checkpoint-1000

qwen3.5-2b-100pct-masked-weighted-v2/checkpoint-1000/README.md

@@ -0,0 +1,207 @@
+---
+base_model: Qwen/Qwen3.5-2B
+library_name: peft
+pipeline_tag: text-generation
+tags:
+- base_model:adapter:Qwen/Qwen3.5-2B
+- lora
+- transformers
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.21.0

qwen3.5-2b-100pct-masked-weighted-v2/checkpoint-1000/adapter_config.json

@@ -0,0 +1,56 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": null,
+  "base_model_name_or_path": "Qwen/Qwen3.5-2B",
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "kasa_config": null,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.05,
+  "lora_ga_config": null,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "monteclora_config": null,
+  "peft_type": "LORA",
+  "peft_version": "0.21.0",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "q_proj",
+    "in_proj_b",
+    "o_proj",
+    "down_proj",
+    "v_proj",
+    "up_proj",
+    "out_proj",
+    "in_proj_z",
+    "k_proj",
+    "gate_proj",
+    "in_proj_qkv",
+    "in_proj_a"
+  ],
+  "target_parameters": null,
+  "task_type": "CAUSAL_LM",
+  "trainable_token_indices": null,
+  "use_bdlora": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false,
+  "velora_config": null
+}

qwen3.5-2b-100pct-masked-weighted-v2/checkpoint-1000/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:553036a1dd99e151d6841e41d70997633c92dce94b3b9d20022d28970e6bf775
+size 134604152

qwen3.5-2b-100pct-masked-weighted-v2/checkpoint-1000/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e9953e1c0a37725e09d083aad8edbeea67e9c7915cf73ef194614b266949d676
+size 269426431

qwen3.5-2b-100pct-masked-weighted-v2/checkpoint-1000/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b439dcb80c6fda7953c7f35026a6fd217e60f59e2c4a05d93a95d3fd100a66a1
+size 14645

qwen3.5-2b-100pct-masked-weighted-v2/checkpoint-1000/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:184f621bbfabb6cf7bc829ecc1d00613784e93e7add9a224dc6dcf5964c929ee
+size 1465

