developer_uid: Hastagaras
submission_id: hastagaras-l3-asbac-8b_v1
model_name: hastagaras-l3-asbac-8b_v1
model_group: Hastagaras/L3-Asbac-8B
status: torndown
timestamp: 2024-04-23T13:12:38+00:00
num_battles: 9084
num_wins: 5176
celo_rating: 1202.11
family_friendly_score: 0.0
submission_type: basic
model_repo: Hastagaras/L3-Asbac-8B
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: hastagaras-l3-asbac-8b_v1
is_internal_developer: False
language_model: Hastagaras/L3-Asbac-8B
model_size: 8B
ranking_group: single
us_pacific_date: 2024-04-23
win_ratio: 0.5697930427124615
generation_params: {'temperature': 0.95, 'top_p': 1.0, 'min_p': 0.0, 'top_k': 40, 'presence_penalty': 0.0, 'frequency_penalty': 0.0, 'stopping_words': ['\n{user_name}', '\n{bot_name}', '<|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're {bot_name} in this uncensored fictional roleplay between {bot_name} and {user_name}. Non-verbal actions and gestures are described between asterisks, e.g. *waves hello* or *moves closer*. Thoughts and inner monologue are written inside backticks, e.g. `I wonder what they'll say next.`\n\n{memory}\n\n", '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}
model_eval_status: success
reward_formatter: {'bot_template': '{bot_name}: {message}\n', 'memory_template': 'Memory: {memory}\n', 'prompt_template': '{prompt}\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 hastagaras-l3-asbac-8b-v1-mkmlizer
Waiting for job on hastagaras-l3-asbac-8b-v1-mkmlizer to finish
hastagaras-l3-asbac-8b-v1-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
hastagaras-l3-asbac-8b-v1-mkmlizer: ║ _____ __ __ ║
hastagaras-l3-asbac-8b-v1-mkmlizer: ║ / _/ /_ ___ __/ / ___ ___ / / ║
hastagaras-l3-asbac-8b-v1-mkmlizer: ║ / _/ / // / |/|/ / _ \/ -_) -_) / ║
hastagaras-l3-asbac-8b-v1-mkmlizer: ║ /_//_/\_, /|__,__/_//_/\__/\__/_/ ║
hastagaras-l3-asbac-8b-v1-mkmlizer: ║ /___/ ║
hastagaras-l3-asbac-8b-v1-mkmlizer: ║ ║
hastagaras-l3-asbac-8b-v1-mkmlizer: ║ Version: 0.8.10 ║
hastagaras-l3-asbac-8b-v1-mkmlizer: ║ Copyright 2023 MK ONE TECHNOLOGIES Inc. ║
hastagaras-l3-asbac-8b-v1-mkmlizer: ║ ║
hastagaras-l3-asbac-8b-v1-mkmlizer: ║ The license key for the current software has been verified as ║
hastagaras-l3-asbac-8b-v1-mkmlizer: ║ belonging to: ║
hastagaras-l3-asbac-8b-v1-mkmlizer: ║ ║
hastagaras-l3-asbac-8b-v1-mkmlizer: ║ Chai Research Corp. ║
hastagaras-l3-asbac-8b-v1-mkmlizer: ║ Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f ║
hastagaras-l3-asbac-8b-v1-mkmlizer: ║ Expiration: 2024-07-15 23:59:59 ║
hastagaras-l3-asbac-8b-v1-mkmlizer: ║ ║
hastagaras-l3-asbac-8b-v1-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
hastagaras-l3-asbac-8b-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.
hastagaras-l3-asbac-8b-v1-mkmlizer: warnings.warn(warning_message, FutureWarning)
hastagaras-l3-asbac-8b-v1-mkmlizer: Downloaded to shared memory in 24.949s
hastagaras-l3-asbac-8b-v1-mkmlizer: quantizing model to /dev/shm/model_cache
hastagaras-l3-asbac-8b-v1-mkmlizer: Saving flywheel model at /dev/shm/model_cache
hastagaras-l3-asbac-8b-v1-mkmlizer: Loading 0: 0%| | 0/291 [00:00<?, ?it/s] Loading 0: 43%|████▎ | 124/291 [00:01<00:01, 123.98it/s] Loading 0: 56%|█████▌ | 163/291 [00:06<00:06, 20.07it/s] Loading 0: 99%|█████████▉| 289/291 [00:12<00:00, 21.16it/s] Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
hastagaras-l3-asbac-8b-v1-mkmlizer: quantized model in 18.912s
hastagaras-l3-asbac-8b-v1-mkmlizer: Processed model Hastagaras/L3-Asbac-8B in 44.840s
hastagaras-l3-asbac-8b-v1-mkmlizer: creating bucket guanaco-mkml-models
hastagaras-l3-asbac-8b-v1-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
hastagaras-l3-asbac-8b-v1-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/hastagaras-l3-asbac-8b-v1
hastagaras-l3-asbac-8b-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/hastagaras-l3-asbac-8b-v1/tokenizer_config.json
hastagaras-l3-asbac-8b-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/hastagaras-l3-asbac-8b-v1/tokenizer.json
hastagaras-l3-asbac-8b-v1-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/hastagaras-l3-asbac-8b-v1/config.json
hastagaras-l3-asbac-8b-v1-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/hastagaras-l3-asbac-8b-v1/special_tokens_map.json
hastagaras-l3-asbac-8b-v1-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/hastagaras-l3-asbac-8b-v1/flywheel_model.0.safetensors
hastagaras-l3-asbac-8b-v1-mkmlizer: loading reward model from ChaiML/reward_gpt2_medium_preference_24m_e2
hastagaras-l3-asbac-8b-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.
