
Toredown service nousresearch-meta-llama-4941-v51
Deleting key nousresearch-meta-llama-4941-v49/config.json from bucket guanaco-mkml-models
Pipeline stage ISVCDeleter completed in 6.82s
Deleting key nousresearch-meta-llama-4941-v49/flywheel_model.0.safetensors from bucket guanaco-mkml-models
Running pipeline stage MKMLModelDeleter
Cleaning model data from S3
Cleaning model data from model cache
Deleting key nousresearch-meta-llama-4941-v51/config.json from bucket guanaco-mkml-models
Deleting key nousresearch-meta-llama-4941-v49/special_tokens_map.json from bucket guanaco-mkml-models
Deleting key nousresearch-meta-llama-4941-v51/flywheel_model.0.safetensors from bucket guanaco-mkml-models
Deleting key nousresearch-meta-llama-4941-v49/tokenizer.json from bucket guanaco-mkml-models
Deleting key nousresearch-meta-llama-4941-v49/tokenizer_config.json from bucket guanaco-mkml-models
Cleaning model data from model cache
Deleting key nousresearch-meta-llama-4941-v49_reward/config.json from bucket guanaco-reward-models
Deleting key nousresearch-meta-llama-4941-v51/special_tokens_map.json from bucket guanaco-mkml-models
Deleting key nousresearch-meta-llama-4941-v49_reward/merges.txt from bucket guanaco-reward-models
Deleting key nousresearch-meta-llama-4941-v51/tokenizer.json from bucket guanaco-mkml-models
Deleting key nousresearch-meta-llama-4941-v49_reward/reward.tensors from bucket guanaco-reward-models
Deleting key nousresearch-meta-llama-4941-v51/tokenizer_config.json from bucket guanaco-mkml-models
Deleting key nousresearch-meta-llama-4941-v49_reward/special_tokens_map.json from bucket guanaco-reward-models
Cleaning model data from model cache
Deleting key nousresearch-meta-llama-4941-v49_reward/tokenizer.json from bucket guanaco-reward-models
Deleting key nousresearch-meta-llama-4941-v51_reward/config.json from bucket guanaco-reward-models
Deleting key nousresearch-meta-llama-4941-v49_reward/tokenizer_config.json from bucket guanaco-reward-models
Deleting key nousresearch-meta-llama-4941-v51_reward/merges.txt from bucket guanaco-reward-models
Deleting key nousresearch-meta-llama-4941-v49_reward/vocab.json from bucket guanaco-reward-models
Deleting key nousresearch-meta-llama-4941-v51_reward/reward.tensors from bucket guanaco-reward-models
Pipeline stage MKMLModelDeleter completed in 3.43s
Deleting key nousresearch-meta-llama-4941-v51_reward/special_tokens_map.json from bucket guanaco-reward-models
nousresearch-meta-llama_4941_v49 status is now torndown due to DeploymentManager action
Deleting key nousresearch-meta-llama-4941-v51_reward/tokenizer_config.json from bucket guanaco-reward-models
Deleting key nousresearch-meta-llama-4941-v51_reward/vocab.json from bucket guanaco-reward-models
Pipeline stage MKMLModelDeleter completed in 2.69s
nousresearch-meta-llama_4941_v51 status is now torndown due to DeploymentManager action