qwen3.5-2b-100pct-masked-weighted-v2/checkpoint-1000/trainer_state.json

@@ -0,0 +1,771 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 0.17944663145026524,
+  "eval_steps": 500,
+  "global_step": 1000,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.0017944663145026524,
+      "grad_norm": 2.403308153152466,
+      "learning_rate": 3.225806451612903e-06,
+      "loss": 1.210298442840576,
+      "step": 10
+    },
+    {
+      "epoch": 0.003588932629005305,
+      "grad_norm": 1.3837767839431763,
+      "learning_rate": 6.810035842293908e-06,
+      "loss": 1.238302230834961,
+      "step": 20
+    },
+    {
+      "epoch": 0.005383398943507958,
+      "grad_norm": 1.152842402458191,
+      "learning_rate": 1.039426523297491e-05,
+      "loss": 1.1493020057678223,
+      "step": 30
+    },
+    {
+      "epoch": 0.00717786525801061,
+      "grad_norm": 1.2282543182373047,
+      "learning_rate": 1.3978494623655914e-05,
+      "loss": 1.2698699951171875,
+      "step": 40
+    },
+    {
+      "epoch": 0.008972331572513263,
+      "grad_norm": 4.5811686515808105,
+      "learning_rate": 1.7562724014336916e-05,
+      "loss": 1.336978054046631,
+      "step": 50
+    },
+    {
+      "epoch": 0.010766797887015915,
+      "grad_norm": 0.7280638217926025,
+      "learning_rate": 2.1146953405017922e-05,
+      "loss": 0.9929692268371582,
+      "step": 60
+    },
+    {
+      "epoch": 0.012561264201518567,
+      "grad_norm": 0.8359633684158325,
+      "learning_rate": 2.4731182795698928e-05,
+      "loss": 0.987311840057373,
+      "step": 70
+    },
+    {
+      "epoch": 0.01435573051602122,
+      "grad_norm": 1.061148762702942,
+      "learning_rate": 2.831541218637993e-05,
+      "loss": 1.0299980163574218,
+      "step": 80
+    },
+    {
+      "epoch": 0.01615019683052387,
+      "grad_norm": 1.4723000526428223,
+      "learning_rate": 3.1899641577060935e-05,
+      "loss": 1.1110344886779786,
+      "step": 90
+    },
+    {
+      "epoch": 0.017944663145026525,
+      "grad_norm": 5.412300109863281,
+      "learning_rate": 3.548387096774194e-05,
+      "loss": 1.1978090286254883,
+      "step": 100
+    },
+    {
+      "epoch": 0.019739129459529176,
+      "grad_norm": 0.7945532202720642,
+      "learning_rate": 3.906810035842295e-05,
+      "loss": 0.8953464508056641,
+      "step": 110
+    },
+    {
+      "epoch": 0.02153359577403183,
+      "grad_norm": 0.9241706728935242,
+      "learning_rate": 4.265232974910394e-05,
+      "loss": 0.8933357238769531,
+      "step": 120
+    },
+    {
+      "epoch": 0.023328062088534482,
+      "grad_norm": 1.2824897766113281,
+      "learning_rate": 4.6236559139784944e-05,
+      "loss": 0.9129752159118653,
+      "step": 130
+    },
+    {
+      "epoch": 0.025122528403037133,
+      "grad_norm": 1.8574696779251099,
+      "learning_rate": 4.982078853046595e-05,
+      "loss": 1.0019177436828612,
+      "step": 140
+    },
+    {
+      "epoch": 0.026916994717539788,
+      "grad_norm": 4.865994930267334,
+      "learning_rate": 5.340501792114696e-05,
+      "loss": 1.1639716148376464,
+      "step": 150
+    },
+    {
+      "epoch": 0.02871146103204244,
+      "grad_norm": 0.8981343507766724,
+      "learning_rate": 5.6989247311827965e-05,
+      "loss": 0.8498629570007324,
+      "step": 160
+    },
+    {
+      "epoch": 0.03050592734654509,
+      "grad_norm": 0.8773536086082458,
+      "learning_rate": 6.057347670250897e-05,
+      "loss": 0.8652203559875489,
+      "step": 170
+    },
+    {
+      "epoch": 0.03230039366104774,
