Hugging Face's logo

PedramR
/
ASR_Post-processing

TensorBoard
Safetensors
Model card Files
xet
 Metrics Community
PedramR commited on
Commit
d6e28c4
·
verified ·
1 Parent(s): e2e6ff0

checkpoint-1000

Browse files
Files changed (30) hide show
  1. qwen3.5-2b-50pct-masked-weighted/best_checkpoint_wer/README.md +207 -0
  2. qwen3.5-2b-50pct-masked-weighted/best_checkpoint_wer/adapter_config.json +56 -0
  3. qwen3.5-2b-50pct-masked-weighted/best_checkpoint_wer/adapter_model.safetensors +3 -0
  4. qwen3.5-2b-50pct-masked-weighted/best_checkpoint_wer/best_metrics.json +4 -0
  5. qwen3.5-2b-50pct-masked-weighted/best_checkpoint_wer/optimizer.pt +3 -0
  6. qwen3.5-2b-50pct-masked-weighted/best_checkpoint_wer/rng_state.pth +3 -0
  7. qwen3.5-2b-50pct-masked-weighted/best_checkpoint_wer/scheduler.pt +3 -0
  8. qwen3.5-2b-50pct-masked-weighted/best_checkpoint_wer/trainer_state.json +734 -0
  9. qwen3.5-2b-50pct-masked-weighted/best_checkpoint_wer/training_args.bin +3 -0
  10. qwen3.5-2b-50pct-masked-weighted/checkpoint-1000/README.md +207 -0
  11. qwen3.5-2b-50pct-masked-weighted/checkpoint-1000/adapter_config.json +56 -0
  12. qwen3.5-2b-50pct-masked-weighted/checkpoint-1000/adapter_model.safetensors +3 -0
  13. qwen3.5-2b-50pct-masked-weighted/checkpoint-1000/optimizer.pt +3 -0
  14. qwen3.5-2b-50pct-masked-weighted/checkpoint-1000/rng_state.pth +3 -0
  15. qwen3.5-2b-50pct-masked-weighted/checkpoint-1000/scheduler.pt +3 -0
  16. qwen3.5-2b-50pct-masked-weighted/checkpoint-1000/trainer_state.json +734 -0
  17. qwen3.5-2b-50pct-masked-weighted/checkpoint-1000/training_args.bin +3 -0
  18. qwen3.5-2b-50pct-masked-weighted/config.yaml +61 -0
  19. qwen3.5-2b-50pct-masked-weighted/resolved_config.yaml +54 -0
  20. qwen3.5-2b-50pct-masked-weighted/tb/events.out.tfevents.1790426135.b0aacb224392.4711.0 +3 -0
  21. qwen3.5-2b-50pct-masked-weighted/test_eval_baseline/metrics.json +15 -0
  22. qwen3.5-2b-50pct-masked-weighted/test_eval_baseline/predictions.jsonl +0 -0
  23. qwen3.5-2b-50pct-masked-weighted/test_eval_entity/metrics.json +15 -0
  24. qwen3.5-2b-50pct-masked-weighted/test_eval_entity/predictions.jsonl +0 -0
  25. qwen3.5-2b-50pct-masked-weighted/test_eval_entity_baseline/metrics.json +15 -0
  26. qwen3.5-2b-50pct-masked-weighted/test_eval_entity_baseline/predictions.jsonl +0 -0
  27. qwen3.5-2b-50pct-masked-weighted/test_eval_typo/metrics.json +15 -0
  28. qwen3.5-2b-50pct-masked-weighted/test_eval_typo/predictions.jsonl +0 -0
  29. qwen3.5-2b-50pct-masked-weighted/test_eval_typo_baseline/metrics.json +15 -0
  30. qwen3.5-2b-50pct-masked-weighted/test_eval_typo_baseline/predictions.jsonl +0 -0
qwen3.5-2b-50pct-masked-weighted/best_checkpoint_wer/README.md ADDED
@@ -0,0 +1,207 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ base_model: Qwen/Qwen3.5-2B
3
+ library_name: peft
4
+ pipeline_tag: text-generation
5
+ tags:
6
+ - base_model:adapter:Qwen/Qwen3.5-2B
7
+ - lora
8
+ - transformers
9
+ ---
10
+
11
+ # Model Card for Model ID
12
+
13
+ <!-- Provide a quick summary of what the model is/does. -->
14
+
15
+
16
+
17
+ ## Model Details
18
+
19
+ ### Model Description
20
+
21
+ <!-- Provide a longer summary of what this model is. -->
22
+
23
+
24
+
25
+ - **Developed by:** [More Information Needed]
26
+ - **Funded by [optional]:** [More Information Needed]
27
+ - **Shared by [optional]:** [More Information Needed]
28
+ - **Model type:** [More Information Needed]
29
+ - **Language(s) (NLP):** [More Information Needed]
30
+ - **License:** [More Information Needed]
31
+ - **Finetuned from model [optional]:** [More Information Needed]
32
+
33
+ ### Model Sources [optional]
34
+
35
+ <!-- Provide the basic links for the model. -->
36
+
37
+ - **Repository:** [More Information Needed]
38
+ - **Paper [optional]:** [More Information Needed]
39
+ - **Demo [optional]:** [More Information Needed]
40
+
41
+ ## Uses
42
+
43
+ <!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
44
+
45
+ ### Direct Use
46
+
47
+ <!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
48
+
49
+ [More Information Needed]
50
+
51
+ ### Downstream Use [optional]
52
+
53
+ <!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
54
+
55
+ [More Information Needed]
56
+
57
+ ### Out-of-Scope Use
58
+
59
+ <!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
60
+
61
+ [More Information Needed]
62
+
63
+ ## Bias, Risks, and Limitations
64
+
65
+ <!-- This section is meant to convey both technical and sociotechnical limitations. -->
66
+
67
+ [More Information Needed]
68
+
69
+ ### Recommendations
70
+
71
+ <!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
72
+
73
+ Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
74
+
75
+ ## How to Get Started with the Model
76
+
77
+ Use the code below to get started with the model.
