submission_id: undi95-llama-3-unholy-8b_v3
developer_uid: Undi95
status: torndown
model_repo: Undi95/Llama-3-Unholy-8B
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': 8, 'max_output_tokens': 64}
formatter: {'memory_template': '<|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\nThis is an entertaining conversation. You are {bot_name} who has the persona: {memory}.\nEngage in a chat with {user_name} while staying in character. Try to flirt with {user_name}. Engage in *roleplay* actions. Describe the scene dramatically.', '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:', 'truncate_by_message': False}
timestamp: 2024-04-20T13:14:47+00:00
model_name: undi95-llama-3-unholy-8b_v2
model_eval_status: success
model_group: Undi95/Llama-3-Unholy-8B
num_battles: 6415
num_wins: 3283
celo_rating: 1157.97
propriety_score: 0.0
propriety_total_count: 0.0
submission_type: basic
model_architecture: LlamaForCausalLM
model_num_parameters: 8030261248.0
best_of: 8
max_input_tokens: 512
max_output_tokens: 64
display_name: undi95-llama-3-unholy-8b_v2
ineligible_reason: propriety_total_count < 800
language_model: Undi95/Llama-3-Unholy-8B
model_size: 8B
reward_model: ChaiML/reward_gpt2_medium_preference_24m_e2
us_pacific_date: 2024-04-20
win_ratio: 0.5117692907248635
preference_data_url: None
Resubmit model
Running pipeline stage MKMLizer
Starting job with name undi95-llama-3-unholy-8b-v3-mkmlizer
Waiting for job on undi95-llama-3-unholy-8b-v3-mkmlizer to finish
Stopping job with name undi95-llama-3-unholy-8b-v3-mkmlizer
%s, retrying in %s seconds...
Starting job with name undi95-llama-3-unholy-8b-v3-mkmlizer
Waiting for job on undi95-llama-3-unholy-8b-v3-mkmlizer to finish
Stopping job with name undi95-llama-3-unholy-8b-v3-mkmlizer
%s, retrying in %s seconds...
Starting job with name undi95-llama-3-unholy-8b-v3-mkmlizer
Waiting for job on undi95-llama-3-unholy-8b-v3-mkmlizer to finish
undi95-llama-3-unholy-8b-v3-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
undi95-llama-3-unholy-8b-v3-mkmlizer: ║ _____ __ __ ║
undi95-llama-3-unholy-8b-v3-mkmlizer: ║ / _/ /_ ___ __/ / ___ ___ / / ║
undi95-llama-3-unholy-8b-v3-mkmlizer: ║ / _/ / // / |/|/ / _ \/ -_) -_) / ║
undi95-llama-3-unholy-8b-v3-mkmlizer: ║ /_//_/\_, /|__,__/_//_/\__/\__/_/ ║
undi95-llama-3-unholy-8b-v3-mkmlizer: ║ /___/ ║
undi95-llama-3-unholy-8b-v3-mkmlizer: ║ ║
undi95-llama-3-unholy-8b-v3-mkmlizer: ║ Version: 0.8.10 ║
undi95-llama-3-unholy-8b-v3-mkmlizer: ║ Copyright 2023 MK ONE TECHNOLOGIES Inc. ║
undi95-llama-3-unholy-8b-v3-mkmlizer: ║ ║
undi95-llama-3-unholy-8b-v3-mkmlizer: ║ The license key for the current software has been verified as ║
undi95-llama-3-unholy-8b-v3-mkmlizer: ║ belonging to: ║
undi95-llama-3-unholy-8b-v3-mkmlizer: ║ ║
undi95-llama-3-unholy-8b-v3-mkmlizer: ║ Chai Research Corp. ║
undi95-llama-3-unholy-8b-v3-mkmlizer: ║ Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f ║
undi95-llama-3-unholy-8b-v3-mkmlizer: ║ Expiration: 2024-07-15 23:59:59 ║
undi95-llama-3-unholy-8b-v3-mkmlizer: ║ ║
undi95-llama-3-unholy-8b-v3-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
undi95-llama-3-unholy-8b-v3-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-llama-3-unholy-8b-v3-mkmlizer: warnings.warn(warning_message, FutureWarning)
undi95-llama-3-unholy-8b-v3-mkmlizer: Downloaded to shared memory in 19.463s
undi95-llama-3-unholy-8b-v3-mkmlizer: quantizing model to /dev/shm/model_cache
undi95-llama-3-unholy-8b-v3-mkmlizer: Saving flywheel model at /dev/shm/model_cache
