Upload folder using huggingface_hub

.gitattributes

@@ -33,3 +33,6 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+checkpoint-2000/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+checkpoint-3500/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+tokenizer.json filter=lfs diff=lfs merge=lfs -text

README.md

@@ -0,0 +1,199 @@
+---
+language:
+- en
+- ur
+license: mit
+tags:
+- text-classification
+- fake-news-detection
+- misinformation
+- xlm-roberta
+- urdu
+- english
+- multilingual
+datasets:
+- liar
+- GonzaloA/fake_news
+- UrduFakeNews-AxToGrind
+metrics:
+- accuracy
+- f1
+model-index:
+- name: sifta-fakenews-xlmr
+  results:
+  - task:
+      type: text-classification
+      name: Fake News Detection
+    metrics:
+    - type: accuracy
+      value: 0.9696
+    - type: f1
+      value: 0.9695
+      name: Macro F1
+---
+
+# SIFTA — Multilingual Fake News Detector (EN + UR)
+
+**Model:** Fine-tuned `xlm-roberta-base` for binary fake news classification
+**Languages:** English · Urdu (اردو)
+**Labels:** `TRUE` (0) · `FAKE` (1)
+**Student:** Muhammad Umair Khan · 2023F-BCET-010
+**Institute:** Sir Syed University of Engineering and Technology, Karachi
+**Project:** SIFTA — AI Misinformation Detection Platform (Final Year Project 2025–26)
+
+---
+
+## Model Description
+
+This model fine-tunes [XLM-RoBERTa base](https://huggingface.co/xlm-roberta-base)
+on a combined English + Urdu fake news corpus. It is the core ML component of
+SIFTA — a multi-modal AI misinformation detection platform that also performs
+voice clone detection, document forgery analysis, and live source verification.
+
+The raw probability output is passed through SIFTA's Decision Engine alongside
+live source verification signals to produce a five-class Trust Score verdict:
+Verified Real · Likely Real · Uncertain · Unable to Verify · Likely Fake.
+
+---
+
+## Training Data
+
+| Dataset | Samples (train) | Language |
+|---|---|---|
+| ISOT | 32482 | English |
+| UrduFakeNews-AxToGrind | 8054 | Urdu |
+| **Total** | **40536** | EN + UR |
+
+### Dataset details
+
+**Urdu — Ax-to-Grind UrduFakeNews Corpus**
+- 10,106 news items from Pakistani social media and news outlets
+- Binary labels: TRUE / FAKE
+- Near-balanced: ~5,039 FAKE / ~4,997 TRUE
+- Average length: 351 characters per item
+
+**LIAR (Wang, 2017)**
+- 12,836 political statements from PolitiFact
+- 6-way labels mapped to binary:
+  - pants-fire / false / barely-true → FAKE
+  - half-true / mostly-true / true → TRUE
+
+**ISOT Fake News Dataset (University of Victoria)**
+- ~44,000 real/fake news articles
+- Real: Reuters. Fake: unreliable sources. Already binary labelled.
+
+---
+
+## Training Configuration
+
+| Parameter | Value |
+|---|---|
+| Base model | xlm-roberta-base |
+| Max token length | 256 |
+| Epochs | 5 |
+| Batch size | 16 x 2 grad accum = 32 effective |
+| Learning rate | 2e-05 |
+| LR schedule | Cosine with 10% linear warmup |
+| Weight decay | 0.01 |
+| Dropout | 0.1 |
+| Loss | CrossEntropyLoss with inverse-frequency class weights |
+| Early stopping | Patience 3 on val macro-F1 |
+| Mixed precision | fp16 on T4 GPU |
+| Random seed | 42 |
+
+---
+
+## Evaluation Results
+
+### Overall (held-out test set — 10% stratified)
+
+| Metric | Score |
+|---|---|
+| Accuracy | 0.9696 |
+| Macro F1 | 0.9695 |
+| TRUE (class 0) F1 | 0.9676 |
+| FAKE (class 1) F1 | 0.9713 |
+
+### Per-language macro-F1
+
+| Language | Macro F1 |
+|---|---|
+| EN | 0.9854 |
+| UR | 0.9053 |
+
+### Full classification report
+
+```
+              precision    recall  f1-score   support
+
+        TRUE     0.9632    0.9721    0.9676      2369
+        FAKE     0.9753    0.9674    0.9713      2698
+
+    accuracy                         0.9696      5067
+   macro avg     0.9693    0.9698    0.9695      5067
+weighted avg     0.9697    0.9696    0.9696      5067
+
+```
+### Confusion Matrix
+
+![Confusion Matrix](confusion_matrix.png)
+
+| | Predicted TRUE | Predicted FAKE |
+|---|---|---|
+| **Actual TRUE** | 2,303 (TN) | 66 (FP) |
+| **Actual FAKE** | 88 (FN) | 2,610 (TP) |
+
+- **TRUE recall:** 97.2%
+- **FAKE recall:** 96.7%
+
+---
+
+## Usage
+
+### HuggingFace Inference API (used by SIFTA backend)
+
+```python
+import httpx
+
+API_URL = "https://api-inference.huggingface.co/models/zodumair/sifta-fakenews-xlmr"
+headers = {"Authorization": "Bearer hf_your_token_here"}
+
+def classify(text):
+    return httpx.post(API_URL, headers=headers, json={'inputs': text}).json()
+
+print(classify("Scientists confirm new vaccine is 77% effective."))
+print(classify("حکومت نے خفیہ طور پر ملک کا سارا سونا بیرون ملک بھیج دیا"))
+```
+### Local inference
+
+```python
+from transformers import pipeline
+
+classifier = pipeline("text-classification", model="zodumair/sifta-fakenews-xlmr")
+print(classifier("SHOCKING: Scientists hiding cancer cure from the public."))
+# [{'label': 'FAKE', 'score': 0.93}]
+```
+---
+
+## Limitations
+
+- Optimised for headlines and short excerpts under 256 tokens
+- Urdu performance is lower than English due to smaller Urdu training set
+- Satire and opinion pieces may be misclassified
+- Use alongside source verification — not as a standalone decision tool
+
+---
+
+## Citation
+
+```
+@misc{sifta2026,
+  author      = {Muhammad Umair Khan},
+  title       = {SIFTA: AI Misinformation Detection Platform},
+  year        = {2026},
+  institution = {Sir Syed University of Engineering and Technology, Karachi},
+  note        = {Final Year Project, Batch 2023F}
+}
+```
+---
+*SIFTA FYP SSUET Karachi · Batch 2023F · 2023F-BCET-010*