78
+
79
+ [More Information Needed]
80
+
81
+ ## Training Details
82
+
83
+ ### Training Data
84
+
85
+ <!-- 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. -->
86
+
87
+ [More Information Needed]
88
+
89
+ ### Training Procedure
90
+
91
+ <!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
92
+
93
+ #### Preprocessing [optional]
94
+
95
+ [More Information Needed]
96
+
97
+
98
+ #### Training Hyperparameters
99
+
100
+ - **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
101
+
102
+ #### Speeds, Sizes, Times [optional]
103
+
104
+ <!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
105
+
106
+ [More Information Needed]
107
+
108
+ ## Evaluation
109
+
110
+ <!-- This section describes the evaluation protocols and provides the results. -->
111
+
112
+ ### Testing Data, Factors & Metrics
113
+
114
+ #### Testing Data
115
+
116
+ <!-- This should link to a Dataset Card if possible. -->
117
+
118
+ [More Information Needed]
119
+
120
+ #### Factors
121
+
122
+ <!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
123
+
124
+ [More Information Needed]
125
+
126
+ #### Metrics
127
+
128
+ <!-- These are the evaluation metrics being used, ideally with a description of why. -->
129
+
130
+ [More Information Needed]
131
+
132
+ ### Results
133
+
134
+ [More Information Needed]
135
+
136
+ #### Summary
137
+
138
+
139
+
140
+ ## Model Examination [optional]
141
+
142
+ <!-- Relevant interpretability work for the model goes here -->
143
+
144
+ [More Information Needed]
145
+
146
+ ## Environmental Impact
147
+
148
+ <!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
149
+
150
+ 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).
151
+
152
+ - **Hardware Type:** [More Information Needed]
153
+ - **Hours used:** [More Information Needed]
154
+ - **Cloud Provider:** [More Information Needed]
155
+ - **Compute Region:** [More Information Needed]
156
+ - **Carbon Emitted:** [More Information Needed]
157
+
158
+ ## Technical Specifications [optional]
159
+
160
+ ### Model Architecture and Objective
161
+
162
+ [More Information Needed]
163
+
164
+ ### Compute Infrastructure
165
+
166
+ [More Information Needed]
167
+
168
+ #### Hardware
169
+
170
+ [More Information Needed]
171
+
172
+ #### Software
173
+
174
+ [More Information Needed]
175
+
176
+ ## Citation [optional]
177
+
178
+ <!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
179
+
180
+ **BibTeX:**
181
+
182
+ [More Information Needed]
183
+
184
+ **APA:**
185
+
186
+ [More Information Needed]
187
+
188
+ ## Glossary [optional]
189
+
190
+ <!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
191
+
192
+ [More Information Needed]
193
+
194
+ ## More Information [optional]
195
+
196
+ [More Information Needed]
197
+
198
+ ## Model Card Authors [optional]
199
+
200
+ [More Information Needed]
201
+
202
+ ## Model Card Contact
203
+
204
+ [More Information Needed]
205
+ ### Framework versions
206
+
207
+ - PEFT 0.21.0
qwen3.5-2b-50pct-masked-weighted/best_checkpoint_wer/adapter_config.json ADDED
@@ -0,0 +1,56 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "alora_invocation_tokens": null,
3
+ "alpha_pattern": {},
4
+ "arrow_config": null,
5
+ "auto_mapping": null,
6
+ "base_model_name_or_path": "Qwen/Qwen3.5-2B",
7
+ "bias": "none",
8
+ "corda_config": null,
9
+ "ensure_weight_tying": false,
10
+ "eva_config": null,
11
+ "exclude_modules": null,
12
+ "fan_in_fan_out": false,
13
+ "inference_mode": true,
14
+ "init_lora_weights": true,
15
+ "kasa_config": null,
16
+ "layer_replication": null,
17
+ "layers_pattern": null,
18
+ "layers_to_transform": null,
19
+ "loftq_config": {},
20
+ "lora_alpha": 64,
21
+ "lora_bias": false,
22
+ "lora_dropout": 0.05,
23
+ "lora_ga_config": null,
24
+ "megatron_config": null,
25
+ "megatron_core": "megatron.core",
26
+ "modules_to_save": null,
27
+ "monteclora_config": null,
28
+ "peft_type": "LORA",
29
+ "peft_version": "0.21.0",
30
+ "qalora_group_size": 16,
31
+ "r": 32,
32
+ "rank_pattern": {},
33
+ "revision": null,
34
+ "target_modules": [
35
+ "down_proj",
36
+ "o_proj",
37
+ "out_proj",
38
+ "v_proj",
39
+ "up_proj",
40
+ "in_proj_qkv",
41
+ "in_proj_b",
42
+ "in_proj_z",
43
+ "k_proj",
44
+ "q_proj",
45
+ "in_proj_a",
46
+ "gate_proj"
47
+ ],
48
+ "target_parameters": null,
49
+ "task_type": "CAUSAL_LM",
50
+ "trainable_token_indices": null,
51
+ "use_bdlora": null,