undi95-llama-3-unholy-8b-v3-mkmlizer: Loading 0: 0%| | 0/291 [00:00<?, ?it/s] Loading 0: 20%|██ | 59/291 [00:01<00:03, 58.22it/s] Loading 0: 33%|███▎ | 95/291 [00:02<00:04, 45.11it/s] Loading 0: 57%|█████▋ | 166/291 [00:03<00:02, 56.78it/s] Loading 0: 73%|███████▎ | 212/291 [00:04<00:01, 52.40it/s] Loading 0: 99%|█████████▊| 287/291 [00:10<00:00, 20.24it/s] Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
undi95-llama-3-unholy-8b-v3-mkmlizer: quantized model in 22.344s
undi95-llama-3-unholy-8b-v3-mkmlizer: Processed model Undi95/Llama-3-Unholy-8B in 43.526s
undi95-llama-3-unholy-8b-v3-mkmlizer: creating bucket guanaco-mkml-models
undi95-llama-3-unholy-8b-v3-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
undi95-llama-3-unholy-8b-v3-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/undi95-llama-3-unholy-8b-v3
undi95-llama-3-unholy-8b-v3-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/undi95-llama-3-unholy-8b-v3/tokenizer_config.json
undi95-llama-3-unholy-8b-v3-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/undi95-llama-3-unholy-8b-v3/special_tokens_map.json
undi95-llama-3-unholy-8b-v3-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/undi95-llama-3-unholy-8b-v3/config.json
undi95-llama-3-unholy-8b-v3-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/undi95-llama-3-unholy-8b-v3/tokenizer.json
undi95-llama-3-unholy-8b-v3-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/undi95-llama-3-unholy-8b-v3/flywheel_model.0.safetensors
undi95-llama-3-unholy-8b-v3-mkmlizer: loading reward model from ChaiML/reward_gpt2_medium_preference_24m_e2
undi95-llama-3-unholy-8b-v3-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-llama-3-unholy-8b-v3-mkmlizer: warnings.warn(
undi95-llama-3-unholy-8b-v3-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-llama-3-unholy-8b-v3-mkmlizer: warnings.warn(
undi95-llama-3-unholy-8b-v3-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-llama-3-unholy-8b-v3-mkmlizer: warnings.warn(
undi95-llama-3-unholy-8b-v3-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()
undi95-llama-3-unholy-8b-v3-mkmlizer: return self.fget.__get__(instance, owner)()
undi95-llama-3-unholy-8b-v3-mkmlizer: Saving model to /tmp/reward_cache/reward.tensors
undi95-llama-3-unholy-8b-v3-mkmlizer: Saving duration: 0.313s
undi95-llama-3-unholy-8b-v3-mkmlizer: Processed model ChaiML/reward_gpt2_medium_preference_24m_e2 in 7.528s
undi95-llama-3-unholy-8b-v3-mkmlizer: creating bucket guanaco-reward-models
undi95-llama-3-unholy-8b-v3-mkmlizer: Bucket 's3://guanaco-reward-models/' created
undi95-llama-3-unholy-8b-v3-mkmlizer: uploading /tmp/reward_cache to s3://guanaco-reward-models/undi95-llama-3-unholy-8b-v3_reward
undi95-llama-3-unholy-8b-v3-mkmlizer: cp /tmp/reward_cache/config.json s3://guanaco-reward-models/undi95-llama-3-unholy-8b-v3_reward/config.json
undi95-llama-3-unholy-8b-v3-mkmlizer: cp /tmp/reward_cache/tokenizer_config.json s3://guanaco-reward-models/undi95-llama-3-unholy-8b-v3_reward/tokenizer_config.json
undi95-llama-3-unholy-8b-v3-mkmlizer: cp /tmp/reward_cache/special_tokens_map.json s3://guanaco-reward-models/undi95-llama-3-unholy-8b-v3_reward/special_tokens_map.json
undi95-llama-3-unholy-8b-v3-mkmlizer: cp /tmp/reward_cache/merges.txt s3://guanaco-reward-models/undi95-llama-3-unholy-8b-v3_reward/merges.txt
undi95-llama-3-unholy-8b-v3-mkmlizer: cp /tmp/reward_cache/vocab.json s3://guanaco-reward-models/undi95-llama-3-unholy-8b-v3_reward/vocab.json
undi95-llama-3-unholy-8b-v3-mkmlizer: cp /tmp/reward_cache/tokenizer.json s3://guanaco-reward-models/undi95-llama-3-unholy-8b-v3_reward/tokenizer.json
undi95-llama-3-unholy-8b-v3-mkmlizer: cp /tmp/reward_cache/reward.tensors s3://guanaco-reward-models/undi95-llama-3-unholy-8b-v3_reward/reward.tensors
Job undi95-llama-3-unholy-8b-v3-mkmlizer completed after 79.67s with status: succeeded
Stopping job with name undi95-llama-3-unholy-8b-v3-mkmlizer