+      "grad_norm": 1.2524876594543457,
+      "learning_rate": 6.415770609318996e-05,
+      "loss": 0.9427430152893066,
+      "step": 180
+    },
+    {
+      "epoch": 0.034094859975550396,
+      "grad_norm": 1.7148643732070923,
+      "learning_rate": 6.774193548387096e-05,
+      "loss": 0.9581910133361816,
+      "step": 190
+    },
+    {
+      "epoch": 0.03588932629005305,
+      "grad_norm": 5.458459854125977,
+      "learning_rate": 7.132616487455197e-05,
+      "loss": 1.1793695449829102,
+      "step": 200
+    },
+    {
+      "epoch": 0.037683792604555705,
+      "grad_norm": 0.7586544752120972,
+      "learning_rate": 7.491039426523297e-05,
+      "loss": 0.8673212051391601,
+      "step": 210
+    },
+    {
+      "epoch": 0.03947825891905835,
+      "grad_norm": 0.8580747246742249,
+      "learning_rate": 7.849462365591398e-05,
+      "loss": 0.8158726692199707,
+      "step": 220
+    },
+    {
+      "epoch": 0.04127272523356101,
+      "grad_norm": 1.0292130708694458,
+      "learning_rate": 8.207885304659499e-05,
+      "loss": 0.8816629409790039,
+      "step": 230
+    },
+    {
+      "epoch": 0.04306719154806366,
+      "grad_norm": 1.2191781997680664,
+      "learning_rate": 8.566308243727598e-05,
+      "loss": 0.9256319999694824,
+      "step": 240
+    },
+    {
+      "epoch": 0.04486165786256631,
+      "grad_norm": 4.896722316741943,
+      "learning_rate": 8.924731182795699e-05,
+      "loss": 1.054029369354248,
+      "step": 250
+    },
+    {
+      "epoch": 0.046656124177068964,
+      "grad_norm": 0.6555370688438416,
+      "learning_rate": 9.2831541218638e-05,
+      "loss": 0.8240686416625976,
+      "step": 260
+    },
+    {
+      "epoch": 0.04845059049157162,
+      "grad_norm": 0.8895808458328247,
+      "learning_rate": 9.6415770609319e-05,
+      "loss": 0.7868500232696534,
+      "step": 270
+    },
+    {
+      "epoch": 0.050245056806074266,
+      "grad_norm": 0.9253164529800415,
+      "learning_rate": 0.0001,
+      "loss": 0.821414852142334,
+      "step": 280
+    },
+    {
+      "epoch": 0.05203952312057692,
+      "grad_norm": 1.134040355682373,
+      "learning_rate": 9.981110691348697e-05,
+      "loss": 0.9013632774353028,
+      "step": 290
+    },
+    {
+      "epoch": 0.053833989435079575,
+      "grad_norm": 3.446653366088867,
+      "learning_rate": 9.962221382697394e-05,
+      "loss": 1.0495364189147949,
+      "step": 300
+    },
+    {
+      "epoch": 0.05562845574958222,
+      "grad_norm": 0.7303311824798584,
+      "learning_rate": 9.94333207404609e-05,
+      "loss": 0.7989195346832275,
+      "step": 310
+    },
+    {
+      "epoch": 0.05742292206408488,
+      "grad_norm": 0.8196483850479126,
+      "learning_rate": 9.924442765394787e-05,
+      "loss": 0.7937421798706055,
+      "step": 320
+    },
+    {
+      "epoch": 0.05921738837858753,
+      "grad_norm": 0.9366778135299683,
+      "learning_rate": 9.905553456743484e-05,
+      "loss": 0.809019660949707,
+      "step": 330
+    },
+    {
+      "epoch": 0.06101185469309018,
+      "grad_norm": 1.2467617988586426,
+      "learning_rate": 9.88666414809218e-05,
+      "loss": 0.930391502380371,
+      "step": 340
+    },
+    {
+      "epoch": 0.06280632100759284,
+      "grad_norm": 3.261760950088501,
+      "learning_rate": 9.867774839440876e-05,
+      "loss": 1.0620959281921387,
+      "step": 350
+    },
+    {
+      "epoch": 0.06460078732209548,
+      "grad_norm": 0.6342303156852722,
+      "learning_rate": 9.848885530789574e-05,
+      "loss": 0.8134325981140137,