checkpoint-2000/config.json

@@ -0,0 +1,38 @@
+{
+  "add_cross_attention": false,
+  "architectures": [
+    "XLMRobertaForSequenceClassification"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "bos_token_id": 0,
+  "classifier_dropout": null,
+  "dtype": "float32",
+  "eos_token_id": 2,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 768,
+  "id2label": {
+    "0": "TRUE",
+    "1": "FAKE"
+  },
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "is_decoder": false,
+  "label2id": {
+    "FAKE": 1,
+    "TRUE": 0
+  },
+  "layer_norm_eps": 1e-05,
+  "max_position_embeddings": 514,
+  "model_type": "xlm-roberta",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 12,
+  "output_past": true,
+  "pad_token_id": 1,
+  "position_embedding_type": "absolute",
+  "tie_word_embeddings": true,
+  "transformers_version": "5.12.1",
+  "type_vocab_size": 1,
+  "use_cache": false,
+  "vocab_size": 250002
+}

checkpoint-2000/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:01813e8912e5a6e170cb27bcdf268caf1ef3dde7339e114d0129c3bc844605af
+size 1112205008

checkpoint-2000/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e200b5ad8efb51e246ba634f7083181e99180c6bd9faf3adb60abee982ca6c9d
+size 2224532875

checkpoint-2000/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:81b9b59ff6c6e428c304847bdb69165ada5d6bf5d44b1a91844e7a3e386ab3e3
+size 14645

