Upload folder using huggingface_hub

README.md

@@ -0,0 +1,165 @@
+---
+language: en
+license: cc-by-nc-sa-4.0
+base_model: ProsusAI/finbert
+pipeline_tag: text-classification
+tags:
+  - finance
+  - financial-nlp
+  - sentiment-analysis
+  - macroeconomics
+  - bert
+metrics:
+  - f1
+  - accuracy
+---
+
+# finbert-macro
+
+FinBERT fine-tuned to read macroeconomic headlines by their effect on equities, without losing the
+base model's grip on company news.
+
+ProsusAI/finbert scores 39.0 macro-F1 on the macro test set, which is near chance for three classes.
+This checkpoint scores 68.8, and corporate news improves at the same time, 70.2 to 84.0.
+
+## Labels
+
+`0 = positive`, `1 = negative`, `2 = neutral`, the same ids as the base model.
+
+On macro text the label is market impact, not tone. Dovish maps to positive, hawkish maps to
+negative. A hot CPI print or a strong payrolls number comes back `negative` here because it implies
+tighter policy. If you want literal sentiment, use ProsusAI/finbert instead.
+
+## Usage
+
+```python
+from transformers import AutoTokenizer, AutoModelForSequenceClassification
+import torch
+
+name = "ryousuf569/finbert-macro"
+tok = AutoTokenizer.from_pretrained(name)
+model = AutoModelForSequenceClassification.from_pretrained(name).eval()
+
+texts = ["Powell signals rate cuts are coming as price pressures ease",
+         "Payrolls blow past forecasts, wage growth accelerates",
+         "GDP growth unchanged from prior quarter"]
+with torch.no_grad():
+    logits = model(**tok(texts, padding=True, truncation=True,
+                         max_length=64, return_tensors="pt")).logits
+for t, p in zip(texts, logits.argmax(-1).tolist()):
+    print(model.config.id2label[p], "|", t)
+
+# positive | Powell signals rate cuts are coming as price pressures ease
+# negative | Payrolls blow past forecasts, wage growth accelerates
+# neutral  | GDP growth unchanged from prior quarter
+```
+
+"Jobless claims rise, cooling a tight labor market" also comes back positive, which is the mapping
+working as intended rather than a bug.
+
+## Results
+
+Macro and corporate are always reported separately. Averaging them would hide a macro gain paid for
+by forgetting corporate news.
+
+| model | macro acc | macro F1 | corporate acc | corporate F1 |
+|---|---|---|---|---|
+| ProsusAI/finbert | 38.06 | 39.03 | 71.60 | 70.24 |
+| finbert-macro | 68.89 | 68.82 | 85.03 | 83.97 |
+| change | +30.83 | +29.79 | +13.43 | +13.73 |
+
+Macro test set is 6,101 real held-out rows, corporate is 4,169. No class collapsed on either set.
+Lowest per-class F1 is 65.9 on macro, 81.3 on corporate.
+
+## Seed replication
+
+One seed cannot separate a real gap from run-to-run noise, so the winning configuration (u6@32k: 6
+unfrozen encoder blocks, 32,000 training rows) and the runner-up (u4@20k: 4 blocks, 20,000 rows)
+were each retrained from scratch at seeds 42, 1 and 2.
+
+**u6@32k beats u4@20k by 1.87 points of macro-F1 and 1.67 points of corporate F1. Both gaps survive
+the change of seed, and both domains improve at once, so this is not a macro-for-corporate trade.**
+
+| config | seed 42 | seed 1 | seed 2 | mean | std | min to max |
+|---|---|---|---|---|---|---|
+| macro-F1, u4@20k | 67.59 | 67.29 | 67.41 | 67.43 | 0.15 | 67.29 to 67.59 |
+| macro-F1, u6@32k | 68.82 | 69.34 | 69.73 | 69.30 | 0.46 | 68.82 to 69.73 |
+| corporate F1, u4@20k | 81.39 | 82.06 | 82.32 | 81.92 | 0.48 | 81.39 to 82.32 |
+| corporate F1, u6@32k | 83.97 | 82.66 | 84.16 | 83.60 | 0.82 | 82.66 to 84.16 |
+
+The macro gap of 1.87 points is wider than the widest seed range of 0.91, and the corporate gap of
+1.67 is wider than its widest range of 1.50. Every u6@32k seed beats every u4@20k seed on both
+metrics. Three seeds gives a range, not a significance test, so there is no p-value here and none
+should be inferred.