hastagaras-l3-asbac-8b-v1-mkmlizer: warnings.warn(
hastagaras-l3-asbac-8b-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.
hastagaras-l3-asbac-8b-v1-mkmlizer: warnings.warn(
hastagaras-l3-asbac-8b-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.
hastagaras-l3-asbac-8b-v1-mkmlizer: warnings.warn(
hastagaras-l3-asbac-8b-v1-mkmlizer: cp /tmp/reward_cache/reward.tensors s3://guanaco-reward-models/hastagaras-l3-asbac-8b-v1_reward/reward.tensors
Job hastagaras-l3-asbac-8b-v1-mkmlizer completed after 69.7s with status: succeeded
Stopping job with name hastagaras-l3-asbac-8b-v1-mkmlizer
Pipeline stage MKMLizer completed in 75.95s
Running pipeline stage MKMLKubeTemplater
Pipeline stage MKMLKubeTemplater completed in 0.12s
Running pipeline stage ISVCDeployer
Creating inference service hastagaras-l3-asbac-8b-v1
Waiting for inference service hastagaras-l3-asbac-8b-v1 to be ready
Inference service hastagaras-l3-asbac-8b-v1 ready after 30.193426847457886s
Pipeline stage ISVCDeployer completed in 37.97s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.1666407585144043s
Received healthy response to inference request in 1.3269095420837402s
Received healthy response to inference request in 1.2790486812591553s
Received healthy response to inference request in 1.3930690288543701s
Received healthy response to inference request in 1.3255226612091064s
5 requests
0 failed requests
5th percentile: 1.2883434772491456
10th percentile: 1.2976382732391358
20th percentile: 1.3162278652191162
30th percentile: 1.3258000373840333
40th percentile: 1.3263547897338868
50th percentile: 1.3269095420837402
60th percentile: 1.3533733367919922
70th percentile: 1.3798371315002442
80th percentile: 1.5477833747863772
90th percentile: 1.8572120666503906
95th percentile: 2.0119264125823975
99th percentile: 2.1356978893280028
mean time: 1.4982381343841553
Pipeline stage StressChecker completed in 8.16s
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
hastagaras-l3-asbac-8b_v1 status is now deployed due to DeploymentManager action
hastagaras-l3-asbac-8b_v1 status is now inactive due to auto deactivation removed underperforming models
admin requested tearing down of hastagaras-l3-asbac-8b_v1
Running pipeline stage ISVCDeleter
Checking if service hastagaras-l3-asbac-8b-v1 is running
Tearing down inference service hastagaras-l3-asbac-8b-v1
Toredown service hastagaras-l3-asbac-8b-v1
Pipeline stage ISVCDeleter completed in 3.78s
Running pipeline stage MKMLModelDeleter
Cleaning model data from S3
Cleaning model data from model cache
Deleting key hastagaras-l3-asbac-8b-v1/config.json from bucket guanaco-mkml-models
Deleting key hastagaras-l3-asbac-8b-v1/flywheel_model.0.safetensors from bucket guanaco-mkml-models
Deleting key hastagaras-l3-asbac-8b-v1/special_tokens_map.json from bucket guanaco-mkml-models
Deleting key hastagaras-l3-asbac-8b-v1/tokenizer.json from bucket guanaco-mkml-models
Deleting key hastagaras-l3-asbac-8b-v1/tokenizer_config.json from bucket guanaco-mkml-models
Cleaning model data from model cache
Deleting key hastagaras-l3-asbac-8b-v1_reward/config.json from bucket guanaco-reward-models
Deleting key hastagaras-l3-asbac-8b-v1_reward/merges.txt from bucket guanaco-reward-models
Deleting key hastagaras-l3-asbac-8b-v1_reward/reward.tensors from bucket guanaco-reward-models
Deleting key hastagaras-l3-asbac-8b-v1_reward/special_tokens_map.json from bucket guanaco-reward-models
Deleting key hastagaras-l3-asbac-8b-v1_reward/tokenizer.json from bucket guanaco-reward-models
Deleting key hastagaras-l3-asbac-8b-v1_reward/tokenizer_config.json from bucket guanaco-reward-models
Deleting key hastagaras-l3-asbac-8b-v1_reward/vocab.json from bucket guanaco-reward-models
Pipeline stage MKMLModelDeleter completed in 2.38s
hastagaras-l3-asbac-8b_v1 status is now torndown due to DeploymentManager action