checkpoint-2000/scaler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:52fdbe03c6ee3fc7688e2854d1f87d9e898a5909ac109776c7735a230667f733
+size 1383

checkpoint-2000/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fd9d66016d89a43de634092495adc9298d13f83cbdd115358b14dbacb22d239b
+size 1465

checkpoint-2000/tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1d237c4974cfea16dffbc5035028fea5b47ad8b653af90c908b7a35bdc8b0d90
+size 17098184

checkpoint-2000/tokenizer_config.json

@@ -0,0 +1,15 @@
+{
+  "add_prefix_space": true,
+  "backend": "tokenizers",
+  "bos_token": "<s>",
+  "cls_token": "<s>",
+  "eos_token": "</s>",
+  "is_local": false,
+  "local_files_only": false,
+  "mask_token": "<mask>",
+  "model_max_length": 512,
+  "pad_token": "<pad>",
+  "sep_token": "</s>",
+  "tokenizer_class": "XLMRobertaTokenizer",
+  "unk_token": "<unk>"
+}

checkpoint-2000/trainer_state.json

@@ -0,0 +1,223 @@
+{
+  "best_global_step": 2000,
+  "best_metric": 0.9721,
+  "best_model_checkpoint": "./sifta-fakenews-xlmr/checkpoint-2000",
+  "epoch": 1.5785319652722967,
+  "eval_steps": 500,
+  "global_step": 2000,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.07892659826361484,
+      "grad_norm": 7.519743919372559,
+      "learning_rate": 3.1230283911671926e-06,
+      "loss": 1.3593978881835938,
+      "step": 100
+    },
+    {
+      "epoch": 0.15785319652722968,
+      "grad_norm": 44.37847137451172,
+      "learning_rate": 6.2776025236593065e-06,
+      "loss": 0.6375294876098633,
+      "step": 200
+    },
+    {
+      "epoch": 0.23677979479084452,
+      "grad_norm": 3.3530941009521484,
+      "learning_rate": 9.43217665615142e-06,
+      "loss": 0.3285886001586914,
+      "step": 300
+    },
+    {
+      "epoch": 0.31570639305445936,
+      "grad_norm": 4.9190993309021,
+      "learning_rate": 1.2586750788643533e-05,
+      "loss": 0.31189311981201173,
+      "step": 400
+    },
+    {
+      "epoch": 0.39463299131807417,
+      "grad_norm": 13.304952621459961,
+      "learning_rate": 1.574132492113565e-05,
+      "loss": 0.28032915115356444,
+      "step": 500
+    },
+    {
+      "epoch": 0.39463299131807417,
+      "eval_accuracy": 0.9513,
+      "eval_f1": 0.951,
+      "eval_loss": 0.11800144612789154,
+      "eval_runtime": 17.0078,
+      "eval_samples_per_second": 297.922,
+      "eval_steps_per_second": 9.349,
+      "step": 500
+    },
+    {
+      "epoch": 0.47355958958168903,
+      "grad_norm": 69.24589538574219,
+      "learning_rate": 1.889589905362776e-05,
+      "loss": 0.23949861526489258,
+      "step": 600
+    },
+    {
+      "epoch": 0.5524861878453039,
+      "grad_norm": 77.79669952392578,
+      "learning_rate": 1.9993585719507573e-05,
+      "loss": 0.26106679916381836,
+      "step": 700
+    },
+    {
+      "epoch": 0.6314127861089187,
+      "grad_norm": 9.278132438659668,
+      "learning_rate": 1.9958691794328414e-05,
+      "loss": 0.26748144149780273,
+      "step": 800
+    },
+    {
+      "epoch": 0.7103393843725335,
+      "grad_norm": 10.259608268737793,
+      "learning_rate": 1.9893564256008604e-05,
+      "loss": 0.2700900650024414,
+      "step": 900
+    },
+    {
+      "epoch": 0.7892659826361483,
+      "grad_norm": 1.331714153289795,
+      "learning_rate": 1.9798400825377246e-05,
+      "loss": 0.23025428771972656,
+      "step": 1000
+    },
+    {
+      "epoch": 0.7892659826361483,
