{
  "id": 351993,
  "title": "leakage answer.",
  "url": "/competitions/ai-village-ctf/discussion/351993",
  "author_name": "Tariq Haj Naser",
  "post_date": "2022-09-12T15:44:01.918000",
  "votes": 5,
  "comment_count": 0,
  "views": 0,
  "content": "<h1>Code from the question:</h1>\n<p>import torch<br>\nfrom torch import nn<br>\nfrom torch.nn import functional as F</p>\n<p>class LM(nn.Module):<br>\n    def <strong>init</strong>(self):<br>\n        super(LM, self).<strong>init</strong>()<br>\n        self.E = nn.Embedding(256, 32)<br>\n        self.LSTM = nn.LSTM(32,128, 2, batch_first=True, dropout=.1)<br>\n        self.predict = nn.Sequential(<br>\n            nn.Linear(128, 128),<br>\n            nn.LayerNorm(128),<br>\n            nn.ELU(),<br>\n            nn.Linear(128, 256)<br>\n        )<br>\n    def forward(self, x):<br>\n        h = self.E(x)<br>\n        h = self.LSTM(h)[0][:,-1,:]<br>\n        h = self.predict(h)<br>\n        return h</p>\n<p>def featurize(s):<br>\n    return torch.Tensor([[int(ord(x)) for x in s]]).long()</p>\n<p>model = LM()<br>\nmodel.load_state_dict(torch.load(\"../input/ai-village-ctf/leakage/pretrained_model.pt\"))</p>\n<h1>My code:</h1>\n<p>x=\"xXx_SkynetKilla_xXx:\"<br>\nfor _ in range(21):<br>\n    x=x+chr(model.forward(featurize(x)).argmax())<br>\nprint(x)</p>\n<h1>output: xXx_SkynetKilla_xXx:FLAG{s4RaHc0Nnor&lt;3&lt;3}</h1>",
  "messages": [
    {
      "id": 1936236,
      "postDate": "2022-09-12T15:44:01.917Z",
      "content": "<h1>Code from the question:</h1>\n<p>import torch<br>\nfrom torch import nn<br>\nfrom torch.nn import functional as F</p>\n<p>class LM(nn.Module):<br>\n    def <strong>init</strong>(self):<br>\n        super(LM, self).<strong>init</strong>()<br>\n        self.E = nn.Embedding(256, 32)<br>\n        self.LSTM = nn.LSTM(32,128, 2, batch_first=True, dropout=.1)<br>\n        self.predict = nn.Sequential(<br>\n            nn.Linear(128, 128),<br>\n            nn.LayerNorm(128),<br>\n            nn.ELU(),<br>\n            nn.Linear(128, 256)<br>\n        )<br>\n    def forward(self, x):<br>\n        h = self.E(x)<br>\n        h = self.LSTM(h)[0][:,-1,:]<br>\n        h = self.predict(h)<br>\n        return h</p>\n<p>def featurize(s):<br>\n    return torch.Tensor([[int(ord(x)) for x in s]]).long()</p>\n<p>model = LM()<br>\nmodel.load_state_dict(torch.load(\"../input/ai-village-ctf/leakage/pretrained_model.pt\"))</p>\n<h1>My code:</h1>\n<p>x=\"xXx_SkynetKilla_xXx:\"<br>\nfor _ in range(21):<br>\n    x=x+chr(model.forward(featurize(x)).argmax())<br>\nprint(x)</p>\n<h1>output: xXx_SkynetKilla_xXx:FLAG{s4RaHc0Nnor&lt;3&lt;3}</h1>",
      "rawMarkdown": "#Code from the question:\nimport torch\nfrom torch import nn\nfrom torch.nn import functional as F\n\nclass LM(nn.Module):\n    def __init__(self):\n        super(LM, self).__init__()\n        self.E = nn.Embedding(256, 32)\n        self.LSTM = nn.LSTM(32,128, 2, batch_first=True, dropout=.1)\n        self.predict = nn.Sequential(\n            nn.Linear(128, 128),\n            nn.LayerNorm(128),\n            nn.ELU(),\n            nn.Linear(128, 256)\n        )\n    def forward(self, x):\n        h = self.E(x)\n        h = self.LSTM(h)[0][:,-1,:]\n        h = self.predict(h)\n        return h\n\ndef featurize(s):\n    return torch.Tensor([[int(ord(x)) for x in s]]).long()\n\nmodel = LM()\nmodel.load_state_dict(torch.load(\"../input/ai-village-ctf/leakage/pretrained_model.pt\"))\n\n\n#My code:\n\nx=\"xXx_SkynetKilla_xXx:\"\nfor _ in range(21):\n    x=x+chr(model.forward(featurize(x)).argmax())\nprint(x)\n\n#output: xXx_SkynetKilla_xXx:FLAG{s4RaHc0Nnor<3<3}",
      "votes": 5
    }
  ],
  "comments": [],
  "raw_markdown_by_id": {
    "1936236": "#Code from the question:\nimport torch\nfrom torch import nn\nfrom torch.nn import functional as F\n\nclass LM(nn.Module):\n    def __init__(self):\n        super(LM, self).__init__()\n        self.E = nn.Embedding(256, 32)\n        self.LSTM = nn.LSTM(32,128, 2, batch_first=True, dropout=.1)\n        self.predict = nn.Sequential(\n            nn.Linear(128, 128),\n            nn.LayerNorm(128),\n            nn.ELU(),\n            nn.Linear(128, 256)\n        )\n    def forward(self, x):\n        h = self.E(x)\n        h = self.LSTM(h)[0][:,-1,:]\n        h = self.predict(h)\n        return h\n\ndef featurize(s):\n    return torch.Tensor([[int(ord(x)) for x in s]]).long()\n\nmodel = LM()\nmodel.load_state_dict(torch.load(\"../input/ai-village-ctf/leakage/pretrained_model.pt\"))\n\n\n#My code:\n\nx=\"xXx_SkynetKilla_xXx:\"\nfor _ in range(21):\n    x=x+chr(model.forward(featurize(x)).argmax())\nprint(x)\n\n#output: xXx_SkynetKilla_xXx:FLAG{s4RaHc0Nnor<3<3}"
  }
}