+
+The published weights are the seed 42 run, the seed fixed before the replication study. Seed 2
+scored higher at 69.73, but picking it would be selecting a checkpoint on the test set.
+
+## What did not replicate
+
+Four other metrics moved more between seeds than between configurations, so they were dropped from
+the decision and are listed here only as description.
+
+| metric | gap between configs | widest seed range |
+|---|---|---|
+| probe accuracy (48 probes) | 8.33 | 14.58 |
+| minimal pairs solved (of 18) | 1.67 | 7.00 |
+| polarity-agreeing probe accuracy (14 probes) | 7.14 | 7.14 |
+| Financial PhraseBank F1 | 12.31 | 13.29 |
+
+The probe suite is small enough that this is structural. With 14 polarity-agreeing probes, one probe
+changing its answer is worth 7.14 points, which is the entire gap. More probes would fix this. More
+seeds would not.
+
+## Limitations
+
+The PhraseBank comparison is not clean. This model gets 73.1 PhraseBank macro-F1 against the base
+model's 96.3, but ProsusAI/finbert was itself fine-tuned on Financial PhraseBank, so 96.3 is close
+to its training score. PhraseBank was never trained on here and every pool was filtered against it
+for leakage. Read 73.1 as retained company-level ability rather than a 23-point regression.
+
+Two things changed at once between u6@32k and u4@20k, unfreeze depth and training rows. The gain is
+real and repeatable, but this experiment cannot say which of the two produced it.
+
+1,817 of the 20,800 macro training rows are synthetic templates. They are scored as a separate
+evaluation set and never folded into the reported macro number, but they are in the training mix.
+
+Scope is short English headlines and policy sentences, 64 tokens. Central-bank language from 1996 to
+2024 dominates the macro side. Not built for filings, transcripts, long documents or other
+languages. The output is a text label, not a market forecast.
+
+## Training
+
+32,000 rows, 20,800 macro and 11,200 corporate, sampled without replacement. The corporate share is
+an anchor against forgetting. Everything is frozen except the top 6 encoder blocks, the pooler and
+the head, since full fine-tuning is the fastest way to lose corporate competence. Cross-entropy with
+inverse-frequency class weights, lr 2e-5, weight decay 0.01, warmup 0.1, batch 32, fp16.
+
+Both a macro and a corporate validation set are scored each epoch. The selected epoch maximizes
+macro-F1 while corporate validation F1 stays within 0.01 of the base checkpoint. Epoch 5 of 7 won
+and its weights were restored, against a ceiling of 16 epochs with early stopping. Training took 145
+minutes on one GTX 1660 Ti (6 GB, Turing).
+
+On that card `torch.cuda.is_bf16_supported()` returns True but counts emulated bf16, which is
+unusably slow, so this was trained in fp16.
+
+## Data and license
+
+Macro pool: World Central Banks, 16,870 rows, CC BY-NC-SA 4.0. FOMC / Trillion Dollar Words, 1,660
+rows, CC BY-NC 4.0. FinMarBa headlines, 749 rows, MIT. Synthetic templates, 1,848 rows.
+
+Corporate pool: SEntFiN, Twitter financial news and FiQA, 14,591 rows, all MIT.
+
+Splits are date-blocked per source where dates exist and grouped-random otherwise.
+
+The largest macro source is CC BY-NC-SA 4.0, so this model is released under CC BY-NC-SA 4.0,
+non-commercial and share-alike. That is stricter than the base model's Apache-2.0 licence, and the
+restriction comes from the training data.