+      "eval_accuracy": 0.9637,
+      "eval_f1": 0.9636,
+      "eval_loss": 0.11573408544063568,
+      "eval_runtime": 16.2817,
+      "eval_samples_per_second": 311.209,
+      "eval_steps_per_second": 9.766,
+      "step": 1000
+    },
+    {
+      "epoch": 0.8681925808997633,
+      "grad_norm": 0.1167285144329071,
+      "learning_rate": 1.9673490409291412e-05,
+      "loss": 0.22350259780883788,
+      "step": 1100
+    },
+    {
+      "epoch": 0.9471191791633781,
+      "grad_norm": 21.29085922241211,
+      "learning_rate": 1.9519212223543213e-05,
+      "loss": 0.24957496643066407,
+      "step": 1200
+    },
+    {
+      "epoch": 1.0260457774269929,
+      "grad_norm": 6.630244731903076,
+      "learning_rate": 1.933603464159779e-05,
+      "loss": 0.2235790252685547,
+      "step": 1300
+    },
+    {
+      "epoch": 1.1049723756906078,
+      "grad_norm": 3.18111252784729,
+      "learning_rate": 1.912451377265733e-05,
+      "loss": 0.17377975463867187,
+      "step": 1400
+    },
+    {
+      "epoch": 1.1838989739542225,
+      "grad_norm": 1.588996410369873,
+      "learning_rate": 1.888529177336798e-05,
+      "loss": 0.18398305892944336,
+      "step": 1500
+    },
+    {
+      "epoch": 1.1838989739542225,
+      "eval_accuracy": 0.9692,
+      "eval_f1": 0.9691,
+      "eval_loss": 0.08150316029787064,
+      "eval_runtime": 16.4485,
+      "eval_samples_per_second": 308.053,
+      "eval_steps_per_second": 9.667,
+      "step": 1500
+    },
+    {
+      "epoch": 1.2628255722178374,
+      "grad_norm": 2.542539119720459,
+      "learning_rate": 1.861909489829517e-05,
+      "loss": 0.20040679931640626,
+      "step": 1600
+    },
+    {
+      "epoch": 1.3417521704814521,
+      "grad_norm": 1.4010111093521118,
+      "learning_rate": 1.8326731295085838e-05,
+      "loss": 0.1535040855407715,
+      "step": 1700
+    },
+    {
+      "epoch": 1.420678768745067,
+      "grad_norm": 2.9616997241973877,
+      "learning_rate": 1.800908855101132e-05,
+      "loss": 0.1694472312927246,
+      "step": 1800
+    },
+    {
+      "epoch": 1.499605367008682,
+      "grad_norm": 8.78850269317627,
+      "learning_rate": 1.766713099833931e-05,
+      "loss": 0.18262014389038086,
+      "step": 1900
+    },
+    {
+      "epoch": 1.5785319652722967,
+      "grad_norm": 5.830899715423584,
+      "learning_rate": 1.730189678671552e-05,
+      "loss": 0.15452383995056151,
+      "step": 2000
+    },
+    {
+      "epoch": 1.5785319652722967,
+      "eval_accuracy": 0.9722,
+      "eval_f1": 0.9721,
+      "eval_loss": 0.08517046272754669,
+      "eval_runtime": 17.0389,
+      "eval_samples_per_second": 297.379,
+      "eval_steps_per_second": 9.332,
+      "step": 2000
+    }
+  ],
+  "logging_steps": 100,
+  "max_steps": 6335,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 5,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "EarlyStoppingCallback": {
+      "args": {
+        "early_stopping_patience": 3,
+        "early_stopping_threshold": 0.0
+      },
+      "attributes": {
+        "early_stopping_patience_counter": 0
+      }
+    },
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 8418501327298560.0,
+  "train_batch_size": 16,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-2000/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f6399d38d06ebeb459d7534aa989f92e96c4483d112f9ce5d7f695596af7f00d
+size 5201

