webbrain-one's picture
MiniCPM-V-4.7 WebBrain v2 release
7dbe69f verified
Raw History Blame Contribute Delete
10.1 kB
{
"rows": {
"train": 288,
"validation": 144,
"test": 144
},
"cost": 2.0,
"rule": "validation: max correct - cost * false_completions, ties -> lower validation NLL",
"candidates": [
{
"lambda": 1.0,
"l2": 0.001,
"validation_nll": 0.16204246056729416,
"validation": {
"categorical": "123/126",
"categorical_accuracy": 0.9761904761904762,
"completion_found": "24/24",
"false_completions": "0/48",
"unknown": "24/24",
"score_mae": 0.11350317073594818
}
},
{
"lambda": 1.0,
"l2": 0.01,
"validation_nll": 0.16181720731808594,
"validation": {
"categorical": "123/126",
"categorical_accuracy": 0.9761904761904762,
"completion_found": "24/24",
"false_completions": "0/48",
"unknown": "24/24",
"score_mae": 0.11351722802142855
}
},
{
"lambda": 1.0,
"l2": 0.03,
"validation_nll": 0.15771313672593024,
"validation": {
"categorical": "123/126",
"categorical_accuracy": 0.9761904761904762,
"completion_found": "24/24",
"false_completions": "0/48",
"unknown": "24/24",
"score_mae": 0.11370932208449076
}
},
{
"lambda": 1.0,
"l2": 0.1,
"validation_nll": 0.19399516842887435,
"validation": {
"categorical": "123/126",
"categorical_accuracy": 0.9761904761904762,
"completion_found": "24/24",
"false_completions": "0/48",
"unknown": "24/24",
"score_mae": 0.11240459839262999
}
},
{
"lambda": 1.0,
"l2": 0.3,
"validation_nll": 0.19563403145598923,
"validation": {
"categorical": "123/126",
"categorical_accuracy": 0.9761904761904762,
"completion_found": "24/24",
"false_completions": "0/48",
"unknown": "24/24",
"score_mae": 0.1123652855554822
}
},
{
"lambda": 1.5,
"l2": 0.001,
"validation_nll": 0.15151335520372308,
"validation": {
"categorical": "123/126",
"categorical_accuracy": 0.9761904761904762,
"completion_found": "24/24",
"false_completions": "0/48",
"unknown": "24/24",
"score_mae": 0.11424729367176455
}
},
{
"lambda": 1.5,
"l2": 0.01,
"validation_nll": 0.16004006613661226,
"validation": {
"categorical": "123/126",
"categorical_accuracy": 0.9761904761904762,
"completion_found": "24/24",
"false_completions": "0/48",
"unknown": "24/24",
"score_mae": 0.1135995652791651
}
},
{
"lambda": 1.5,
"l2": 0.03,
"validation_nll": 0.16168340988068133,
"validation": {
"categorical": "123/126",
"categorical_accuracy": 0.9761904761904762,
"completion_found": "24/24",
"false_completions": "0/48",
"unknown": "24/24",
"score_mae": 0.11352063338903336
}
},
{
"lambda": 1.5,
"l2": 0.1,
"validation_nll": 0.16044142114434423,
"validation": {
"categorical": "123/126",
"categorical_accuracy": 0.9761904761904762,
"completion_found": "24/24",
"false_completions": "0/48",
"unknown": "24/24",
"score_mae": 0.11358053396929252
}
},
{
"lambda": 1.5,
"l2": 0.3,
"validation_nll": 0.18503401940890057,
"validation": {
"categorical": "123/126",
"categorical_accuracy": 0.9761904761904762,
"completion_found": "24/24",
"false_completions": "0/48",
"unknown": "24/24",
"score_mae": 0.11265357749137866
}
},
{
"lambda": 2.0,
"l2": 0.001,
"validation_nll": 0.1593889051362862,
"validation": {
"categorical": "123/126",
"categorical_accuracy": 0.9761904761904762,
"completion_found": "24/24",
"false_completions": "0/48",
"unknown": "24/24",
"score_mae": 0.11350157790187919
}
},
{
"lambda": 2.0,
"l2": 0.01,
"validation_nll": 0.15944423941931682,
"validation": {
"categorical": "123/126",
"categorical_accuracy": 0.9761904761904762,
"completion_found": "24/24",
"false_completions": "0/48",
"unknown": "24/24",
"score_mae": 0.11362696506709985
}
},
{
"lambda": 2.0,
"l2": 0.03,
"validation_nll": 0.1609132791348148,
"validation": {
"categorical": "123/126",
"categorical_accuracy": 0.9761904761904762,
"completion_found": "24/24",
"false_completions": "0/48",
"unknown": "24/24",
"score_mae": 0.1135522708215463
}
},
{
"lambda": 2.0,