+
+## Reproducing the seed study
+
+```bash
+python seed_rep.py --criterion macro_floored --seeds 42 1 2 \
+  --config "u4@20k:unfreeze=4,total=20000" \
+  --config "u6@32k:unfreeze=6,total=32000" --yes
+```

config.json

@@ -0,0 +1,40 @@
+{
+  "add_cross_attention": false,
+  "architectures": [
+    "BertForSequenceClassification"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "bos_token_id": null,
+  "classifier_dropout": null,
+  "dtype": "float32",
+  "eos_token_id": null,
+  "gradient_checkpointing": false,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 768,
+  "id2label": {
+    "0": "positive",
+    "1": "negative",
+    "2": "neutral"
+  },
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "is_decoder": false,
+  "label2id": {
+    "negative": 1,
+    "neutral": 2,
+    "positive": 0
+  },
+  "layer_norm_eps": 1e-12,
+  "max_position_embeddings": 512,
+  "model_type": "bert",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 12,
+  "pad_token_id": 0,
+  "position_embedding_type": "absolute",
+  "tie_word_embeddings": true,
+  "transformers_version": "5.14.1",
+  "type_vocab_size": 2,
+  "use_cache": false,
+  "vocab_size": 30522
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e4f123018d76a4f6a0227298b27e7c7f684c4d603f4850d72c27bb4dca00de14
+size 437961724

run.json

@@ -0,0 +1,110 @@
+{
+  "tag": "mix65_unfreeze6_full32k",
+  "model_dir": "models\\mix65_unfreeze6_full32k",
+  "macro_share": 0.65,
+  "unfreeze_top_layers": 6,
+  "seed": 42,
+  "epochs_ceiling": 16,
+  "epochs_run": 7,
+  "criterion": "macro_floored",
+  "tolerance": 0.01,
+  "mix": {
+    "macro_share": 0.65,
+    "policy": "fixed_total",
+    "total": 32000,
+    "n_macro": 20800,
+    "n_corporate": 11200,
+    "macro_upsample": 1.0,
+    "corporate_upsample": 1.0,
+    "n_macro_synthetic": 1817,
+    "label_counts": {
+      "2": 13584,
+      "0": 10537,
+      "1": 7879
+    }
+  },
+  "precision": {
+    "fp16": true,
+    "bf16": false,
+    "reason": "NVIDIA GeForce GTX 1660 Ti sm_75 no native bf16"
+  },
+  "selection": {
+    "criterion": "macro_floored",
+    "corporate_floor": 0.712656632790316,
+    "best_score": 1.715931047218585,
+    "best_epoch": 5.0,
+    "history": [
+      {
+        "epoch": 1.0,
+        "step": 1000,
+        "macro_val_acc": 0.6313853698406212,
+        "macro_val_f1": 0.6331539523789201,
+        "corporate_val_acc": 0.8095009596928983,
+        "corporate_val_f1": 0.797987307011303,
+        "score": 1.6331539523789202,
+        "is_best": true
+      },
+      {
+        "epoch": 2.0,
+        "step": 2000,
+        "macro_val_acc": 0.6947282386595831,
+        "macro_val_f1": 0.6940865713786276,
+        "corporate_val_acc": 0.8406909788867563,
+        "corporate_val_f1": 0.8312950922765591,
+        "score": 1.6940865713786275,
+        "is_best": true
+      },
+      {
+        "epoch": 3.0,
+        "step": 3000,
+        "macro_val_acc": 0.7151614221495709,
+        "macro_val_f1": 0.7150946803506621,
+        "corporate_val_acc": 0.8522072936660269,
+        "corporate_val_f1": 0.8468924856094882,
+        "score": 1.7150946803506621,
+        "is_best": true
+      },
+      {
+        "epoch": 4.0,
+        "step": 4000,
+        "macro_val_acc": 0.7131181038005722,
+        "macro_val_f1": 0.7131539477621897,
+        "corporate_val_acc": 0.8574856046065259,
+        "corporate_val_f1": 0.8496864160901788,
+        "score": 1.7131539477621898,
+        "is_best": false
+      },
+      {
+        "epoch": 5.0,
+        "step": 5000,
+        "macro_val_acc": 0.7151614221495709,
+        "macro_val_f1": 0.7159310472185848,
+        "corporate_val_acc": 0.8555662188099808,
