developer_uid: Hastagaras
submission_id: hastagaras-test-gmn-cld-dpo_v1
model_name: hastagaras-test-gmn-cld-dpo_v1
model_group: Hastagaras/test-gmn-cld-
status: torndown
timestamp: 2024-04-26T17:09:53+00:00
num_battles: 7737
num_wins: 3752
celo_rating: 1140.18
family_friendly_score: 0.0
submission_type: basic
model_repo: Hastagaras/test-gmn-cld-dpo
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-test-gmn-cld-dpo_v1
is_internal_developer: False
language_model: Hastagaras/test-gmn-cld-dpo
model_size: 8B
ranking_group: single
us_pacific_date: 2024-04-26
win_ratio: 0.48494248416698976
generation_params: {'temperature': 0.95, 'top_p': 1.0, 'min_p': 0.0, 'top_k': 40, 'presence_penalty': 0.0, 'frequency_penalty': 1.1, 'stopping_words': ['<|eot_id|>'], 'max_input_tokens': 512, 'best_of': 16, 'max_output_tokens': 64}
formatter: {'memory_template': '<|start_header_id|>system<|end_header_id|>\n\nYou are {bot_name}.\n\n{memory}<|eot_id|>', 'prompt_template': "<|start_header_id|>user<|end_header_id|>\n\nLet's begin.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n{bot_name}: {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': "{bot_name}'s Persona: {memory}\n####\n", 'prompt_template': '{bot_name}: {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-test-gmn-cld-dpo-v1-mkmlizer
Waiting for job on hastagaras-test-gmn-cld-dpo-v1-mkmlizer to finish
hastagaras-test-gmn-cld-dpo-v1-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
hastagaras-test-gmn-cld-dpo-v1-mkmlizer: ║ _____ __ __ ║
hastagaras-test-gmn-cld-dpo-v1-mkmlizer: ║ / _/ /_ ___ __/ / ___ ___ / / ║
hastagaras-test-gmn-cld-dpo-v1-mkmlizer: ║ / _/ / // / |/|/ / _ \/ -_) -_) / ║
hastagaras-test-gmn-cld-dpo-v1-mkmlizer: ║ /_//_/\_, /|__,__/_//_/\__/\__/_/ ║
hastagaras-test-gmn-cld-dpo-v1-mkmlizer: ║ /___/ ║
hastagaras-test-gmn-cld-dpo-v1-mkmlizer: ║ ║
hastagaras-test-gmn-cld-dpo-v1-mkmlizer: ║ Version: 0.8.10 ║
hastagaras-test-gmn-cld-dpo-v1-mkmlizer: ║ Copyright 2023 MK ONE TECHNOLOGIES Inc. ║
hastagaras-test-gmn-cld-dpo-v1-mkmlizer: ║ ║
hastagaras-test-gmn-cld-dpo-v1-mkmlizer: ║ The license key for the current software has been verified as ║
hastagaras-test-gmn-cld-dpo-v1-mkmlizer: ║ belonging to: ║
hastagaras-test-gmn-cld-dpo-v1-mkmlizer: ║ ║
hastagaras-test-gmn-cld-dpo-v1-mkmlizer: ║ Chai Research Corp. ║
hastagaras-test-gmn-cld-dpo-v1-mkmlizer: ║ Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f ║
hastagaras-test-gmn-cld-dpo-v1-mkmlizer: ║ Expiration: 2024-07-15 23:59:59 ║
hastagaras-test-gmn-cld-dpo-v1-mkmlizer: ║ ║
hastagaras-test-gmn-cld-dpo-v1-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
hastagaras-test-gmn-cld-dpo-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-test-gmn-cld-dpo-v1-mkmlizer: warnings.warn(warning_message, FutureWarning)
hastagaras-test-gmn-cld-dpo-v1-mkmlizer: Downloaded to shared memory in 24.728s
hastagaras-test-gmn-cld-dpo-v1-mkmlizer: quantizing model to /dev/shm/model_cache
hastagaras-test-gmn-cld-dpo-v1-mkmlizer: Saving flywheel model at /dev/shm/model_cache
hastagaras-test-gmn-cld-dpo-v1-mkmlizer: Loading 0: 0%| | 0/291 [00:00<?, ?it/s] Loading 0: 1%| | 2/291 [00:04<10:27, 2.17s/it] Loading 0: 36%|███▌ | 105/291 [00:05<00:07, 25.76it/s] Loading 0: 70%|██████▉ | 203/291 [00:06<00:02, 43.12it/s] Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
hastagaras-test-gmn-cld-dpo-v1-mkmlizer: quantized model in 18.422s