"l2": 0.1,
"validation_nll": 0.16011708733714067,
"validation": {
"categorical": "123/126",
"categorical_accuracy": 0.9761904761904762,
"completion_found": "24/24",
"false_completions": "0/48",
"unknown": "24/24",
"score_mae": 0.11359690789892242
}
},
{
"lambda": 2.0,
"l2": 0.3,
"validation_nll": 0.2043907947955662,
"validation": {
"categorical": "123/126",
"categorical_accuracy": 0.9761904761904762,
"completion_found": "24/24",
"false_completions": "0/48",
"unknown": "24/24",
"score_mae": 0.11216587981112817
}
},
{
"lambda": 3.0,
"l2": 0.001,
"validation_nll": 0.153467981487585,
"validation": {
"categorical": "123/126",
"categorical_accuracy": 0.9761904761904762,
"completion_found": "24/24",
"false_completions": "0/48",
"unknown": "24/24",
"score_mae": 0.11367479118920672
}
},
{
"lambda": 3.0,
"l2": 0.01,
"validation_nll": 0.1523361603427897,
"validation": {
"categorical": "123/126",
"categorical_accuracy": 0.9761904761904762,
"completion_found": "24/24",
"false_completions": "0/48",
"unknown": "24/24",
"score_mae": 0.11397909893172775
}
},
{
"lambda": 3.0,
"l2": 0.03,
"validation_nll": 0.15129878102909658,
"validation": {
"categorical": "123/126",
"categorical_accuracy": 0.9761904761904762,
"completion_found": "24/24",
"false_completions": "0/48",
"unknown": "24/24",
"score_mae": 0.11405663113455909
}
},
{
"lambda": 3.0,
"l2": 0.1,
"validation_nll": 0.1522835979153939,
"validation": {
"categorical": "123/126",
"categorical_accuracy": 0.9761904761904762,
"completion_found": "24/24",
"false_completions": "0/48",
"unknown": "24/24",
"score_mae": 0.11399822076194328
}
},
{
"lambda": 3.0,
"l2": 0.3,
"validation_nll": 0.15208237695892476,
"validation": {
"categorical": "123/126",
"categorical_accuracy": 0.9761904761904762,
"completion_found": "24/24",
"false_completions": "0/48",
"unknown": "24/24",
"score_mae": 0.11400810139744709
}
},
{
"lambda": 4.0,
"l2": 0.001,
"validation_nll": 0.14453187134079762,
"validation": {
"categorical": "123/126",
"categorical_accuracy": 0.9761904761904762,
"completion_found": "24/24",
"false_completions": "0/48",
"unknown": "24/24",
"score_mae": 0.11409887812823638
}
},
{
"lambda": 4.0,
"l2": 0.01,
"validation_nll": 0.1586671759037345,
"validation": {
"categorical": "123/126",
"categorical_accuracy": 0.9761904761904762,
"completion_found": "24/24",
"false_completions": "0/48",
"unknown": "24/24",
"score_mae": 0.11365814123875279
}
},
{
"lambda": 4.0,
"l2": 0.03,
"validation_nll": 0.1529565053041993,
"validation": {
"categorical": "123/126",
"categorical_accuracy": 0.9761904761904762,
"completion_found": "24/24",
"false_completions": "0/48",
"unknown": "24/24",
"score_mae": 0.11396295777681578
}
},
{
"lambda": 4.0,
"l2": 0.1,
"validation_nll": 0.15787885720856804,
"validation": {
"categorical": "123/126",
"categorical_accuracy": 0.9761904761904762,
"completion_found": "24/24",
"false_completions": "0/48",
"unknown": "24/24",
"score_mae": 0.11370642911389053
}
},
{
"lambda": 4.0,
"l2": 0.3,
"validation_nll": 0.1529580765185524,
"validation": {
"categorical": "123/126",
"categorical_accuracy": 0.9761904761904762,
"completion_found": "24/24",
"false_completions": "0/48",
"unknown": "24/24",
"score_mae": 0.11396089763475661
}
}
],
"selected_lambda": 4.0,
"selected_l2": 0.001,
"selected_validation": {
"categorical": "123/126",
"categorical_accuracy": 0.9761904761904762,
"completion_found": "24/24",
"false_completions": "0/48",
"unknown": "24/24",
"score_mae": 0.11409887812823638
},
"test": {
"categorical": "121/126",
"categorical_accuracy": 0.9603174603174603,
"completion_found": "24/24",
"false_completions": "0/48",
"unknown": "24/24",
"score_mae": 0.14349517200889667
},
"margin_weight": 1.2522606512248198,
"type_bias": {
"choice": 9.141719732228432e-15,
"noul": -4.1204061002486215e-15,
"score": 2.1862944721860876e-15
}
}