+        "corporate_val_f1": 0.8467967784591366,
+        "score": 1.715931047218585,
+        "is_best": true
+      },
+      {
+        "epoch": 6.0,
+        "step": 6000,
+        "macro_val_acc": 0.6975888843481814,
+        "macro_val_f1": 0.6972553572452421,
+        "corporate_val_acc": 0.8459692898272553,
+        "corporate_val_f1": 0.8387254784627061,
+        "score": 1.697255357245242,
+        "is_best": false
+      },
+      {
+        "epoch": 7.0,
+        "step": 7000,
+        "macro_val_acc": 0.7024928483857785,
+        "macro_val_f1": 0.7025371437055846,
+        "corporate_val_acc": 0.8550863723608445,
+        "corporate_val_f1": 0.8473720194671822,
+        "score": 1.7025371437055847,
+        "is_best": false
+      }
+    ]
+  },
+  "minutes": 144.56
+}

seed_replication.json

@@ -0,0 +1,380 @@
+{
+  "generated": "2026-08-20 11:17:21",
+  "rule": "macro and corporate are carried side by side as separate columns and are NEVER averaged",
+  "note": "Ranges over a handful of seeds. No significance test, no p-value.",
+  "seeds": [
+    42,
+    1,
+    2
+  ],
+  "tolerance": 0.01,
+  "precision": {
+    "fp16": true,
+    "bf16": false,
+    "reason": "NVIDIA GeForce GTX 1660 Ti sm_75 no native bf16"
+  },
+  "metrics": {
+    "macro_f1": {
+      "name": "macro real held-out F1",
+      "units": "pts",
+      "scale": 100.0
+    },
+    "corp_f1": {
+      "name": "corporate F1",
+      "units": "pts",
+      "scale": 100.0
+    },
+    "probe_acc": {
+      "name": "probe accuracy",
+      "units": "pts",
+      "scale": 100.0
+    },
+    "pairs_solved": {
+      "name": "minimal pairs solved",
+      "units": "pairs",
+      "scale": 1.0
+    },
+    "agree_acc": {
+      "name": "polarity-agreeing probe acc",
+      "units": "pts",
+      "scale": 100.0
+    },
+    "pb_f1": {
+      "name": "PhraseBank F1",
+      "units": "pts",
+      "scale": 100.0
+    }
+  },
+  "configs": [
+    {
+      "name": "u4@20k",
+      "macro_share": 0.65,
+      "unfreeze": 4,
+      "total": 20000,
+      "epochs": 16,
+      "criterion": "macro_floored",
+      "mix_policy": "fixed_total"
+    },
+    {
+      "name": "u6@32k",
+      "macro_share": 0.65,
+      "unfreeze": 6,
+      "total": 32000,
+      "epochs": 16,
+      "criterion": "macro_floored",
+      "mix_policy": "fixed_total"
+    }
+  ],
+  "per_seed": [
+    {
+      "config": "u4@20k",
+      "seed": 42,
+      "status": "LOADED",
+      "dir": "models/mix65_unfreeze4_macro_floored_s42",
+      "how": "run.json match in mix65_unfreeze4_macro_floored_s42",
+      "metrics": {
+        "macro_f1": 67.59165949395553,
+        "corp_f1": 81.38670066473745,
+        "probe_acc": 79.16666666666666,
+        "pairs_solved": 12.0,
+        "agree_acc": 57.14285714285714,
+        "pb_f1": 86.91964125668869
+      }
+    },
+    {
+      "config": "u4@20k",
+      "seed": 1,
+      "status": "TRAINED",
+      "dir": "models/rep_u4_20k_s1",
+      "how": "trained by this script, 102.94 min",
+      "metrics": {
+        "macro_f1": 67.28569186992823,
+        "corp_f1": 82.06353062408495,
+        "probe_acc": 70.83333333333334,
+        "pairs_solved": 7.0,
+        "agree_acc": 50.0,
+        "pb_f1": 85.065296617682
+      }
+    },
+    {
+      "config": "u4@20k",
+      "seed": 2,
+      "status": "TRAINED",
+      "dir": "models/rep_u4_20k_s2",
+      "how": "trained by this script, 136.91 min",
+      "metrics": {
+        "macro_f1": 67.41373179920018,
+        "corp_f1": 82.31776419246067,
+        "probe_acc": 64.58333333333334,
+        "pairs_solved": 5.0,
+        "agree_acc": 57.14285714285714,
+        "pb_f1": 73.63131856850693
+      }
+    },
+    {
+      "config": "u6@32k",