checkpoint-3500/config.json

@@ -0,0 +1,38 @@
+{
+  "add_cross_attention": false,
+  "architectures": [
+    "XLMRobertaForSequenceClassification"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "bos_token_id": 0,
+  "classifier_dropout": null,
+  "dtype": "float32",
+  "eos_token_id": 2,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 768,
+  "id2label": {
+    "0": "TRUE",
+    "1": "FAKE"
+  },
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "is_decoder": false,
+  "label2id": {
+    "FAKE": 1,
+    "TRUE": 0
+  },
+  "layer_norm_eps": 1e-05,
+  "max_position_embeddings": 514,
+  "model_type": "xlm-roberta",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 12,
+  "output_past": true,
+  "pad_token_id": 1,
+  "position_embedding_type": "absolute",
+  "tie_word_embeddings": true,
+  "transformers_version": "5.12.1",
+  "type_vocab_size": 1,
+  "use_cache": false,
+  "vocab_size": 250002
+}

checkpoint-3500/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1dd37401db7582c6eff4c1cae02a6d72d7dca5176df8338bbb291a95731bf024
+size 1112205008

checkpoint-3500/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c78b337dd4d1e59808e2712116499d183ae4ba71920263c402015a08ac5435c1
+size 2224532875

checkpoint-3500/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b88525b5b027c40c3d45a6400a47ebb8f274fbe218c8a908da74a4d6d8f1675a
+size 14645

checkpoint-3500/scaler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c7eec9530a8e94fb0be5f11a65bc58a564506ff2ec7f427721f35bf80ff0727b
+size 1383

checkpoint-3500/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8f71cd3abc387204e04085410364338ff66913c9db2ae75282da4165bf5307c4
+size 1465

checkpoint-3500/tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1d237c4974cfea16dffbc5035028fea5b47ad8b653af90c908b7a35bdc8b0d90
+size 17098184

checkpoint-3500/tokenizer_config.json

@@ -0,0 +1,15 @@
+{
+  "add_prefix_space": true,
+  "backend": "tokenizers",
+  "bos_token": "<s>",
+  "cls_token": "<s>",
+  "eos_token": "</s>",
+  "is_local": false,
+  "local_files_only": false,
+  "mask_token": "<mask>",
+  "model_max_length": 512,
+  "pad_token": "<pad>",
+  "sep_token": "</s>",
+  "tokenizer_class": "XLMRobertaTokenizer",
+  "unk_token": "<unk>"
+}

