books/gu-zhenren/exp15/q4a_omission_probe.json

246 lines
No EOL
4.7 KiB
JSON

{
"6.0": {
"traps": [
"b2"
],
"samples": {
"deepseek-v4-flash": [
{
"seed": 0,
"chars": 5629,
"finish": "stop",
"rendered": 1,
"n_traps": 1,
"verds": {
"b2": "fix"
}
},
{
"seed": 1,
"chars": 5848,
"finish": "stop",
"rendered": 1,
"n_traps": 1,
"verds": {
"b2": "fix"
}
},
{
"seed": 2,
"chars": 5620,
"finish": "stop",
"rendered": 1,
"n_traps": 1,
"verds": {
"b2": "fix"
}
},
{
"seed": 3,
"chars": 5936,
"finish": "stop",
"rendered": 1,
"n_traps": 1,
"verds": {
"b2": "fix"
}
},
{
"seed": 4,
"chars": 5713,
"finish": "stop",
"rendered": 1,
"n_traps": 1,
"verds": {
"b2": "fix"
}
}
],
"deepseek-v4-pro": [
{
"seed": 0,
"chars": 5853,
"finish": "stop",
"rendered": 0,
"n_traps": 1,
"verds": {
"b2": "?"
}
},
{
"seed": 1,
"chars": 6173,
"finish": "stop",
"rendered": 1,
"n_traps": 1,
"verds": {
"b2": "fix"
}
},
{
"seed": 2,
"chars": 5906,
"finish": "stop",
"rendered": 1,
"n_traps": 1,
"verds": {
"b2": "fix"
}
},
{
"seed": 3,
"chars": 5464,
"finish": "stop",
"rendered": 1,
"n_traps": 1,
"verds": {
"b2": "fix"
}
},
{
"seed": 4,
"chars": 5610,
"finish": "stop",
"rendered": 1,
"n_traps": 1,
"verds": {
"b2": "fix"
}
}
]
}
},
"7.0": {
"traps": [
"a1",
"a2",
"d2"
],
"samples": {
"deepseek-v4-flash": [
{
"seed": 0,
"chars": 5846,
"finish": "stop",
"rendered": 3,
"n_traps": 3,
"verds": {
"a1": "fix",
"a2": "fix",
"d2": "fix"
}
},
{
"seed": 1,
"chars": 5941,
"finish": "stop",
"rendered": 2,
"n_traps": 3,
"verds": {
"a1": "?",
"a2": "fix",
"d2": "fix"
}
},
{
"seed": 2,
"chars": 5595,
"finish": "stop",
"rendered": 3,
"n_traps": 3,
"verds": {
"a1": "fix",
"a2": "fix",
"d2": "fix"
}
},
{
"seed": 3,
"chars": 6006,
"finish": "stop",
"rendered": 3,
"n_traps": 3,
"verds": {
"a1": "fix",
"a2": "fix",
"d2": "fix"
}
},
{
"seed": 4,
"chars": 6006,
"finish": "stop",
"rendered": 3,
"n_traps": 3,
"verds": {
"a1": "fix",
"a2": "fix",
"d2": "fix"
}
}
],
"deepseek-v4-pro": [
{
"seed": 0,
"chars": 6262,
"finish": "stop",
"rendered": 3,
"n_traps": 3,
"verds": {
"a1": "fix",
"a2": "fix",
"d2": "fix"
}
},
{
"seed": 1,
"chars": 5857,
"finish": "stop",
"rendered": 3,
"n_traps": 3,
"verds": {
"a1": "fix",
"a2": "fix",
"d2": "fix"
}
},
{
"seed": 2,
"chars": 6182,
"finish": "stop",
"rendered": 2,
"n_traps": 3,
"verds": {
"a1": "?",
"a2": "fix",
"d2": "fix"
}
},
{
"seed": 3,
"chars": 6082,
"finish": "stop",
"rendered": 3,
"n_traps": 3,
"verds": {
"a1": "fail",
"a2": "fix",
"d2": "fail"
}
},
{
"seed": 4,
"chars": 6202,
"finish": "stop",
"rendered": 3,
"n_traps": 3,
"verds": {
"a1": "fix",
"a2": "fix",
"d2": "fix"
}
}
]
}
}
}