+      "step": 360
+    },
+    {
+      "epoch": 0.06639525363659814,
+      "grad_norm": 0.6608202457427979,
+      "learning_rate": 9.82999622213827e-05,
+      "loss": 0.789882230758667,
+      "step": 370
+    },
+    {
+      "epoch": 0.06818971995110079,
+      "grad_norm": 0.8570723533630371,
+      "learning_rate": 9.811106913486967e-05,
+      "loss": 0.7621271133422851,
+      "step": 380
+    },
+    {
+      "epoch": 0.06998418626560345,
+      "grad_norm": 1.162699818611145,
+      "learning_rate": 9.792217604835664e-05,
+      "loss": 0.8825998306274414,
+      "step": 390
+    },
+    {
+      "epoch": 0.0717786525801061,
+      "grad_norm": 3.485926866531372,
+      "learning_rate": 9.77332829618436e-05,
+      "loss": 1.0364310264587402,
+      "step": 400
+    },
+    {
+      "epoch": 0.07357311889460875,
+      "grad_norm": 0.5732496380805969,
+      "learning_rate": 9.754438987533056e-05,
+      "loss": 0.7694174289703369,
+      "step": 410
+    },
+    {
+      "epoch": 0.07536758520911141,
+      "grad_norm": 0.683313250541687,
+      "learning_rate": 9.735549678881754e-05,
+      "loss": 0.7631448745727539,
+      "step": 420
+    },
+    {
+      "epoch": 0.07716205152361405,
+      "grad_norm": 0.8606892824172974,
+      "learning_rate": 9.716660370230451e-05,
+      "loss": 0.786726999282837,
+      "step": 430
+    },
+    {
+      "epoch": 0.0789565178381167,
+      "grad_norm": 1.1524667739868164,
+      "learning_rate": 9.697771061579146e-05,
+      "loss": 0.8868392944335938,
+      "step": 440
+    },
+    {
+      "epoch": 0.08075098415261936,
+      "grad_norm": 3.888843297958374,
+      "learning_rate": 9.678881752927843e-05,
+      "loss": 0.949896240234375,
+      "step": 450
+    },
+    {
+      "epoch": 0.08254545046712201,
+      "grad_norm": 0.540230393409729,
+      "learning_rate": 9.65999244427654e-05,
+      "loss": 0.7518480300903321,
+      "step": 460
+    },
+    {
+      "epoch": 0.08433991678162467,
+      "grad_norm": 0.7087560296058655,
+      "learning_rate": 9.641103135625237e-05,
+      "loss": 0.7413692474365234,
+      "step": 470
+    },
+    {
+      "epoch": 0.08613438309612732,
+      "grad_norm": 0.7639017105102539,
+      "learning_rate": 9.622213826973934e-05,
+      "loss": 0.8329146385192872,
+      "step": 480
+    },
+    {
+      "epoch": 0.08792884941062996,
+      "grad_norm": 1.0710383653640747,
+      "learning_rate": 9.60332451832263e-05,
+      "loss": 0.8491949081420899,
+      "step": 490
+    },
+    {
+      "epoch": 0.08972331572513262,
+      "grad_norm": 3.929804563522339,
+      "learning_rate": 9.584435209671326e-05,
+      "loss": 1.014235496520996,
+      "step": 500
+    },
+    {
+      "epoch": 0.08972331572513262,
+      "step": 500,
+      "test_entity_exact_match": 0.0039603960396039604,
+      "test_entity_fix_rate": 0.2154554278736262,
+      "test_entity_fix_rate_lenient": 0.20545155993431855,
+      "test_entity_fix_rate_recoverable": 0.2082457720404283,
+      "test_entity_hallucination_rate": 0.006930693069306931,
+      "test_entity_premature_stop_rate": 0.04950495049504951,
+      "test_entity_preservation_rate": 0.8851729129805372,
+      "test_entity_preservation_rate_lenient": 0.8867249143604942,
+      "test_entity_targeted_score": 0.5503141704270818,
+      "test_entity_targeted_score_lenient": 0.5460882371474064,
+      "test_entity_wer": 0.25481154129439265,
+      "test_entity_wer_zwnj_normalized": 0.24980518995313886
+    },
+    {
+      "epoch": 0.08972331572513262,