checkpoint-3500/trainer_state.json

@@ -0,0 +1,358 @@
+{
+  "best_global_step": 2000,
+  "best_metric": 0.9721,
+  "best_model_checkpoint": "./sifta-fakenews-xlmr/checkpoint-2000",
+  "epoch": 2.7624309392265194,
+  "eval_steps": 500,
+  "global_step": 3500,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.07892659826361484,
+      "grad_norm": 7.519743919372559,
+      "learning_rate": 3.1230283911671926e-06,
+      "loss": 1.3593978881835938,
+      "step": 100
+    },
+    {
+      "epoch": 0.15785319652722968,
+      "grad_norm": 44.37847137451172,
+      "learning_rate": 6.2776025236593065e-06,
+      "loss": 0.6375294876098633,
+      "step": 200
+    },
+    {
+      "epoch": 0.23677979479084452,
+      "grad_norm": 3.3530941009521484,
+      "learning_rate": 9.43217665615142e-06,
+      "loss": 0.3285886001586914,
+      "step": 300
+    },
+    {
+      "epoch": 0.31570639305445936,
+      "grad_norm": 4.9190993309021,
+      "learning_rate": 1.2586750788643533e-05,
+      "loss": 0.31189311981201173,
+      "step": 400
+    },
+    {
+      "epoch": 0.39463299131807417,
+      "grad_norm": 13.304952621459961,
+      "learning_rate": 1.574132492113565e-05,
+      "loss": 0.28032915115356444,
+      "step": 500
+    },
+    {
+      "epoch": 0.39463299131807417,
+      "eval_accuracy": 0.9513,
+      "eval_f1": 0.951,
+      "eval_loss": 0.11800144612789154,
+      "eval_runtime": 17.0078,
+      "eval_samples_per_second": 297.922,
+      "eval_steps_per_second": 9.349,
+      "step": 500
+    },
+    {
+      "epoch": 0.47355958958168903,
+      "grad_norm": 69.24589538574219,
+      "learning_rate": 1.889589905362776e-05,
+      "loss": 0.23949861526489258,
+      "step": 600
+    },
+    {
+      "epoch": 0.5524861878453039,
+      "grad_norm": 77.79669952392578,
+      "learning_rate": 1.9993585719507573e-05,
+      "loss": 0.26106679916381836,
+      "step": 700
+    },
+    {
+      "epoch": 0.6314127861089187,
+      "grad_norm": 9.278132438659668,
+      "learning_rate": 1.9958691794328414e-05,
+      "loss": 0.26748144149780273,
+      "step": 800
+    },
+    {
+      "epoch": 0.7103393843725335,
+      "grad_norm": 10.259608268737793,
+      "learning_rate": 1.9893564256008604e-05,
+      "loss": 0.2700900650024414,
+      "step": 900
+    },
+    {
+      "epoch": 0.7892659826361483,
+      "grad_norm": 1.331714153289795,
+      "learning_rate": 1.9798400825377246e-05,
+      "loss": 0.23025428771972656,
+      "step": 1000
+    },
+    {
+      "epoch": 0.7892659826361483,
+      "eval_accuracy": 0.9637,
+      "eval_f1": 0.9636,
+      "eval_loss": 0.11573408544063568,
+      "eval_runtime": 16.2817,
+      "eval_samples_per_second": 311.209,
+      "eval_steps_per_second": 9.766,
+      "step": 1000
+    },
+    {
+      "epoch": 0.8681925808997633,
+      "grad_norm": 0.1167285144329071,
+      "learning_rate": 1.9673490409291412e-05,
+      "loss": 0.22350259780883788,
+      "step": 1100
+    },
+    {
+      "epoch": 0.9471191791633781,
+      "grad_norm": 21.29085922241211,
+      "learning_rate": 1.9519212223543213e-05,
+      "loss": 0.24957496643066407,
+      "step": 1200
+    },
+    {
+      "epoch": 1.0260457774269929,
+      "grad_norm": 6.630244731903076,
+      "learning_rate": 1.933603464159779e-05,
+      "loss": 0.2235790252685547,
+      "step": 1300
+    },
+    {
+      "epoch": 1.1049723756906078,
+      "grad_norm": 3.18111252784729,
+      "learning_rate": 1.912451377265733e-05,
+      "loss": 0.17377975463867187,
+      "step": 1400
+    },
+    {
+      "epoch": 1.1838989739542225,
+      "grad_norm": 1.588996410369873,
+      "learning_rate": 1.888529177336798e-05,
+      "loss": 0.18398305892944336,
+      "step": 1500
+    },
+    {
+      "epoch": 1.1838989739542225,
+      "eval_accuracy": 0.9692,
+      "eval_f1": 0.9691,
+      "eval_loss": 0.08150316029787064,
+      "eval_runtime": 16.4485,
+      "eval_samples_per_second": 308.053,
+      "eval_steps_per_second": 9.667,
+      "step": 1500
+    },
+    {
+      "epoch": 1.2628255722178374,
+      "grad_norm": 2.542539119720459,
+      "learning_rate": 1.861909489829517e-05,
+      "loss": 0.20040679931640626,
+      "step": 1600
+    },
+    {
+      "epoch": 1.3417521704814521,
+      "grad_norm": 1.4010111093521118,
+      "learning_rate": 1.8326731295085838e-05,
+      "loss": 0.1535040855407715,
+      "step": 1700
+    },
+    {
+      "epoch": 1.420678768745067,
+      "grad_norm": 2.9616997241973877,
+      "learning_rate": 1.800908855101132e-05,
+      "loss": 0.1694472312927246,
+      "step": 1800
+    },
+    {
+      "epoch": 1.499605367008682,
+      "grad_norm": 8.78850269317627,
+      "learning_rate": 1.766713099833931e-05,
+      "loss": 0.18262014389038086,
+      "step": 1900
+    },
+    {
+      "epoch": 1.5785319652722967,
