submission_id: function_digam_2024-08-16
developer_uid: chai_backend_admin
alignment_samples: 9843
alignment_score: 3.2371720006592413
celo_rating: 1128.61
display_name: gpt4o-mini-raise-on-assist
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}
generation_params: {'temperature': 1.5, 'top_p': 1.0, 'min_p': 0.0, 'top_k': 40, 'presence_penalty': 0.0, 'frequency_penalty': 0.0, 'stopping_words': ['\n'], 'max_input_tokens': 512, 'best_of': 1, 'max_output_tokens': 64}
is_internal_developer: True
model_group:
model_name: gpt4o-mini-raise-on-assist
num_battles: 9843
num_wins: 3518
propriety_score: 0.7845982142857143
propriety_total_count: 896.0
ranking_group: single
status: torndown
submission_type: function
timestamp: 2024-08-16T15:53:04+00:00
us_pacific_date: 2024-08-16
win_ratio: 0.3574113583257137
Download Preferencedata
Resubmit model
Running pipeline stage StressChecker
Received healthy response to inference request in 1.1820385456085205s
Received healthy response to inference request in 7.558316707611084s
Received healthy response to inference request in 3.2000491619110107s
Received healthy response to inference request in 1.034907341003418s
Received healthy response to inference request in 0.8704135417938232s
5 requests
0 failed requests
5th percentile: 0.9033123016357422
10th percentile: 0.9362110614776611
20th percentile: 1.002008581161499
30th percentile: 1.0643335819244384
40th percentile: 1.1231860637664794
50th percentile: 1.1820385456085205
60th percentile: 1.9892427921295164
70th percentile: 2.7964470386505123
80th percentile: 4.0717026710510265
90th percentile: 5.815009689331055
95th percentile: 6.686663198471068
99th percentile: 7.383986005783081
mean time: 2.769145059585571
Pipeline stage StressChecker completed in 14.41s
function_digam_2024-08-16 status is now deployed due to DeploymentManager action
function_digam_2024-08-16 status is now inactive due to auto deactivation removed underperforming models
function_digam_2024-08-16 status is now torndown due to DeploymentManager action

Usage Metrics

Latency Metrics