+      "seed": 42,
+      "status": "LOADED",
+      "dir": "models/mix65_unfreeze6_full32k",
+      "how": "run.json match in mix65_unfreeze6_full32k",
+      "metrics": {
+        "macro_f1": 68.81987346760383,
+        "corp_f1": 83.97251903033337,
+        "probe_acc": 77.08333333333334,
+        "pairs_solved": 8.0,
+        "agree_acc": 64.28571428571429,
+        "pb_f1": 73.07008355119325
+      }
+    },
+    {
+      "config": "u6@32k",
+      "seed": 1,
+      "status": "TRAINED",
+      "dir": "models/rep_u6_32k_s1",
+      "how": "trained by this script, 205.69 min",
+      "metrics": {
+        "macro_f1": 69.3386664919747,
+        "corp_f1": 82.66029840632898,
+        "probe_acc": 81.25,
+        "pairs_solved": 11.0,
+        "agree_acc": 57.14285714285714,
+        "pb_f1": 65.74260663190499
+      }
+    },
+    {
+      "config": "u6@32k",
+      "seed": 2,
+      "status": "TRAINED",
+      "dir": "models/rep_u6_32k_s2",
+      "how": "trained by this script, 246.78 min",
+      "metrics": {
+        "macro_f1": 69.728009428257,
+        "corp_f1": 84.15625933222256,
+        "probe_acc": 81.25,
+        "pairs_solved": 10.0,
+        "agree_acc": 64.28571428571429,
+        "pb_f1": 69.87313438147625
+      }
+    }
+  ],
+  "vanilla": {
+    "macro_f1": 39.03300821786394,
+    "corp_f1": 70.23790161183014,
+    "probe_acc": 41.66666666666667,
+    "pairs_solved": 6.0,
+    "agree_acc": 71.42857142857143,
+    "pb_f1": 96.24820403323632
+  },
+  "aggregate": {
+    "u4@20k": {
+      "macro_f1": {
+        "n": 3,
+        "mean": 67.43036105436131,
+        "std": 0.15366016345702269,
+        "min": 67.28569186992823,
+        "max": 67.59165949395553,
+        "range": 0.3059676240273035
+      },
+      "corp_f1": {
+        "n": 3,
+        "mean": 81.9226651604277,
+        "std": 0.4812505920023736,
+        "min": 81.38670066473745,
+        "max": 82.31776419246067,
+        "range": 0.9310635277232251
+      },
+      "probe_acc": {
+        "n": 3,
+        "mean": 71.52777777777779,
+        "std": 7.316426217258837,
+        "min": 64.58333333333334,
+        "max": 79.16666666666666,
+        "range": 14.583333333333314
+      },
+      "pairs_solved": {
+        "n": 3,
+        "mean": 8.0,
+        "std": 3.605551275463989,
+        "min": 5.0,
+        "max": 12.0,
+        "range": 7.0
+      },
+      "agree_acc": {
+        "n": 3,
+        "mean": 54.76190476190476,
+        "std": 4.12393049421161,
+        "min": 50.0,
+        "max": 57.14285714285714,
+        "range": 7.142857142857139
+      },
+      "pb_f1": {
+        "n": 3,
+        "mean": 81.8720854809592,
+        "std": 7.196688682239868,
+        "min": 73.63131856850693,
+        "max": 86.91964125668869,
+        "range": 13.288322688181765
+      }
+    },
+    "u6@32k": {
+      "macro_f1": {
+        "n": 3,
+        "mean": 69.29551646261184,
+        "std": 0.4556030888101787,
+        "min": 68.81987346760383,
+        "max": 69.728009428257,
+        "range": 0.9081359606531691
+      },
+      "corp_f1": {
+        "n": 3,
+        "mean": 83.59635892296164,
+        "std": 0.815841340417078,
+        "min": 82.66029840632898,
+        "max": 84.15625933222256,
+        "range": 1.4959609258935842
+      },
+      "probe_acc": {
+        "n": 3,
+        "mean": 79.86111111111111,
+        "std": 2.405626121623435,
+        "min": 77.08333333333334,
+        "max": 81.25,
+        "range": 4.166666666666657
+      },
+      "pairs_solved": {
+        "n": 3,
+        "mean": 9.666666666666666,
+        "std": 1.5275252316519468,
+        "min": 8.0,
+        "max": 11.0,
+        "range": 3.0
+      },