+      "grad_norm": 5.830899715423584,
+      "learning_rate": 1.730189678671552e-05,
+      "loss": 0.15452383995056151,
+      "step": 2000
+    },
+    {
+      "epoch": 1.5785319652722967,
+      "eval_accuracy": 0.9722,
+      "eval_f1": 0.9721,
+      "eval_loss": 0.08517046272754669,
+      "eval_runtime": 17.0389,
+      "eval_samples_per_second": 297.379,
+      "eval_steps_per_second": 9.332,
+      "step": 2000
+    },
+    {
+      "epoch": 1.6574585635359116,
+      "grad_norm": 14.781915664672852,
+      "learning_rate": 1.6914494731443005e-05,
+      "loss": 0.17684141159057618,
+      "step": 2100
+    },
+    {
+      "epoch": 1.7363851617995265,
+      "grad_norm": 14.542154312133789,
+      "learning_rate": 1.6506100947227483e-05,
+      "loss": 0.16984281539916993,
+      "step": 2200
+    },
+    {
+      "epoch": 1.8153117600631412,
+      "grad_norm": 2.6693317890167236,
+      "learning_rate": 1.607795527760824e-05,
+      "loss": 0.16369407653808593,
+      "step": 2300
+    },
+    {
+      "epoch": 1.8942383583267561,
+      "grad_norm": 35.68703842163086,
+      "learning_rate": 1.5631357530914576e-05,
+      "loss": 0.16996910095214843,
+      "step": 2400
+    },
+    {
+      "epoch": 1.973164956590371,
+      "grad_norm": 8.133594512939453,
+      "learning_rate": 1.5167663534174963e-05,
+      "loss": 0.17478582382202149,
+      "step": 2500
+    },
+    {
+      "epoch": 1.973164956590371,
+      "eval_accuracy": 0.9722,
+      "eval_f1": 0.9721,
+      "eval_loss": 0.07743066549301147,
+      "eval_runtime": 17.1006,
+      "eval_samples_per_second": 296.305,
+      "eval_steps_per_second": 9.298,
+      "step": 2500
+    },
+    {
+      "epoch": 2.0520915548539858,
+      "grad_norm": 3.4671425819396973,
+      "learning_rate": 1.468828101695897e-05,
+      "loss": 0.14399107933044433,
+      "step": 2600
+    },
+    {
+      "epoch": 2.1310181531176005,
+      "grad_norm": 29.677196502685547,
+      "learning_rate": 1.4194665337648121e-05,
+      "loss": 0.15213555335998535,
+      "step": 2700
+    },
+    {
+      "epoch": 2.2099447513812156,
+      "grad_norm": 7.875465393066406,
+      "learning_rate": 1.368831506511036e-05,
+      "loss": 0.14315860748291015,
+      "step": 2800
+    },
+    {
+      "epoch": 2.2888713496448303,
+      "grad_norm": 20.07367706298828,
+      "learning_rate": 1.3170767429191709e-05,
+      "loss": 0.13439299583435058,
+      "step": 2900
+    },
+    {
+      "epoch": 2.367797947908445,
+      "grad_norm": 0.5996111631393433,
+      "learning_rate": 1.2643593653836975e-05,
+      "loss": 0.15156855583190917,
+      "step": 3000
+    },
+    {
+      "epoch": 2.367797947908445,
+      "eval_accuracy": 0.9714,
+      "eval_f1": 0.9712,
+      "eval_loss": 0.08889847248792648,
+      "eval_runtime": 17.0683,
+      "eval_samples_per_second": 296.866,
+      "eval_steps_per_second": 9.316,
+      "step": 3000
+    },
+    {
+      "epoch": 2.44672454617206,
+      "grad_norm": 9.044219970703125,
+      "learning_rate": 1.2108394187007715e-05,
+      "loss": 0.12777873992919922,
+      "step": 3100
+    },
+    {
+      "epoch": 2.525651144435675,
+      "grad_norm": 0.1941675990819931,
+      "learning_rate": 1.1566793841878931e-05,
+      "loss": 0.13878602981567384,
+      "step": 3200
+    },
+    {
+      "epoch": 2.6045777426992895,
+      "grad_norm": 8.129409790039062,
+      "learning_rate": 1.1020436864065368e-05,
+      "loss": 0.14199438095092773,
+      "step": 3300
+    },
+    {
+      "epoch": 2.6835043409629042,
+      "grad_norm": 0.1269533634185791,
+      "learning_rate": 1.0470981939852842e-05,
+      "loss": 0.12948278427124024,
+      "step": 3400
+    },
+    {
+      "epoch": 2.7624309392265194,
+      "grad_norm": 4.474359512329102,
+      "learning_rate": 9.92009716058909e-06,
+      "loss": 0.12795443534851075,
+      "step": 3500
+    },
+    {
+      "epoch": 2.7624309392265194,
+      "eval_accuracy": 0.9722,
+      "eval_f1": 0.972,
+      "eval_loss": 0.08592783659696579,
+      "eval_runtime": 17.0846,
+      "eval_samples_per_second": 296.584,
+      "eval_steps_per_second": 9.307,
+      "step": 3500
+    }
+  ],
+  "logging_steps": 100,
+  "max_steps": 6335,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 5,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "EarlyStoppingCallback": {
+      "args": {
+        "early_stopping_patience": 3,
+        "early_stopping_threshold": 0.0
+      },
+      "attributes": {
+        "early_stopping_patience_counter": 3
+      }
+    },
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 1.473211421171712e+16,
+  "train_batch_size": 16,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-3500/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f6399d38d06ebeb459d7534aa989f92e96c4483d112f9ce5d7f695596af7f00d
+size 5201