Pipeline stage MKMLizer completed in 89.44s
Running pipeline stage MKMLKubeTemplater
Pipeline stage MKMLKubeTemplater completed in 0.22s
Running pipeline stage ISVCDeployer
Creating inference service undi95-llama-3-unholy-8b-v3
Waiting for inference service undi95-llama-3-unholy-8b-v3 to be ready
Inference service undi95-llama-3-unholy-8b-v3 ready after 30.190364837646484s
Pipeline stage ISVCDeployer completed in 40.10s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.097435712814331s
Received healthy response to inference request in 1.228832483291626s
Received healthy response to inference request in 1.201415777206421s
Received healthy response to inference request in 1.1879832744598389s
Received healthy response to inference request in 0.6416189670562744s
5 requests
0 failed requests
5th percentile: 0.7508918285369873
10th percentile: 0.8601646900177002
20th percentile: 1.078710412979126
30th percentile: 1.1906697750091553
40th percentile: 1.196042776107788
50th percentile: 1.201415777206421
60th percentile: 1.2123824596405028
70th percentile: 1.223349142074585
80th percentile: 1.4025531291961673
90th percentile: 1.7499944210052492
95th percentile: 1.9237150669097898
99th percentile: 2.062691583633423
mean time: 1.2714572429656983
Pipeline stage StressChecker completed in 8.50s
Running pipeline stage DaemonicModelEvalScorer
Pipeline stage DaemonicModelEvalScorer completed in 0.03s
Running M-Eval for topic stay_in_character
Running pipeline stage DaemonicSafetyScorer
M-Eval Dataset for topic stay_in_character is loaded
Pipeline stage DaemonicSafetyScorer completed in 0.09s
undi95-llama-3-unholy-8b_v3 status is now deployed due to DeploymentManager action
Scoring model output for bot %s
Scoring model output for bot %s
Scoring model output for bot %s
Scoring model output for bot %s
undi95-llama-3-unholy-8b_v3 status is now inactive due to auto deactivation removed underperforming models
admin requested tearing down of undi95-llama-3-unholy-8b_v3
Running pipeline stage ISVCDeleter
Checking if service undi95-llama-3-unholy-8b-v3 is running
Tearing down inference service undi95-llama-3-unholy-8b-v3
Toredown service undi95-llama-3-unholy-8b-v3
Pipeline stage ISVCDeleter completed in 3.56s
Running pipeline stage MKMLModelDeleter
Cleaning model data from S3
Cleaning model data from model cache
Deleting key undi95-llama-3-unholy-8b-v3/config.json from bucket guanaco-mkml-models
Deleting key undi95-llama-3-unholy-8b-v3/flywheel_model.0.safetensors from bucket guanaco-mkml-models
Deleting key undi95-llama-3-unholy-8b-v3/special_tokens_map.json from bucket guanaco-mkml-models
Deleting key undi95-llama-3-unholy-8b-v3/tokenizer.json from bucket guanaco-mkml-models
Deleting key undi95-llama-3-unholy-8b-v3/tokenizer_config.json from bucket guanaco-mkml-models
Cleaning model data from model cache
Deleting key undi95-llama-3-unholy-8b-v3_reward/config.json from bucket guanaco-reward-models
Deleting key undi95-llama-3-unholy-8b-v3_reward/merges.txt from bucket guanaco-reward-models
Deleting key undi95-llama-3-unholy-8b-v3_reward/reward.tensors from bucket guanaco-reward-models
Deleting key undi95-llama-3-unholy-8b-v3_reward/special_tokens_map.json from bucket guanaco-reward-models
Deleting key undi95-llama-3-unholy-8b-v3_reward/tokenizer.json from bucket guanaco-reward-models
Deleting key undi95-llama-3-unholy-8b-v3_reward/tokenizer_config.json from bucket guanaco-reward-models
Deleting key undi95-llama-3-unholy-8b-v3_reward/vocab.json from bucket guanaco-reward-models
Pipeline stage MKMLModelDeleter completed in 2.26s
undi95-llama-3-unholy-8b_v3 status is now torndown due to DeploymentManager action

Usage Metrics

Latency Metrics