+      "agree_acc": {
+        "n": 3,
+        "mean": 61.904761904761905,
+        "std": 4.123930494211619,
+        "min": 57.14285714285714,
+        "max": 64.28571428571429,
+        "range": 7.142857142857153
+      },
+      "pb_f1": {
+        "n": 3,
+        "mean": 69.56194152152483,
+        "std": 3.673637195986982,
+        "min": 65.74260663190499,
+        "max": 73.07008355119325,
+        "range": 7.327476919288259
+      }
+    }
+  },
+  "comparisons": [
+    {
+      "a": "u4@20k",
+      "b": "u6@32k",
+      "metric": "macro_f1",
+      "units": "pts",
+      "gap": 1.86515540825053,
+      "spread": 0.9081359606531691,
+      "half_spread": 0.4540679803265846,
+      "n_seeds": 3,
+      "leader": "u6@32k",
+      "ranges_overlap": false,
+      "verdict": "RESOLVED",
+      "a_min": 67.28569186992823,
+      "a_max": 67.59165949395553,
+      "b_min": 68.81987346760383,
+      "b_max": 69.728009428257
+    },
+    {
+      "a": "u4@20k",
+      "b": "u6@32k",
+      "metric": "corp_f1",
+      "units": "pts",
+      "gap": 1.6736937625339436,
+      "spread": 1.4959609258935842,
+      "half_spread": 0.7479804629467921,
+      "n_seeds": 3,
+      "leader": "u6@32k",
+      "ranges_overlap": false,
+      "verdict": "RESOLVED",
+      "a_min": 81.38670066473745,
+      "a_max": 82.31776419246067,
+      "b_min": 82.66029840632898,
+      "b_max": 84.15625933222256
+    },
+    {
+      "a": "u4@20k",
+      "b": "u6@32k",
+      "metric": "probe_acc",
+      "units": "pts",
+      "gap": 8.333333333333329,
+      "spread": 14.583333333333314,
+      "half_spread": 7.291666666666657,
+      "n_seeds": 3,
+      "leader": null,
+      "ranges_overlap": true,
+      "verdict": "NOT resolved",
+      "a_min": 64.58333333333334,
+      "a_max": 79.16666666666666,
+      "b_min": 77.08333333333334,
+      "b_max": 81.25
+    },
+    {
+      "a": "u4@20k",
+      "b": "u6@32k",
+      "metric": "pairs_solved",
+      "units": "pairs",
+      "gap": 1.666666666666666,
+      "spread": 7.0,
+      "half_spread": 3.5,
+      "n_seeds": 3,
+      "leader": null,
+      "ranges_overlap": true,
+      "verdict": "NOT resolved",
+      "a_min": 5.0,
+      "a_max": 12.0,
+      "b_min": 8.0,
+      "b_max": 11.0
+    },
+    {
+      "a": "u4@20k",
+      "b": "u6@32k",
+      "metric": "agree_acc",
+      "units": "pts",
+      "gap": 7.142857142857146,
+      "spread": 7.142857142857153,
+      "half_spread": 3.5714285714285765,
+      "n_seeds": 3,
+      "leader": null,
+      "ranges_overlap": true,
+      "verdict": "NOT resolved",
+      "a_min": 50.0,
+      "a_max": 57.14285714285714,
+      "b_min": 57.14285714285714,
+      "b_max": 64.28571428571429
+    },
+    {
+      "a": "u4@20k",
+      "b": "u6@32k",
+      "metric": "pb_f1",
+      "units": "pts",
+      "gap": 12.31014395943437,
+      "spread": 13.288322688181765,
+      "half_spread": 6.644161344090882,
+      "n_seeds": 3,
+      "leader": null,
+      "ranges_overlap": false,
+      "verdict": "NOT resolved",
+      "a_min": 73.63131856850693,
+      "a_max": 86.91964125668869,
+      "b_min": 65.74260663190499,
+      "b_max": 73.07008355119325
+    }
+  ],
+  "unrankable_metrics": [
+    "probe_acc",
+    "pairs_solved",
+    "agree_acc",
+    "pb_f1"
+  ]
+}

tokenizer_config.json

@@ -0,0 +1,15 @@
+{
+  "backend": "tokenizers",
+  "cls_token": "[CLS]",
+  "do_lower_case": true,
+  "is_local": false,
+  "local_files_only": false,
+  "mask_token": "[MASK]",
+  "model_max_length": 512,
+  "pad_token": "[PAD]",
+  "sep_token": "[SEP]",
+  "strip_accents": null,
+  "tokenize_chinese_chars": true,
+  "tokenizer_class": "BertTokenizer",
+  "unk_token": "[UNK]"
+}