52
+ "use_dora": false,
53
+ "use_qalora": false,
54
+ "use_rslora": false,
55
+ "velora_config": null
56
+ }
qwen3.5-2b-50pct-masked-weighted/best_checkpoint_wer/adapter_model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5c355b9758d25da25cd198145fefdcbabf96c8381dc3e00789d628ff12d935b2
3
+ size 134604152
qwen3.5-2b-50pct-masked-weighted/best_checkpoint_wer/best_metrics.json ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ {
2
+ "step": 1000,
3
+ "wer": 0.6495641332727022
4
+ }
qwen3.5-2b-50pct-masked-weighted/best_checkpoint_wer/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:87a42ad7610cdce89a745a148c3b9f478ca80c2d6be4bbec707af20112091abe
3
+ size 269426431
qwen3.5-2b-50pct-masked-weighted/best_checkpoint_wer/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8372e63a426c4c2855a22732d06e8275ff2b5b5b4accbf12ba6a522510c9ee7d
3
+ size 14645
qwen3.5-2b-50pct-masked-weighted/best_checkpoint_wer/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:34532159d3083a204b1ddf71a34daca30b5d985bb7d15860a5e34b3afa0bfec2
3
+ size 1465
qwen3.5-2b-50pct-masked-weighted/best_checkpoint_wer/trainer_state.json ADDED
@@ -0,0 +1,734 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": null,
3
+ "best_metric": null,
4
+ "best_model_checkpoint": null,
5
+ "epoch": 0.17946474639388024,
6
+ "eval_steps": 500,
7
+ "global_step": 1000,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.0017946474639388024,
14
+ "grad_norm": 3.4698801040649414,
15
+ "learning_rate": 1.6129032258064516e-06,
16
+ "loss": 1.4609528541564942,
17
+ "step": 10
18
+ },
19
+ {
20
+ "epoch": 0.003589294927877605,
21
+ "grad_norm": 2.894972085952759,
22
+ "learning_rate": 3.405017921146954e-06,
23
+ "loss": 1.5060235977172851,
24
+ "step": 20
25
+ },
26
+ {
27
+ "epoch": 0.0053839423918164075,
28
+ "grad_norm": 2.3209266662597656,
29
+ "learning_rate": 5.197132616487455e-06,
30
+ "loss": 1.4531757354736328,
31
+ "step": 30
32
+ },
33
+ {
34
+ "epoch": 0.00717858985575521,
35
+ "grad_norm": 2.2354202270507812,
36
+ "learning_rate": 6.989247311827957e-06,
37
+ "loss": 1.5934924125671386,
38
+ "step": 40
39
+ },
40
+ {
41
+ "epoch": 0.008973237319694013,
42
+ "grad_norm": 6.718564033508301,
43
+ "learning_rate": 8.781362007168458e-06,
44
+ "loss": 1.6818513870239258,
45
+ "step": 50
46
+ },
47
+ {
48
+ "epoch": 0.010767884783632815,
49
+ "grad_norm": 1.1820789575576782,
50
+ "learning_rate": 1.0573476702508961e-05,
51
+ "loss": 1.2959606170654296,
52
+ "step": 60
53
+ },
54
+ {
55
+ "epoch": 0.012562532247571617,
56
+ "grad_norm": 1.3072574138641357,
57
+ "learning_rate": 1.2365591397849464e-05,
58
+ "loss": 1.1890168190002441,
59
+ "step": 70
60
+ },
61
+ {
62
+ "epoch": 0.01435717971151042,
63
+ "grad_norm": 1.6472913026809692,
64
+ "learning_rate": 1.4157706093189965e-05,
65
+ "loss": 1.2888777732849122,
66
+ "step": 80
67
+ },
68
+ {
69
+ "epoch": 0.01615182717544922,
70
+ "grad_norm": 2.2778408527374268,
71
+ "learning_rate": 1.5949820788530468e-05,
72
+ "loss": 1.382105255126953,
73
+ "step": 90
74
+ },
75
+ {
76
+ "epoch": 0.017946474639388026,
77
+ "grad_norm": 6.462679386138916,
78
+ "learning_rate": 1.774193548387097e-05,
79
+ "loss": 1.3651474952697753,
80
+ "step": 100
81
+ },
82
+ {
83
+ "epoch": 0.019741122103326826,
84
+ "grad_norm": 1.3327974081039429,
85
+ "learning_rate": 1.9534050179211473e-05,
86
+ "loss": 1.2019264221191406,
87
+ "step": 110
88
+ },
89
+ {
90
+ "epoch": 0.02153576956726563,
91
+ "grad_norm": 1.625113606452942,
92
+ "learning_rate": 2.132616487455197e-05,
93
+ "loss": 1.1140929222106934,
94
+ "step": 120
95
+ },
96
+ {
97
+ "epoch": 0.023330417031204434,
98
+ "grad_norm": 2.114356279373169,
99
+ "learning_rate": 2.3118279569892472e-05,
100
+ "loss": 1.194014072418213,
101
+ "step": 130
102
+ },
103
+ {
104
+ "epoch": 0.025125064495143234,
105
+ "grad_norm": 2.738748073577881,
106
+ "learning_rate": 2.4910394265232977e-05,
107
+ "loss": 1.310903263092041,
108
+ "step": 140
109
+ },
110
+ {
111
+ "epoch": 0.02691971195908204,
112
+ "grad_norm": 7.28310489654541,
113
+ "learning_rate": 2.670250896057348e-05,
114
+ "loss": 1.341808032989502,
115
+ "step": 150
116
+ },
117
+ {
118
+ "epoch": 0.02871435942302084,
119
+ "grad_norm": 1.4109766483306885,
120
+ "learning_rate": 2.8494623655913982e-05,
121
+ "loss": 1.0624680519104004,
122
+ "step": 160
123
+ },
124
+ {
125