config.json

@@ -0,0 +1,38 @@
+{
+  "add_cross_attention": false,
+  "architectures": [
+    "XLMRobertaForSequenceClassification"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "bos_token_id": 0,
+  "classifier_dropout": null,
+  "dtype": "float32",
+  "eos_token_id": 2,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 768,
+  "id2label": {
+    "0": "TRUE",
+    "1": "FAKE"
+  },
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "is_decoder": false,
+  "label2id": {
+    "FAKE": 1,
+    "TRUE": 0
+  },
+  "layer_norm_eps": 1e-05,
+  "max_position_embeddings": 514,
+  "model_type": "xlm-roberta",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 12,
+  "output_past": true,
+  "pad_token_id": 1,
+  "position_embedding_type": "absolute",
+  "tie_word_embeddings": true,
+  "transformers_version": "5.12.1",
+  "type_vocab_size": 1,
+  "use_cache": false,
+  "vocab_size": 250002
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:01813e8912e5a6e170cb27bcdf268caf1ef3dde7339e114d0129c3bc844605af
+size 1112205008

tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1d237c4974cfea16dffbc5035028fea5b47ad8b653af90c908b7a35bdc8b0d90
+size 17098184

tokenizer_config.json

@@ -0,0 +1,15 @@
+{
+  "add_prefix_space": true,
+  "backend": "tokenizers",
+  "bos_token": "<s>",
+  "cls_token": "<s>",
+  "eos_token": "</s>",
+  "is_local": false,
+  "local_files_only": false,
+  "mask_token": "<mask>",
+  "model_max_length": 512,
+  "pad_token": "<pad>",
+  "sep_token": "</s>",
+  "tokenizer_class": "XLMRobertaTokenizer",
+  "unk_token": "<unk>"
+}