{
  "type": "tweet",
  "id": "2113657683689812476",
  "url": "https://x.com/ChrisGPotts/status/2113657683689812476",
  "twitterUrl": "https://twitter.com/ChrisGPotts/status/2113657683689812476",
  "text": "The wonderful “Don’t just Fix it in Post” paper identifies predicting memorization as a key open problem. \n\nA progress report: we can now accurately predict verbatim memorization using just three well-understood variables. Experiments by Jing Huang, using the Hubble model suite: https://t.co/BdnqtoeX45",
  "source": "",
  "retweetCount": 11,
  "replyCount": 5,
  "likeCount": 56,
  "quoteCount": 2,
  "viewCount": 3729,
  "createdAt": "Fri Oct 23 15:43:56 +0000 2026",
  "lang": "en",
  "bookmarkCount": 44,
  "isReply": false,
  "inReplyToId": null,
  "conversationId": "2113657683689812476",
  "displayTextRange": [
    0,
    279
  ],
  "inReplyToUserId": null,
  "inReplyToUsername": null,
  "author": {
    "type": "user",
    "userName": "ChrisGPotts",
    "url": "https://x.com/ChrisGPotts",
    "twitterUrl": "https://twitter.com/ChrisGPotts",
    "id": "408714449",
    "name": "Christopher Potts",
    "isVerified": false,
    "isBlueVerified": true,
    "verifiedType": null,
    "profilePicture": "https://pbs.twimg.com/profile_images/2489657624/elm9xc0grliqd8e3w9mg_normal.png",
    "coverPicture": "https://pbs.twimg.com/profile_banners/408714449/1437858474",
    "description": "",
    "location": "",
    "followers": 16844,
    "following": 746,
    "status": "",
    "canDm": false,
    "canMediaTag": true,
    "createdAt": "Wed Nov 09 19:59:28 +0000 2011",
    "entities": {
      "description": {
        "urls": []
      },
      "url": {}
    },
    "fastFollowersCount": 0,
    "favouritesCount": 3567,
    "hasCustomTimelines": true,
    "isTranslator": false,
    "mediaCount": 172,
    "statusesCount": 2995,
    "withheldInCountries": [],
    "affiliatesHighlightedLabel": {},
    "possiblySensitive": false,
    "pinnedTweetIds": [
      "2074142172635627593"
    ],
    "profile_bio": {
      "description": "Stanford Professor of Linguistics and, by courtesy, of Computer Science. Member of technical staff @stanfordnlp and @StanfordAILab. Co-founder @ Bigspin AI.",
      "entities": {
        "description": {
          "user_mentions": [
            {
              "id_str": "",
              "indices": [
                99,
                111
              ],
              "name": "",
              "screen_name": "stanfordnlp"
            },
            {
              "id_str": "",
              "indices": [
                116,
                130
              ],
              "name": "",
              "screen_name": "StanfordAILab"
            }
          ]
        },
        "url": {
          "urls": [
            {
              "display_url": "web.stanford.edu/~cgpotts/",
              "expanded_url": "http://web.stanford.edu/~cgpotts/",
              "indices": [
                0,
                23
              ],
              "url": "https://t.co/3zBrAqL7VI"
            }
          ]
        }
      }
    },
    "isAutomated": false,
    "automatedBy": null,
    "protected": false
  },
  "extendedEntities": {
    "media": [
      {
        "display_url": "pic.x.com/BdnqtoeX45",
        "expanded_url": "https://x.com/ChrisGPotts/status/2113657683689812476/photo/1",
        "ext_alt_text": "Dot plot addressing whether memorized sequences can be predicted without training the full model. Headline: “Largely yes, with Frequency and Recency doing most of the work.” Four rows show AUC-ROC (0.5 = chance, 1.0 = perfect) for a linear predictor of verbatim extraction, one mark per Hubble model (1B or 8B params; 100B or 500B tokens). All three features: 0.82–0.94. Frequency alone: 0.81–0.93. Recency alone: 0.82–0.90. Hardness alone: 0.51–0.55, barely above chance. In every row the 1B/500B model scores lowest and the 1B/100B model highest. Footnote: held-out strings, mean of 5 seeds; Frequency = copies of the string, Recency = steps since last copy, Hardness = NLL on models that never saw it.",
        "ext_master_playlist_only": [],
        "ext_media_availability": {
          "status": "Available"
        },
        "ext_playlists": [],
        "features": {
          "large": {
            "faces": []
          },
          "orig": {
            "faces": []
          }
        },
        "id_str": "2113657480161239040",
        "indices": [
          280,
          303
        ],
        "media_key": "3_2113657480161239040",
        "media_results": {
          "id": "QXBpTWVkaWFSZXN1bHRzOgwAAQoAAR1VOEeR26AACgACHVU4dvUbMfwAAA==",
          "result": {
            "__typename": "ApiMedia",
            "id": "QXBpTWVkaWE6DAABCgABHVU4R5HboAAKAAIdVTh29Rsx/AAA",
            "media_key": "3_2113657480161239040"
          }
        },
        "media_url_https": "https://pbs.twimg.com/media/HVU4R5HboAArz3v.jpg",
        "original_info": {
          "focus_rects": [
            {
              "h": 1344,
              "w": 2400,
              "x": 0,
              "y": 0
            },
            {
              "h": 1350,
              "w": 1350,
              "x": 0,
              "y": 0
            },
            {
              "h": 1350,
              "w": 1184,
              "x": 0,
              "y": 0
            },
            {
              "h": 1350,
              "w": 675,
              "x": 83,
              "y": 0
            },
            {
              "h": 1350,
              "w": 2400,
              "x": 0,
              "y": 0
            }
          ],
          "height": 1350,
          "width": 2400
        },
        "sizes": {
          "large": {
            "h": 1152,
            "w": 2048
          }
        },
        "type": "photo",
        "url": "https://t.co/BdnqtoeX45"
      }
    ]
  },
  "card": null,
  "place": {},
  "entities": {},
  "quoted_tweet": null,
  "retweeted_tweet": null,
  "isLimitedReply": false,
  "communityInfo": null,
  "article": null
}