+ "epoch": 0.030509006886959643,
126
+ "grad_norm": 1.8665562868118286,
127
+ "learning_rate": 3.0286738351254483e-05,
128
+ "loss": 1.1256374359130858,
129
+ "step": 170
130
+ },
131
+ {
132
+ "epoch": 0.03230365435089844,
133
+ "grad_norm": 2.103581190109253,
134
+ "learning_rate": 3.207885304659498e-05,
135
+ "loss": 1.1144988059997558,
136
+ "step": 180
137
+ },
138
+ {
139
+ "epoch": 0.03409830181483725,
140
+ "grad_norm": 2.7976624965667725,
141
+ "learning_rate": 3.387096774193548e-05,
142
+ "loss": 1.2549245834350586,
143
+ "step": 190
144
+ },
145
+ {
146
+ "epoch": 0.03589294927877605,
147
+ "grad_norm": 6.750956058502197,
148
+ "learning_rate": 3.566308243727599e-05,
149
+ "loss": 1.2815749168395996,
150
+ "step": 200
151
+ },
152
+ {
153
+ "epoch": 0.03768759674271485,
154
+ "grad_norm": 1.6264805793762207,
155
+ "learning_rate": 3.7455197132616484e-05,
156
+ "loss": 1.0297048568725586,
157
+ "step": 210
158
+ },
159
+ {
160
+ "epoch": 0.03948224420665365,
161
+ "grad_norm": 1.6889760494232178,
162
+ "learning_rate": 3.924731182795699e-05,
163
+ "loss": 1.0584090232849122,
164
+ "step": 220
165
+ },
166
+ {
167
+ "epoch": 0.04127689167059246,
168
+ "grad_norm": 2.0039634704589844,
169
+ "learning_rate": 4.1039426523297493e-05,
170
+ "loss": 1.085252571105957,
171
+ "step": 230
172
+ },
173
+ {
174
+ "epoch": 0.04307153913453126,
175
+ "grad_norm": 2.6522042751312256,
176
+ "learning_rate": 4.283154121863799e-05,
177
+ "loss": 1.2651546478271485,
178
+ "step": 240
179
+ },
180
+ {
181
+ "epoch": 0.04486618659847006,
182
+ "grad_norm": 8.020312309265137,
183
+ "learning_rate": 4.4623655913978496e-05,
184
+ "loss": 1.324000644683838,
185
+ "step": 250
186
+ },
187
+ {
188
+ "epoch": 0.04666083406240887,
189
+ "grad_norm": 1.7270288467407227,
190
+ "learning_rate": 4.6415770609319e-05,
191
+ "loss": 1.0575682640075683,
192
+ "step": 260
193
+ },
194
+ {
195
+ "epoch": 0.04845548152634767,
196
+ "grad_norm": 1.5692397356033325,
197
+ "learning_rate": 4.82078853046595e-05,
198
+ "loss": 1.0243738174438477,
199
+ "step": 270
200
+ },
201
+ {
202
+ "epoch": 0.05025012899028647,
203
+ "grad_norm": 2.029719591140747,
204
+ "learning_rate": 5e-05,
205
+ "loss": 1.0656013488769531,
206
+ "step": 280
207
+ },
208
+ {
209
+ "epoch": 0.052044776454225276,
210
+ "grad_norm": 2.4044599533081055,
211
+ "learning_rate": 5.17921146953405e-05,
212
+ "loss": 1.1308915138244628,
213
+ "step": 290
214
+ },
215
+ {
216
+ "epoch": 0.05383942391816408,
217
+ "grad_norm": 7.324768543243408,
218
+ "learning_rate": 5.3584229390681005e-05,
219
+ "loss": 1.2482977867126466,
220
+ "step": 300
221
+ },
222
+ {
223
+ "epoch": 0.05563407138210288,
224
+ "grad_norm": 1.4643646478652954,
225
+ "learning_rate": 5.53763440860215e-05,
226
+ "loss": 0.984095287322998,
227
+ "step": 310
228
+ },
229
+ {
230
+ "epoch": 0.05742871884604168,
231
+ "grad_norm": 1.6681671142578125,
232
+ "learning_rate": 5.7168458781362014e-05,
233
+ "loss": 1.011151123046875,
234
+ "step": 320
235
+ },
236
+ {
237
+ "epoch": 0.059223366309980485,
238
+ "grad_norm": 1.8012951612472534,
239
+ "learning_rate": 5.8960573476702505e-05,
240
+ "loss": 1.0619875907897949,
241
+ "step": 330
242
+ },
243
+ {
244
+ "epoch": 0.061018013773919286,
245
+ "grad_norm": 2.7174854278564453,
246
+ "learning_rate": 6.0752688172043016e-05,
247
+ "loss": 1.1800737380981445,
248
+ "step": 340
249
+ },
250
+ {
251
+ "epoch": 0.06281266123785809,
252
+ "grad_norm": 7.283389568328857,
253
+ "learning_rate": 6.254480286738351e-05,
254
+ "loss": 1.2749577522277833,
255
+ "step": 350
256
+ },
257
+ {
258
+ "epoch": 0.06460730870179689,
259
+ "grad_norm": 1.4635604619979858,
260
+ "learning_rate": 6.433691756272402e-05,
261
+ "loss": 1.0051876068115235,
262
+ "step": 360
263
+ },
264
+ {
265
+ "epoch": 0.06640195616573569,
266
+ "grad_norm": 1.4372541904449463,
267
+ "learning_rate": 6.612903225806452e-05,
268
+ "loss": 1.0501286506652832,
269
+ "step": 370
270
+ },
271
+ {
272
+ "epoch": 0.0681966036296745,
273
+ "grad_norm": 1.5717954635620117,
274
+ "learning_rate": 6.792114695340503e-05,
275
+ "loss": 1.0269421577453612,
276
+ "step": 380
277
+ },
278
+ {
279
+ "epoch": 0.0699912510936133,
280
+ "grad_norm": 2.4193785190582275,
281
+ "learning_rate": 6.971326164874552e-05,
282
+ "loss": 1.018425464630127,
283
+ "step": 390
284
+ },
285
+ {
286
+ "epoch": 0.0717858985575521,
