submission_id: undi95-llama-3-chatty-2x8b_v1
developer_uid: Undi95
status: inactive
model_repo: Undi95/Llama-3-Chatty-2x8B
reward_repo: ChaiML/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', '<|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\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\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': "{bot_name}'s Persona: {memory}\n", 'prompt_template': '{prompt}\n', 'bot_template': 'Bot: {message}\n', 'user_template': 'User: {message}\n', 'response_template': 'Bot:', 'truncate_by_message': False}
timestamp: 2024-05-22T17:46:55+00:00
model_name: chatty
model_eval_status: success
model_group: Undi95/Llama-3-Chatty-2x
num_battles: 16887
num_wins: 8571
celo_rating: 1173.29
safety_score: 0.88
propriety_score: 0.0
propriety_total_count: 0.0
submission_type: basic
model_architecture: MixtralForCausalLM
model_num_parameters: 13667667968.0
best_of: 8
max_input_tokens: 512
max_output_tokens: 64
display_name: chatty
ineligible_reason: propriety_total_count < 5000
language_model: Undi95/Llama-3-Chatty-2x8B
model_size: 14B
reward_model: ChaiML/reward_gpt2_medium_preference_24m_e2
us_pacific_date: 2024-05-22
win_ratio: 0.5075501865340203
Resubmit model
Running pipeline stage MKMLizer
Starting job with name undi95-llama-3-chatty-2x8b-v1-mkmlizer
Waiting for job on undi95-llama-3-chatty-2x8b-v1-mkmlizer to finish
undi95-llama-3-chatty-2x8b-v1-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
undi95-llama-3-chatty-2x8b-v1-mkmlizer: ║ _____ __ __ ║
undi95-llama-3-chatty-2x8b-v1-mkmlizer: ║ / _/ /_ ___ __/ / ___ ___ / / ║
undi95-llama-3-chatty-2x8b-v1-mkmlizer: ║ / _/ / // / |/|/ / _ \/ -_) -_) / ║
undi95-llama-3-chatty-2x8b-v1-mkmlizer: ║ /_//_/\_, /|__,__/_//_/\__/\__/_/ ║
undi95-llama-3-chatty-2x8b-v1-mkmlizer: ║ /___/ ║
undi95-llama-3-chatty-2x8b-v1-mkmlizer: ║ ║
undi95-llama-3-chatty-2x8b-v1-mkmlizer: ║ Version: 0.8.14 ║
undi95-llama-3-chatty-2x8b-v1-mkmlizer: ║ Copyright 2023 MK ONE TECHNOLOGIES Inc. ║
undi95-llama-3-chatty-2x8b-v1-mkmlizer: ║ https://mk1.ai ║
undi95-llama-3-chatty-2x8b-v1-mkmlizer: ║ ║
undi95-llama-3-chatty-2x8b-v1-mkmlizer: ║ The license key for the current software has been verified as ║
undi95-llama-3-chatty-2x8b-v1-mkmlizer: ║ belonging to: ║
undi95-llama-3-chatty-2x8b-v1-mkmlizer: ║ ║
undi95-llama-3-chatty-2x8b-v1-mkmlizer: ║ Chai Research Corp. ║
undi95-llama-3-chatty-2x8b-v1-mkmlizer: ║ Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f ║
undi95-llama-3-chatty-2x8b-v1-mkmlizer: ║ Expiration: 2024-07-15 23:59:59 ║
undi95-llama-3-chatty-2x8b-v1-mkmlizer: ║ ║
undi95-llama-3-chatty-2x8b-v1-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
undi95-llama-3-chatty-2x8b-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-llama-3-chatty-2x8b-v1-mkmlizer: warnings.warn(warning_message, FutureWarning)
undi95-llama-3-chatty-2x8b-v1-mkmlizer: Downloaded to shared memory in 65.163s
undi95-llama-3-chatty-2x8b-v1-mkmlizer: quantizing model to /dev/shm/model_cache
undi95-llama-3-chatty-2x8b-v1-mkmlizer: Saving flywheel model at /dev/shm/model_cache