+      "step": 500,
+      "test_typo_exact_match": 0.0111731843575419,
+      "test_typo_fix_rate": 0.28247382706475377,
+      "test_typo_fix_rate_lenient": 0.2757193127716829,
+      "test_typo_fix_rate_recoverable": 0.27923577577157255,
+      "test_typo_hallucination_rate": 0.008379888268156424,
+      "test_typo_premature_stop_rate": 0.05027932960893855,
+      "test_typo_preservation_rate": 0.9247640608002823,
+      "test_typo_preservation_rate_lenient": 0.9271587213220428,
+      "test_typo_targeted_score": 0.603618943932518,
+      "test_typo_targeted_score_lenient": 0.6014390170468629,
+      "test_typo_wer": 0.18398815450205508,
+      "test_typo_wer_zwnj_normalized": 0.178116463710398
+    },
+    {
+      "epoch": 0.08972331572513262,
+      "step": 500,
+      "test_best_wer": 0.21939984789822387
+    },
+    {
+      "epoch": 0.09151778203963527,
+      "grad_norm": 0.5897678732872009,
+      "learning_rate": 9.565545901020023e-05,
+      "loss": 0.739581823348999,
+      "step": 510
+    },
+    {
+      "epoch": 0.09331224835413793,
+      "grad_norm": 0.6559104919433594,
+      "learning_rate": 9.54665659236872e-05,
+      "loss": 0.7567477226257324,
+      "step": 520
+    },
+    {
+      "epoch": 0.09510671466864058,
+      "grad_norm": 0.805770993232727,
+      "learning_rate": 9.527767283717416e-05,
+      "loss": 0.7930418968200683,
+      "step": 530
+    },
+    {
+      "epoch": 0.09690118098314324,
+      "grad_norm": 1.1318767070770264,
+      "learning_rate": 9.508877975066113e-05,
+      "loss": 0.905422306060791,
+      "step": 540
+    },
+    {
+      "epoch": 0.09869564729764588,
+      "grad_norm": 2.9629955291748047,
+      "learning_rate": 9.48998866641481e-05,
+      "loss": 0.9469995498657227,
+      "step": 550
+    },
+    {
+      "epoch": 0.10049011361214853,
+      "grad_norm": 0.5787467360496521,
+      "learning_rate": 9.471099357763506e-05,
+      "loss": 0.7495344161987305,
+      "step": 560
+    },
+    {
+      "epoch": 0.10228457992665119,
+      "grad_norm": 0.612440288066864,
+      "learning_rate": 9.452210049112203e-05,
+      "loss": 0.7359298706054688,
+      "step": 570
+    },
+    {
+      "epoch": 0.10407904624115384,
+      "grad_norm": 0.8614351153373718,
+      "learning_rate": 9.4333207404609e-05,
+      "loss": 0.777510404586792,
+      "step": 580
+    },
+    {
+      "epoch": 0.1058735125556565,
+      "grad_norm": 0.9304432272911072,
+      "learning_rate": 9.414431431809596e-05,
+      "loss": 0.7901641845703125,
+      "step": 590
+    },
+    {
+      "epoch": 0.10766797887015915,
+      "grad_norm": 3.34104585647583,
+      "learning_rate": 9.395542123158293e-05,
+      "loss": 0.9468848228454589,
+      "step": 600
+    },
+    {
+      "epoch": 0.1094624451846618,
+      "grad_norm": 0.5620515942573547,
+      "learning_rate": 9.37665281450699e-05,
+      "loss": 0.730808162689209,
+      "step": 610
+    },
+    {
+      "epoch": 0.11125691149916445,
+      "grad_norm": 0.6676825284957886,
+      "learning_rate": 9.357763505855685e-05,
+      "loss": 0.72871413230896,
+      "step": 620
+    },
+    {
+      "epoch": 0.1130513778136671,
+      "grad_norm": 0.795124888420105,
+      "learning_rate": 9.338874197204383e-05,
+      "loss": 0.7552680492401123,
+      "step": 630
+    },
+    {
+      "epoch": 0.11484584412816976,
+      "grad_norm": 0.9903436303138733,
+      "learning_rate": 9.31998488855308e-05,
+      "loss": 0.8315870285034179,
+      "step": 640
+    },
+    {
+      "epoch": 0.11664031044267241,