287
+ "grad_norm": 5.167453289031982,
288
+ "learning_rate": 7.150537634408602e-05,
289
+ "loss": 1.2043580055236816,
290
+ "step": 400
291
+ },
292
+ {
293
+ "epoch": 0.0735805460214909,
294
+ "grad_norm": 1.1322799921035767,
295
+ "learning_rate": 7.329749103942653e-05,
296
+ "loss": 0.954345417022705,
297
+ "step": 410
298
+ },
299
+ {
300
+ "epoch": 0.0753751934854297,
301
+ "grad_norm": 1.3257076740264893,
302
+ "learning_rate": 7.508960573476703e-05,
303
+ "loss": 0.9895898818969726,
304
+ "step": 420
305
+ },
306
+ {
307
+ "epoch": 0.0771698409493685,
308
+ "grad_norm": 1.317622184753418,
309
+ "learning_rate": 7.688172043010752e-05,
310
+ "loss": 0.9647544860839844,
311
+ "step": 430
312
+ },
313
+ {
314
+ "epoch": 0.0789644884133073,
315
+ "grad_norm": 1.8667112588882446,
316
+ "learning_rate": 7.867383512544804e-05,
317
+ "loss": 1.139955997467041,
318
+ "step": 440
319
+ },
320
+ {
321
+ "epoch": 0.08075913587724612,
322
+ "grad_norm": 5.289592266082764,
323
+ "learning_rate": 8.046594982078853e-05,
324
+ "loss": 1.1954397201538085,
325
+ "step": 450
326
+ },
327
+ {
328
+ "epoch": 0.08255378334118492,
329
+ "grad_norm": 1.0380669832229614,
330
+ "learning_rate": 8.225806451612904e-05,
331
+ "loss": 0.9445084571838379,
332
+ "step": 460
333
+ },
334
+ {
335
+ "epoch": 0.08434843080512372,
336
+ "grad_norm": 1.3310823440551758,
337
+ "learning_rate": 8.405017921146954e-05,
338
+ "loss": 0.9562747955322266,
339
+ "step": 470
340
+ },
341
+ {
342
+ "epoch": 0.08614307826906252,
343
+ "grad_norm": 1.417316198348999,
344
+ "learning_rate": 8.584229390681004e-05,
345
+ "loss": 0.9948525428771973,
346
+ "step": 480
347
+ },
348
+ {
349
+ "epoch": 0.08793772573300132,
350
+ "grad_norm": 1.8197181224822998,
351
+ "learning_rate": 8.763440860215054e-05,
352
+ "loss": 1.0975610733032226,
353
+ "step": 490
354
+ },
355
+ {
356
+ "epoch": 0.08973237319694012,
357
+ "grad_norm": 4.941095352172852,
358
+ "learning_rate": 8.942652329749104e-05,
359
+ "loss": 1.2107452392578124,
360
+ "step": 500
361
+ },
362
+ {
363
+ "epoch": 0.09152702066087894,
364
+ "grad_norm": 0.9819939732551575,
365
+ "learning_rate": 9.121863799283154e-05,
366
+ "loss": 0.9593785285949707,
367
+ "step": 510
368
+ },
369
+ {
370
+ "epoch": 0.09332166812481774,
371
+ "grad_norm": 1.1892472505569458,
372
+ "learning_rate": 9.301075268817204e-05,
373
+ "loss": 0.9651047706604003,
374
+ "step": 520
375
+ },
376
+ {
377
+ "epoch": 0.09511631558875654,
378
+ "grad_norm": 1.4324865341186523,
379
+ "learning_rate": 9.480286738351255e-05,
380
+ "loss": 0.9062895774841309,
381
+ "step": 530
382
+ },
383
+ {
384
+ "epoch": 0.09691096305269534,
385
+ "grad_norm": 1.6410809755325317,
386
+ "learning_rate": 9.659498207885304e-05,
387
+ "loss": 1.0246500968933105,
388
+ "step": 540
389
+ },
390
+ {
391
+ "epoch": 0.09870561051663414,
392
+ "grad_norm": 5.0956501960754395,
393
+ "learning_rate": 9.838709677419355e-05,
394
+ "loss": 1.1241219520568848,
395
+ "step": 550
396
+ },
397
+ {
398
+ "epoch": 0.10050025798057294,
399
+ "grad_norm": 0.951851487159729,
400
+ "learning_rate": 9.999055534567435e-05,
401
+ "loss": 0.9464516639709473,
402
+ "step": 560
403
+ },
404
+ {
405
+ "epoch": 0.10229490544451174,
406
+ "grad_norm": 1.1737598180770874,
407
+ "learning_rate": 9.989610880241783e-05,
408
+ "loss": 0.907046890258789,
409
+ "step": 570
410
+ },
411
+ {
412
+ "epoch": 0.10408955290845055,
413
+ "grad_norm": 1.3924814462661743,
414
+ "learning_rate": 9.980166225916132e-05,
415
+ "loss": 0.9977164268493652,
416
+ "step": 580
417
+ },
418
+ {
419
+ "epoch": 0.10588420037238935,
420
+ "grad_norm": 1.6580984592437744,
421
+ "learning_rate": 9.97072157159048e-05,
422
+ "loss": 1.0950872421264648,
423
+ "step": 590
424
+ },
425
+ {
426
+ "epoch": 0.10767884783632815,
427
+ "grad_norm": 4.493084907531738,
428
+ "learning_rate": 9.961276917264828e-05,
429
+ "loss": 1.211144256591797,
430
+ "step": 600
431
+ },
432
+ {
433
+ "epoch": 0.10947349530026695,
434
+ "grad_norm": 0.8613011240959167,
435
+ "learning_rate": 9.951832262939177e-05,
436
+ "loss": 0.9501323699951172,
437
+ "step": 610
438
+ },
439
+ {
440
+ "epoch": 0.11126814276420575,
441
+ "grad_norm": 1.1255444288253784,
442
+ "learning_rate": 9.942387608613525e-05,
443
+ "loss": 0.903736686706543,
444
+ "step": 620
445
+ },
446
+ {
447
+ "epoch": 0.11306279022814456,
448