hastagaras-test-gmn-cld-dpo-v1-mkmlizer: Processed model Hastagaras/test-gmn-cld-dpo in 44.196s
hastagaras-test-gmn-cld-dpo-v1-mkmlizer: creating bucket guanaco-mkml-models
hastagaras-test-gmn-cld-dpo-v1-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
hastagaras-test-gmn-cld-dpo-v1-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/hastagaras-test-gmn-cld-dpo-v1
hastagaras-test-gmn-cld-dpo-v1-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/hastagaras-test-gmn-cld-dpo-v1/config.json
hastagaras-test-gmn-cld-dpo-v1-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/hastagaras-test-gmn-cld-dpo-v1/special_tokens_map.json
hastagaras-test-gmn-cld-dpo-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/hastagaras-test-gmn-cld-dpo-v1/tokenizer_config.json
hastagaras-test-gmn-cld-dpo-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/hastagaras-test-gmn-cld-dpo-v1/tokenizer.json
hastagaras-test-gmn-cld-dpo-v1-mkmlizer: loading reward model from ChaiML/reward_gpt2_medium_preference_24m_e2
hastagaras-test-gmn-cld-dpo-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-test-gmn-cld-dpo-v1-mkmlizer: warnings.warn(
hastagaras-test-gmn-cld-dpo-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-test-gmn-cld-dpo-v1-mkmlizer: warnings.warn(
hastagaras-test-gmn-cld-dpo-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-test-gmn-cld-dpo-v1-mkmlizer: warnings.warn(
hastagaras-test-gmn-cld-dpo-v1-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()
hastagaras-test-gmn-cld-dpo-v1-mkmlizer: return self.fget.__get__(instance, owner)()
hastagaras-test-gmn-cld-dpo-v1-mkmlizer: Saving model to /tmp/reward_cache/reward.tensors
hastagaras-test-gmn-cld-dpo-v1-mkmlizer: Saving duration: 0.322s
hastagaras-test-gmn-cld-dpo-v1-mkmlizer: Processed model ChaiML/reward_gpt2_medium_preference_24m_e2 in 3.458s
hastagaras-test-gmn-cld-dpo-v1-mkmlizer: creating bucket guanaco-reward-models
hastagaras-test-gmn-cld-dpo-v1-mkmlizer: Bucket 's3://guanaco-reward-models/' created
hastagaras-test-gmn-cld-dpo-v1-mkmlizer: uploading /tmp/reward_cache to s3://guanaco-reward-models/hastagaras-test-gmn-cld-dpo-v1_reward
hastagaras-test-gmn-cld-dpo-v1-mkmlizer: cp /tmp/reward_cache/config.json s3://guanaco-reward-models/hastagaras-test-gmn-cld-dpo-v1_reward/config.json
hastagaras-test-gmn-cld-dpo-v1-mkmlizer: cp /tmp/reward_cache/tokenizer_config.json s3://guanaco-reward-models/hastagaras-test-gmn-cld-dpo-v1_reward/tokenizer_config.json
hastagaras-test-gmn-cld-dpo-v1-mkmlizer: cp /tmp/reward_cache/special_tokens_map.json s3://guanaco-reward-models/hastagaras-test-gmn-cld-dpo-v1_reward/special_tokens_map.json
hastagaras-test-gmn-cld-dpo-v1-mkmlizer: cp /tmp/reward_cache/merges.txt s3://guanaco-reward-models/hastagaras-test-gmn-cld-dpo-v1_reward/merges.txt
hastagaras-test-gmn-cld-dpo-v1-mkmlizer: cp /tmp/reward_cache/vocab.json s3://guanaco-reward-models/hastagaras-test-gmn-cld-dpo-v1_reward/vocab.json
hastagaras-test-gmn-cld-dpo-v1-mkmlizer: cp /tmp/reward_cache/tokenizer.json s3://guanaco-reward-models/hastagaras-test-gmn-cld-dpo-v1_reward/tokenizer.json
hastagaras-test-gmn-cld-dpo-v1-mkmlizer: cp /tmp/reward_cache/reward.tensors s3://guanaco-reward-models/hastagaras-test-gmn-cld-dpo-v1_reward/reward.tensors
Job hastagaras-test-gmn-cld-dpo-v1-mkmlizer completed after 73.53s with status: succeeded
Stopping job with name hastagaras-test-gmn-cld-dpo-v1-mkmlizer
Pipeline stage MKMLizer completed in 78.35s