+      "grad_norm": 3.1701719760894775,
+      "learning_rate": 9.301095579901776e-05,
+      "loss": 0.964801025390625,
+      "step": 650
+    },
+    {
+      "epoch": 0.11843477675717506,
+      "grad_norm": 0.5520787835121155,
+      "learning_rate": 9.282206271250473e-05,
+      "loss": 0.7416518688201904,
+      "step": 660
+    },
+    {
+      "epoch": 0.12022924307167772,
+      "grad_norm": 0.6222532987594604,
+      "learning_rate": 9.263316962599168e-05,
+      "loss": 0.758673620223999,
+      "step": 670
+    },
+    {
+      "epoch": 0.12202370938618036,
+      "grad_norm": 0.7712622880935669,
+      "learning_rate": 9.244427653947865e-05,
+      "loss": 0.721858024597168,
+      "step": 680
+    },
+    {
+      "epoch": 0.12381817570068301,
+      "grad_norm": 0.9517074823379517,
+      "learning_rate": 9.225538345296563e-05,
+      "loss": 0.8953560829162598,
+      "step": 690
+    },
+    {
+      "epoch": 0.12561264201518568,
+      "grad_norm": 3.454228162765503,
+      "learning_rate": 9.20664903664526e-05,
+      "loss": 0.9904054641723633,
+      "step": 700
+    },
+    {
+      "epoch": 0.12740710832968832,
+      "grad_norm": 0.5811975002288818,
+      "learning_rate": 9.187759727993955e-05,
+      "loss": 0.7146330833435058,
+      "step": 710
+    },
+    {
+      "epoch": 0.12920157464419096,
+      "grad_norm": 0.5948407649993896,
+      "learning_rate": 9.168870419342652e-05,
+      "loss": 0.723681116104126,
+      "step": 720
+    },
+    {
+      "epoch": 0.13099604095869363,
+      "grad_norm": 0.7696809768676758,
+      "learning_rate": 9.149981110691349e-05,
+      "loss": 0.7829360008239746,
+      "step": 730
+    },
+    {
+      "epoch": 0.13279050727319627,
+      "grad_norm": 1.0465209484100342,
+      "learning_rate": 9.131091802040047e-05,
+      "loss": 0.8694637298583985,
+      "step": 740
+    },
+    {
+      "epoch": 0.13458497358769894,
+      "grad_norm": 2.8256630897521973,
+      "learning_rate": 9.112202493388743e-05,
+      "loss": 0.9362977027893067,
+      "step": 750
+    },
+    {
+      "epoch": 0.13637943990220158,
+      "grad_norm": 0.5560015439987183,
+      "learning_rate": 9.093313184737439e-05,
+      "loss": 0.7227760791778565,
+      "step": 760
+    },
+    {
+      "epoch": 0.13817390621670425,
+      "grad_norm": 0.6404837965965271,
+      "learning_rate": 9.074423876086135e-05,
+      "loss": 0.7460755825042724,
+      "step": 770
+    },
+    {
+      "epoch": 0.1399683725312069,
+      "grad_norm": 0.6959487795829773,
+      "learning_rate": 9.055534567434832e-05,
+      "loss": 0.7522034645080566,
+      "step": 780
+    },
+    {
+      "epoch": 0.14176283884570953,
+      "grad_norm": 0.9924454092979431,
+      "learning_rate": 9.036645258783529e-05,
+      "loss": 0.8138758659362793,
+      "step": 790
+    },
+    {
+      "epoch": 0.1435573051602122,
+      "grad_norm": 2.9993672370910645,
+      "learning_rate": 9.017755950132225e-05,
+      "loss": 0.8775947570800782,
+      "step": 800
+    },
+    {
+      "epoch": 0.14535177147471484,
+      "grad_norm": 0.5515598058700562,
+      "learning_rate": 8.998866641480922e-05,
+      "loss": 0.7623902320861816,
+      "step": 810
+    },
+    {
+      "epoch": 0.1471462377892175,
+      "grad_norm": 0.6749187111854553,
+      "learning_rate": 8.979977332829619e-05,
+      "loss": 0.7583945274353028,
+      "step": 820
+    },
+    {
+      "epoch": 0.14894070410372015,
+      "grad_norm": 0.7289445400238037,
+      "learning_rate": 8.961088024178315e-05,
+      "loss": 0.704672384262085,