+ "grad_norm": 1.4082947969436646,
449
+ "learning_rate": 9.932942954287873e-05,
450
+ "loss": 1.0084752082824706,
451
+ "step": 630
452
+ },
453
+ {
454
+ "epoch": 0.11485743769208336,
455
+ "grad_norm": 1.652286171913147,
456
+ "learning_rate": 9.923498299962222e-05,
457
+ "loss": 0.930430793762207,
458
+ "step": 640
459
+ },
460
+ {
461
+ "epoch": 0.11665208515602217,
462
+ "grad_norm": 5.117873668670654,
463
+ "learning_rate": 9.91405364563657e-05,
464
+ "loss": 1.091403865814209,
465
+ "step": 650
466
+ },
467
+ {
468
+ "epoch": 0.11844673261996097,
469
+ "grad_norm": 1.0188069343566895,
470
+ "learning_rate": 9.904608991310918e-05,
471
+ "loss": 0.883120059967041,
472
+ "step": 660
473
+ },
474
+ {
475
+ "epoch": 0.12024138008389977,
476
+ "grad_norm": 1.1442457437515259,
477
+ "learning_rate": 9.895164336985267e-05,
478
+ "loss": 0.9172245979309082,
479
+ "step": 670
480
+ },
481
+ {
482
+ "epoch": 0.12203602754783857,
483
+ "grad_norm": 1.3536146879196167,
484
+ "learning_rate": 9.885719682659615e-05,
485
+ "loss": 0.9197479248046875,
486
+ "step": 680
487
+ },
488
+ {
489
+ "epoch": 0.12383067501177737,
490
+ "grad_norm": 1.8869231939315796,
491
+ "learning_rate": 9.876275028333963e-05,
492
+ "loss": 1.0764457702636718,
493
+ "step": 690
494
+ },
495
+ {
496
+ "epoch": 0.12562532247571617,
497
+ "grad_norm": 3.6568493843078613,
498
+ "learning_rate": 9.866830374008312e-05,
499
+ "loss": 1.0616437911987304,
500
+ "step": 700
501
+ },
502
+ {
503
+ "epoch": 0.127419969939655,
504
+ "grad_norm": 0.8427063822746277,
505
+ "learning_rate": 9.85738571968266e-05,
506
+ "loss": 0.8856773376464844,
507
+ "step": 710
508
+ },
509
+ {
510
+ "epoch": 0.12921461740359377,
511
+ "grad_norm": 1.0265027284622192,
512
+ "learning_rate": 9.847941065357009e-05,
513
+ "loss": 0.907376766204834,
514
+ "step": 720
515
+ },
516
+ {
517
+ "epoch": 0.1310092648675326,
518
+ "grad_norm": 1.3209590911865234,
519
+ "learning_rate": 9.838496411031357e-05,
520
+ "loss": 1.0000056266784667,
521
+ "step": 730
522
+ },
523
+ {
524
+ "epoch": 0.13280391233147137,
525
+ "grad_norm": 1.7275080680847168,
526
+ "learning_rate": 9.829051756705705e-05,
527
+ "loss": 1.1050853729248047,
528
+ "step": 740
529
+ },
530
+ {
531
+ "epoch": 0.1345985597954102,
532
+ "grad_norm": 4.58080530166626,
533
+ "learning_rate": 9.819607102380054e-05,
534
+ "loss": 1.1597484588623046,
535
+ "step": 750
536
+ },
537
+ {
538
+ "epoch": 0.136393207259349,
539
+ "grad_norm": 0.8872320055961609,
540
+ "learning_rate": 9.810162448054402e-05,
541
+ "loss": 0.917822265625,
542
+ "step": 760
543
+ },
544
+ {
545
+ "epoch": 0.1381878547232878,
546
+ "grad_norm": 0.9287213087081909,
547
+ "learning_rate": 9.80071779372875e-05,
548
+ "loss": 0.8946282386779785,
549
+ "step": 770
550
+ },
551
+ {
552
+ "epoch": 0.1399825021872266,
553
+ "grad_norm": 1.183867335319519,
554
+ "learning_rate": 9.791273139403099e-05,
555
+ "loss": 0.9173077583312989,
556
+ "step": 780
557
+ },
558
+ {
559
+ "epoch": 0.1417771496511654,
560
+ "grad_norm": 1.6546437740325928,
561
+ "learning_rate": 9.781828485077447e-05,
562
+ "loss": 0.962445068359375,
563
+ "step": 790
564
+ },
565
+ {
566
+ "epoch": 0.1435717971151042,
567
+ "grad_norm": 4.307088375091553,
568
+ "learning_rate": 9.772383830751794e-05,
569
+ "loss": 1.1235834121704102,
570
+ "step": 800
571
+ },
572
+ {
573
+ "epoch": 0.145366444579043,
574
+ "grad_norm": 0.8552832007408142,
575
+ "learning_rate": 9.762939176426144e-05,
576
+ "loss": 0.8700121879577637,
577
+ "step": 810
578
+ },
579
+ {
580
+ "epoch": 0.1471610920429818,
581
+ "grad_norm": 0.9782463908195496,
582
+ "learning_rate": 9.753494522100492e-05,
583
+ "loss": 0.8827217102050782,
584
+ "step": 820
585
+ },
586
+ {
587
+ "epoch": 0.14895573950692062,
588
+ "grad_norm": 1.2572635412216187,
589
+ "learning_rate": 9.74404986777484e-05,
590
+ "loss": 0.9147042274475098,
591
+ "step": 830
592
+ },
593
+ {
594
+ "epoch": 0.1507503869708594,
595
+ "grad_norm": 1.6980787515640259,
596
+ "learning_rate": 9.734605213449189e-05,
597
+ "loss": 1.0121471405029296,
598
+ "step": 840
599
+ },
600
+ {
601
+ "epoch": 0.15254503443479822,
602
+ "grad_norm": 4.741994380950928,
603
+ "learning_rate": 9.725160559123536e-05,
604
+ "loss": 1.070129680633545,
605
+ "step": 850
606
+ },
607
+ {
608
+ "epoch": 0.154339681898737,
609
+ "grad_norm": 0.7885052561759949,
610