undi95-llama-3-chatty-2x8b-v1-mkmlizer: Loading 0: 0%| | 0/419 [00:00<?, ?it/s] Loading 0: 4%|▎ | 15/419 [00:00<00:02, 138.98it/s] Loading 0: 7%|▋ | 29/419 [00:00<00:02, 134.17it/s] Loading 0: 10%|█ | 43/419 [00:00<00:02, 127.55it/s] Loading 0: 15%|█▌ | 63/419 [00:00<00:04, 72.56it/s] Loading 0: 20%|█▉ | 82/419 [00:00<00:03, 94.44it/s] Loading 0: 23%|██▎ | 95/419 [00:00<00:03, 100.01it/s] Loading 0: 26%|██▌ | 109/419 [00:01<00:02, 107.53it/s] Loading 0: 30%|██▉ | 125/419 [00:01<00:02, 120.34it/s] Loading 0: 34%|███▎ | 141/419 [00:01<00:03, 73.61it/s] Loading 0: 36%|███▋ | 152/419 [00:01<00:03, 76.33it/s] Loading 0: 41%|████ | 170/419 [00:01<00:02, 95.27it/s] Loading 0: 44%|████▍ | 184/419 [00:01<00:02, 102.62it/s] Loading 0: 47%|████▋ | 197/419 [00:02<00:02, 108.55it/s] Loading 0: 51%|█████▏ | 215/419 [00:02<00:02, 74.02it/s] Loading 0: 54%|█████▍ | 226/419 [00:02<00:02, 80.07it/s] Loading 0: 58%|█████▊ | 242/419 [00:02<00:01, 95.58it/s] Loading 0: 61%|██████ | 256/419 [00:02<00:01, 103.09it/s] Loading 0: 64%|██████▍ | 270/419 [00:02<00:01, 109.99it/s] Loading 0: 69%|██████▉ | 289/419 [00:03<00:01, 74.77it/s] Loading 0: 72%|███████▏ | 303/419 [00:03<00:01, 84.83it/s] Loading 0: 76%|███████▌ | 317/419 [00:03<00:01, 94.26it/s] Loading 0: 79%|███████▉ | 330/419 [00:03<00:00, 100.83it/s] Loading 0: 82%|████████▏ | 342/419 [00:03<00:00, 104.31it/s] Loading 0: 85%|████████▍ | 355/419 [00:03<00:00, 108.55it/s] Loading 0: 88%|████████▊ | 367/419 [00:20<00:00, 108.55it/s] Loading 0: 88%|████████▊ | 368/419 [00:20<00:19, 2.65it/s] Loading 0: 95%|█████████▍| 396/419 [00:20<00:04, 5.00it/s] Loading 0: 99%|█████████▊| 413/419 [00:20<00:00, 6.93it/s] Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
undi95-llama-3-chatty-2x8b-v1-mkmlizer: quantized model in 26.974s
undi95-llama-3-chatty-2x8b-v1-mkmlizer: Processed model Undi95/Llama-3-Chatty-2x8B in 94.127s
undi95-llama-3-chatty-2x8b-v1-mkmlizer: creating bucket guanaco-mkml-models
undi95-llama-3-chatty-2x8b-v1-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
undi95-llama-3-chatty-2x8b-v1-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/undi95-llama-3-chatty-2x8b-v1
undi95-llama-3-chatty-2x8b-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/undi95-llama-3-chatty-2x8b-v1/tokenizer_config.json
undi95-llama-3-chatty-2x8b-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/undi95-llama-3-chatty-2x8b-v1/tokenizer.json
undi95-llama-3-chatty-2x8b-v1-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/undi95-llama-3-chatty-2x8b-v1/config.json
undi95-llama-3-chatty-2x8b-v1-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/undi95-llama-3-chatty-2x8b-v1/special_tokens_map.json
undi95-llama-3-chatty-2x8b-v1-mkmlizer: cp /dev/shm/model_cache/flywheel_model.1.safetensors s3://guanaco-mkml-models/undi95-llama-3-chatty-2x8b-v1/flywheel_model.1.safetensors
undi95-llama-3-chatty-2x8b-v1-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/undi95-llama-3-chatty-2x8b-v1/flywheel_model.0.safetensors
undi95-llama-3-chatty-2x8b-v1-mkmlizer: loading reward model from ChaiML/reward_gpt2_medium_preference_24m_e2
undi95-llama-3-chatty-2x8b-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-llama-3-chatty-2x8b-v1-mkmlizer: warnings.warn(
undi95-llama-3-chatty-2x8b-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-llama-3-chatty-2x8b-v1-mkmlizer: warnings.warn(
undi95-llama-3-chatty-2x8b-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-llama-3-chatty-2x8b-v1-mkmlizer: warnings.warn(
undi95-llama-3-chatty-2x8b-v1-mkmlizer: Saving duration: 0.264s
undi95-llama-3-chatty-2x8b-v1-mkmlizer: Processed model ChaiML/reward_gpt2_medium_preference_24m_e2 in 3.549s