+      "step": 830
+    },
+    {
+      "epoch": 0.15073517041822282,
+      "grad_norm": 1.0323708057403564,
+      "learning_rate": 8.942198715527012e-05,
+      "loss": 0.8557832717895508,
+      "step": 840
+    },
+    {
+      "epoch": 0.15252963673272546,
+      "grad_norm": 2.91811466217041,
+      "learning_rate": 8.923309406875709e-05,
+      "loss": 0.945162582397461,
+      "step": 850
+    },
+    {
+      "epoch": 0.1543241030472281,
+      "grad_norm": 0.54680997133255,
+      "learning_rate": 8.904420098224405e-05,
+      "loss": 0.7189648628234864,
+      "step": 860
+    },
+    {
+      "epoch": 0.15611856936173077,
+      "grad_norm": 0.6474179625511169,
+      "learning_rate": 8.885530789573102e-05,
+      "loss": 0.731608533859253,
+      "step": 870
+    },
+    {
+      "epoch": 0.1579130356762334,
+      "grad_norm": 0.7434310913085938,
+      "learning_rate": 8.866641480921799e-05,
+      "loss": 0.7703955173492432,
+      "step": 880
+    },
+    {
+      "epoch": 0.15970750199073608,
+      "grad_norm": 0.9848499894142151,
+      "learning_rate": 8.847752172270494e-05,
+      "loss": 0.84183349609375,
+      "step": 890
+    },
+    {
+      "epoch": 0.16150196830523872,
+      "grad_norm": 2.488396644592285,
+      "learning_rate": 8.828862863619192e-05,
+      "loss": 0.8863388061523437,
+      "step": 900
+    },
+    {
+      "epoch": 0.16329643461974136,
+      "grad_norm": 0.5121757388114929,
+      "learning_rate": 8.809973554967889e-05,
+      "loss": 0.6949440956115722,
+      "step": 910
+    },
+    {
+      "epoch": 0.16509090093424403,
+      "grad_norm": 0.5928990244865417,
+      "learning_rate": 8.791084246316586e-05,
+      "loss": 0.7060844421386718,
+      "step": 920
+    },
+    {
+      "epoch": 0.16688536724874667,
+      "grad_norm": 0.6861441135406494,
+      "learning_rate": 8.772194937665282e-05,
+      "loss": 0.692753267288208,
+      "step": 930
+    },
+    {
+      "epoch": 0.16867983356324934,
+      "grad_norm": 0.902505099773407,
+      "learning_rate": 8.753305629013978e-05,
+      "loss": 0.8425265312194824,
+      "step": 940
+    },
+    {
+      "epoch": 0.17047429987775198,
+      "grad_norm": 2.991858720779419,
+      "learning_rate": 8.734416320362676e-05,
+      "loss": 0.9827096939086915,
+      "step": 950
+    },
+    {
+      "epoch": 0.17226876619225465,
+      "grad_norm": 0.5504608154296875,
+      "learning_rate": 8.715527011711372e-05,
+      "loss": 0.7078711986541748,
+      "step": 960
+    },
+    {
+      "epoch": 0.1740632325067573,
+      "grad_norm": 0.5990283489227295,
+      "learning_rate": 8.696637703060069e-05,
+      "loss": 0.7095432281494141,
+      "step": 970
+    },
+    {
+      "epoch": 0.17585769882125993,
+      "grad_norm": 0.7148527503013611,
+      "learning_rate": 8.677748394408764e-05,
+      "loss": 0.68252592086792,
+      "step": 980
+    },
+    {
+      "epoch": 0.1776521651357626,
+      "grad_norm": 0.9802461862564087,
+      "learning_rate": 8.658859085757461e-05,
+      "loss": 0.8042885780334472,
+      "step": 990
+    },
+    {
+      "epoch": 0.17944663145026524,
+      "grad_norm": 3.1613829135894775,
+      "learning_rate": 8.639969777106158e-05,
+      "loss": 0.9047529220581054,
+      "step": 1000
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 5573,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 1,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 2.024937921258094e+17,
+  "train_batch_size": 2,
+  "trial_name": null,
+  "trial_params": null
+}