+ "learning_rate": 9.715715904797885e-05,
611
+ "loss": 0.9009537696838379,
612
+ "step": 860
613
+ },
614
+ {
615
+ "epoch": 0.15613432936267582,
616
+ "grad_norm": 0.9653502106666565,
617
+ "learning_rate": 9.706271250472234e-05,
618
+ "loss": 0.8701554298400879,
619
+ "step": 870
620
+ },
621
+ {
622
+ "epoch": 0.1579289768266146,
623
+ "grad_norm": 1.2959074974060059,
624
+ "learning_rate": 9.69682659614658e-05,
625
+ "loss": 0.9180045127868652,
626
+ "step": 880
627
+ },
628
+ {
629
+ "epoch": 0.15972362429055342,
630
+ "grad_norm": 1.5041567087173462,
631
+ "learning_rate": 9.68738194182093e-05,
632
+ "loss": 0.9027626037597656,
633
+ "step": 890
634
+ },
635
+ {
636
+ "epoch": 0.16151827175449224,
637
+ "grad_norm": 5.474041938781738,
638
+ "learning_rate": 9.677937287495277e-05,
639
+ "loss": 1.0739646911621095,
640
+ "step": 900
641
+ },
642
+ {
643
+ "epoch": 0.16331291921843102,
644
+ "grad_norm": 0.933229923248291,
645
+ "learning_rate": 9.668492633169627e-05,
646
+ "loss": 0.881598949432373,
647
+ "step": 910
648
+ },
649
+ {
650
+ "epoch": 0.16510756668236984,
651
+ "grad_norm": 0.997329592704773,
652
+ "learning_rate": 9.659047978843975e-05,
653
+ "loss": 0.8712461471557618,
654
+ "step": 920
655
+ },
656
+ {
657
+ "epoch": 0.16690221414630863,
658
+ "grad_norm": 1.319835901260376,
659
+ "learning_rate": 9.649603324518322e-05,
660
+ "loss": 0.8690865516662598,
661
+ "step": 930
662
+ },
663
+ {
664
+ "epoch": 0.16869686161024744,
665
+ "grad_norm": 1.8019057512283325,
666
+ "learning_rate": 9.640158670192672e-05,
667
+ "loss": 1.1501221656799316,
668
+ "step": 940
669
+ },
670
+ {
671
+ "epoch": 0.17049150907418623,
672
+ "grad_norm": 3.8484811782836914,
673
+ "learning_rate": 9.630714015867019e-05,
674
+ "loss": 1.1916112899780273,
675
+ "step": 950
676
+ },
677
+ {
678
+ "epoch": 0.17228615653812504,
679
+ "grad_norm": 0.8143914937973022,
680
+ "learning_rate": 9.621269361541369e-05,
681
+ "loss": 0.91885404586792,
682
+ "step": 960
683
+ },
684
+ {
685
+ "epoch": 0.17408080400206385,
686
+ "grad_norm": 1.0034821033477783,
687
+ "learning_rate": 9.611824707215716e-05,
688
+ "loss": 0.8618650436401367,
689
+ "step": 970
690
+ },
691
+ {
692
+ "epoch": 0.17587545146600264,
693
+ "grad_norm": 1.3009910583496094,
694
+ "learning_rate": 9.602380052890064e-05,
695
+ "loss": 0.8944579124450683,
696
+ "step": 980
697
+ },
698
+ {
699
+ "epoch": 0.17767009892994146,
700
+ "grad_norm": 1.4715969562530518,
701
+ "learning_rate": 9.592935398564414e-05,
702
+ "loss": 1.0547027587890625,
703
+ "step": 990
704
+ },
705
+ {
706
+ "epoch": 0.17946474639388024,
707
+ "grad_norm": 4.630854606628418,
708
+ "learning_rate": 9.583490744238761e-05,
709
+ "loss": 1.1706391334533692,
710
+ "step": 1000
711
+ }
712
+ ],
713
+ "logging_steps": 10,
714
+ "max_steps": 11146,
715
+ "num_input_tokens_seen": 0,
716
+ "num_train_epochs": 2,
717
+ "save_steps": 1000,
718
+ "stateful_callbacks": {
719
+ "TrainerControl": {
720
+ "args": {
721
+ "should_epoch_stop": false,
722
+ "should_evaluate": false,
723
+ "should_log": false,
724
+ "should_save": true,
725
+ "should_training_stop": false
726
+ },
727
+ "attributes": {}
728
+ }
729
+ },
730
+ "total_flos": 1.014956564093929e+17,
731
+ "train_batch_size": 2,
732
+ "trial_name": null,
733
+ "trial_params": null
734
+ }
qwen3.5-2b-50pct-masked-weighted/best_checkpoint_wer/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:62edea7eb1e0274b5104b0e4cbcf850d6b18e77a951e843e93b693eaf58caa36
3
+ size 5265
qwen3.5-2b-50pct-masked-weighted/checkpoint-1000/README.md ADDED
@@ -0,0 +1,207 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ base_model: Qwen/Qwen3.5-2B
3
+ library_name: peft
4
+ pipeline_tag: text-generation
5
+ tags:
6
+ - base_model:adapter:Qwen/Qwen3.5-2B
7
+ - lora
8
+ - transformers
9
+ ---
10
+
11
+ # Model Card for Model ID
12
+
13
+ <!-- Provide a quick summary of what the model is/does. -->
14
+
15
+
16
+
17
+ ## Model Details
18
+
19
+ ### Model Description
20
+
21
+ <!-- Provide a longer summary of what this model is. -->
22
+
23
+
24
+
25
+ - **Developed by:** [More Information Needed]
26
+ - **Funded by [optional]:** [More Information Needed]
27
+ - **Shared by [optional]:** [More Information Needed]
28
+ - **Model type:** [More Information Needed]
29
+ - **Language(s) (NLP):** [More Information Needed]
30
+ - **License:** [More Information Needed]