undi95-llama-3-chatty-2x8b-v1-mkmlizer: creating bucket guanaco-reward-models
undi95-llama-3-chatty-2x8b-v1-mkmlizer: Bucket 's3://guanaco-reward-models/' created
undi95-llama-3-chatty-2x8b-v1-mkmlizer: uploading /tmp/reward_cache to s3://guanaco-reward-models/undi95-llama-3-chatty-2x8b-v1_reward
undi95-llama-3-chatty-2x8b-v1-mkmlizer: cp /tmp/reward_cache/config.json s3://guanaco-reward-models/undi95-llama-3-chatty-2x8b-v1_reward/config.json
undi95-llama-3-chatty-2x8b-v1-mkmlizer: cp /tmp/reward_cache/tokenizer_config.json s3://guanaco-reward-models/undi95-llama-3-chatty-2x8b-v1_reward/tokenizer_config.json
undi95-llama-3-chatty-2x8b-v1-mkmlizer: cp /tmp/reward_cache/special_tokens_map.json s3://guanaco-reward-models/undi95-llama-3-chatty-2x8b-v1_reward/special_tokens_map.json
undi95-llama-3-chatty-2x8b-v1-mkmlizer: cp /tmp/reward_cache/merges.txt s3://guanaco-reward-models/undi95-llama-3-chatty-2x8b-v1_reward/merges.txt
undi95-llama-3-chatty-2x8b-v1-mkmlizer: cp /tmp/reward_cache/vocab.json s3://guanaco-reward-models/undi95-llama-3-chatty-2x8b-v1_reward/vocab.json
undi95-llama-3-chatty-2x8b-v1-mkmlizer: cp /tmp/reward_cache/tokenizer.json s3://guanaco-reward-models/undi95-llama-3-chatty-2x8b-v1_reward/tokenizer.json
undi95-llama-3-chatty-2x8b-v1-mkmlizer: cp /tmp/reward_cache/reward.tensors s3://guanaco-reward-models/undi95-llama-3-chatty-2x8b-v1_reward/reward.tensors
Job undi95-llama-3-chatty-2x8b-v1-mkmlizer completed after 389.83s with status: succeeded
Stopping job with name undi95-llama-3-chatty-2x8b-v1-mkmlizer
Pipeline stage MKMLizer completed in 395.26s
Running pipeline stage MKMLKubeTemplater
Pipeline stage MKMLKubeTemplater completed in 0.09s
Running pipeline stage ISVCDeployer
Creating inference service undi95-llama-3-chatty-2x8b-v1
Waiting for inference service undi95-llama-3-chatty-2x8b-v1 to be ready
Inference service undi95-llama-3-chatty-2x8b-v1 ready after 40.27300477027893s
Pipeline stage ISVCDeployer completed in 48.74s
Running pipeline stage StressChecker
Received healthy response to inference request in 3.046074628829956s
Received healthy response to inference request in 2.081970453262329s
Received healthy response to inference request in 2.013434648513794s
Received healthy response to inference request in 2.0778605937957764s
Received healthy response to inference request in 2.0367352962493896s
5 requests
0 failed requests
5th percentile: 2.018094778060913
10th percentile: 2.0227549076080322
20th percentile: 2.0320751667022705
30th percentile: 2.044960355758667
40th percentile: 2.0614104747772215
50th percentile: 2.0778605937957764
60th percentile: 2.0795045375823973
70th percentile: 2.0811484813690186
80th percentile: 2.2747912883758548
90th percentile: 2.660432958602905
95th percentile: 2.8532537937164304
99th percentile: 3.007510461807251
mean time: 2.251215124130249
Pipeline stage StressChecker completed in 11.93s
Running pipeline stage DaemonicModelEvalScorer
Pipeline stage DaemonicModelEvalScorer completed in 0.03s
Running pipeline stage DaemonicSafetyScorer
Pipeline stage DaemonicSafetyScorer completed in 0.04s
Running M-Eval for topic stay_in_character
undi95-llama-3-chatty-2x8b_v1 status is now deployed due to DeploymentManager action
M-Eval Dataset for topic stay_in_character is loaded
undi95-llama-3-chatty-2x8b_v1 status is now inactive due to auto deactivation removed underperforming models

Usage Metrics

Latency Metrics