qwen3.5-2b-100pct-masked-weighted-v2/checkpoint-1000/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:36324839c555b723546ca05f552d997ae3f0d4b9b817db61c11bbc2e65ec719c
+size 5265

qwen3.5-2b-100pct-masked-weighted-v2/tb/events.out.tfevents.1790605476.02b48645612f.4580.0

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c89e173c0ed135eb5eb1ac01be76e66fd4ba283cddcfbbbadc58112f683dccce
-size 20087
+oid sha256:15ea34a9ef8d233eb0a7e19bcd46b1e56ebd6c4ad2909d3d08777d2f43d18483
+size 32423

qwen3.5-2b-100pct-masked-weighted-v2/test_eval_entity/metrics.json

@@ -1,16 +1,16 @@
 {
-  "wer": 0.25481154129439265,
-  "wer_zwnj_normalized": 0.24980518995313886,
-  "exact_match": 0.0039603960396039604,
-  "fix_rate": 0.2154554278736262,
-  "preservation_rate": 0.8851729129805372,
-  "targeted_score": 0.5503141704270818,
-  "fix_rate_lenient": 0.20545155993431855,
-  "preservation_rate_lenient": 0.8867249143604942,
-  "targeted_score_lenient": 0.5460882371474064,
-  "fix_rate_recoverable": 0.2082457720404283,
+  "wer": 0.2840543973697975,
+  "wer_zwnj_normalized": 0.2787544966428625,
+  "exact_match": 0.0049504950495049506,
+  "fix_rate": 0.2379685004853305,
+  "preservation_rate": 0.8402123210551713,
+  "targeted_score": 0.5390904107702509,
+  "fix_rate_lenient": 0.2263382594417077,
+  "preservation_rate_lenient": 0.845387013586852,
+  "targeted_score_lenient": 0.5358626365142798,
+  "fix_rate_recoverable": 0.22926048233763635,
   "unrecoverable_targets": 471,
-  "hallucination_rate": 0.006930693069306931,
-  "premature_stop_rate": 0.04950495049504951,
+  "hallucination_rate": 0.0039603960396039604,
+  "premature_stop_rate": 0.07227722772277227,
   "n_examples": 1010
 }

qwen3.5-2b-100pct-masked-weighted-v2/test_eval_typo/metrics.json

@@ -1,16 +1,16 @@
 {
-  "wer": 0.18398815450205508,
-  "wer_zwnj_normalized": 0.178116463710398,
-  "exact_match": 0.0111731843575419,
-  "fix_rate": 0.28247382706475377,
-  "preservation_rate": 0.9247640608002823,
-  "targeted_score": 0.603618943932518,
-  "fix_rate_lenient": 0.2757193127716829,
-  "preservation_rate_lenient": 0.9271587213220428,
-  "targeted_score_lenient": 0.6014390170468629,
-  "fix_rate_recoverable": 0.27923577577157255,
+  "wer": 0.18281381634372368,
+  "wer_zwnj_normalized": 0.17625284011130685,
+  "exact_match": 0.008379888268156424,
+  "fix_rate": 0.32260566110895694,
+  "preservation_rate": 0.9064181342427895,
+  "targeted_score": 0.6145118976758732,
+  "fix_rate_lenient": 0.3113227075139723,
+  "preservation_rate_lenient": 0.9128893316534326,
+  "targeted_score_lenient": 0.6121060195837025,
+  "fix_rate_recoverable": 0.3153474700818812,
   "unrecoverable_targets": 68,
-  "hallucination_rate": 0.008379888268156424,
-  "premature_stop_rate": 0.05027932960893855,
+  "hallucination_rate": 0.00558659217877095,
+  "premature_stop_rate": 0.04748603351955307,
   "n_examples": 358
 }