Running pipeline stage MKMLKubeTemplater
Pipeline stage MKMLKubeTemplater completed in 0.11s
Running pipeline stage ISVCDeployer
Creating inference service hastagaras-test-gmn-cld-dpo-v1
Waiting for inference service hastagaras-test-gmn-cld-dpo-v1 to be ready
Inference service hastagaras-test-gmn-cld-dpo-v1 ready after 40.28023338317871s
Pipeline stage ISVCDeployer completed in 48.44s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.199018955230713s
Received healthy response to inference request in 1.2649812698364258s
Received healthy response to inference request in 1.345703363418579s
Received healthy response to inference request in 1.2786815166473389s
Received healthy response to inference request in 1.2590787410736084s
5 requests
0 failed requests
5th percentile: 1.260259246826172
10th percentile: 1.2614397525787353
20th percentile: 1.2638007640838622
30th percentile: 1.2677213191986083
40th percentile: 1.2732014179229736
50th percentile: 1.2786815166473389
60th percentile: 1.305490255355835
70th percentile: 1.332298994064331
80th percentile: 1.516366481781006
90th percentile: 1.8576927185058594
95th percentile: 2.028355836868286
99th percentile: 2.1648863315582276
mean time: 1.469492769241333
Pipeline stage StressChecker completed in 7.96s
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
%s, retrying in %s seconds...
M-Eval Dataset for topic stay_in_character is loaded
hastagaras-test-gmn-cld-dpo_v1 status is now deployed due to DeploymentManager action
%s, retrying in %s seconds...
hastagaras-test-gmn-cld-dpo_v1 status is now inactive due to auto deactivation removed underperforming models
admin requested tearing down of hastagaras-test-gmn-cld-dpo_v1
Running pipeline stage ISVCDeleter
Checking if service hastagaras-test-gmn-cld-dpo-v1 is running
Tearing down inference service hastagaras-test-gmn-cld-dpo-v1
Toredown service hastagaras-test-gmn-cld-dpo-v1
Pipeline stage ISVCDeleter completed in 4.56s
Running pipeline stage MKMLModelDeleter
Cleaning model data from S3
Cleaning model data from model cache
Deleting key hastagaras-test-gmn-cld-dpo-v1/config.json from bucket guanaco-mkml-models
Deleting key hastagaras-test-gmn-cld-dpo-v1/flywheel_model.0.safetensors from bucket guanaco-mkml-models
Deleting key hastagaras-test-gmn-cld-dpo-v1/special_tokens_map.json from bucket guanaco-mkml-models
Deleting key hastagaras-test-gmn-cld-dpo-v1/tokenizer.json from bucket guanaco-mkml-models
Deleting key hastagaras-test-gmn-cld-dpo-v1/tokenizer_config.json from bucket guanaco-mkml-models
Cleaning model data from model cache
Deleting key hastagaras-test-gmn-cld-dpo-v1_reward/config.json from bucket guanaco-reward-models
Deleting key hastagaras-test-gmn-cld-dpo-v1_reward/merges.txt from bucket guanaco-reward-models
Deleting key hastagaras-test-gmn-cld-dpo-v1_reward/reward.tensors from bucket guanaco-reward-models
Deleting key hastagaras-test-gmn-cld-dpo-v1_reward/special_tokens_map.json from bucket guanaco-reward-models
Deleting key hastagaras-test-gmn-cld-dpo-v1_reward/tokenizer.json from bucket guanaco-reward-models
Deleting key hastagaras-test-gmn-cld-dpo-v1_reward/tokenizer_config.json from bucket guanaco-reward-models
Deleting key hastagaras-test-gmn-cld-dpo-v1_reward/vocab.json from bucket guanaco-reward-models
Pipeline stage MKMLModelDeleter completed in 2.05s
hastagaras-test-gmn-cld-dpo_v1 status is now torndown due to DeploymentManager action