31
+ - **Finetuned from model [optional]:** [More Information Needed]
32
+
33
+ ### Model Sources [optional]
34
+
35
+ <!-- Provide the basic links for the model. -->
36
+
37
+ - **Repository:** [More Information Needed]
38
+ - **Paper [optional]:** [More Information Needed]
39
+ - **Demo [optional]:** [More Information Needed]
40
+
41
+ ## Uses
42
+
43
+ <!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
44
+
45
+ ### Direct Use
46
+
47
+ <!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
48
+
49
+ [More Information Needed]
50
+
51
+ ### Downstream Use [optional]
52
+
53
+ <!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
54
+
55
+ [More Information Needed]
56
+
57
+ ### Out-of-Scope Use
58
+
59
+ <!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
60
+
61
+ [More Information Needed]
62
+
63
+ ## Bias, Risks, and Limitations
64
+
65
+ <!-- This section is meant to convey both technical and sociotechnical limitations. -->
66
+
67
+ [More Information Needed]
68
+
69
+ ### Recommendations
70
+
71
+ <!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
72
+
73
+ Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
74
+
75
+ ## How to Get Started with the Model
76
+
77
+ Use the code below to get started with the model.
78
+
79
+ [More Information Needed]
80
+
81
+ ## Training Details
82
+
83
+ ### Training Data
84
+
85
+ <!-- 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. -->
86
+
87
+ [More Information Needed]
88
+
89
+ ### Training Procedure
90
+
91
+ <!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
92
+
93
+ #### Preprocessing [optional]
94
+
95
+ [More Information Needed]
96
+
97
+
98
+ #### Training Hyperparameters
99
+
100
+ - **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
101
+
102
+ #### Speeds, Sizes, Times [optional]
103
+
104
+ <!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
105
+
106
+ [More Information Needed]
107
+
108
+ ## Evaluation
109
+
110
+ <!-- This section describes the evaluation protocols and provides the results. -->
111
+
112
+ ### Testing Data, Factors & Metrics
113
+
114
+ #### Testing Data
115
+
116
+ <!-- This should link to a Dataset Card if possible. -->
117
+
118
+ [More Information Needed]
119
+
120
+ #### Factors
121
+
122
+ <!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
123
+
124
+ [More Information Needed]
125
+
126
+ #### Metrics
127
+
128
+ <!-- These are the evaluation metrics being used, ideally with a description of why. -->
129
+
130
+ [More Information Needed]
131
+
132
+ ### Results
133
+
134
+ [More Information Needed]
135
+
136
+ #### Summary
137
+
138
+
139
+
140
+ ## Model Examination [optional]
141
+
142
+ <!-- Relevant interpretability work for the model goes here -->
143
+
144
+ [More Information Needed]
145
+
146
+ ## Environmental Impact
147
+
148
+ <!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
149
+
150
+ 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).
151
+
152
+ - **Hardware Type:** [More Information Needed]
153
+ - **Hours used:** [More Information Needed]
154
+ - **Cloud Provider:** [More Information Needed]
155
+ - **Compute Region:** [More Information Needed]
156
+ - **Carbon Emitted:** [More Information Needed]
157
+
158
+ ## Technical Specifications [optional]
159
+
160
+ ### Model Architecture and Objective
161
+
162
+ [More Information Needed]
163
+
164
+ ### Compute Infrastructure
165
+
166
+ [More Information Needed]
167
+
168
+ #### Hardware
169
+
170
+ [More Information Needed]
171
+
172
+ #### Software
173
+
174
+ [More Information Needed]
175
+
176
+ ## Citation [optional]
177
+
178
+ <!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
179
+
180
+ **BibTeX:**
181
+
182
+ [More Information Needed]
183
+
184
+ **APA:**
185
+
186
+ [More Information Needed]
187
+
188
+ ## Glossary [optional]
189
+
190
+ <!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
191
+
192
+ [More Information Needed]
193
+
194
+ ## More Information [optional]
195
+
196
+ [More Information Needed]
197
+
198
+ ## Model Card Authors [optional]
199
+
200
+ [More Information Needed]
201
+
202
+ ## Model Card Contact
203
+
204
+ [More Information Needed]
205
+ ### Framework versions
206
+
207
+ - PEFT 0.21.0
qwen3.5-2b-50pct-masked-weighted/checkpoint-1000/adapter_config.json ADDED
@@ -0,0 +1,56 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "alora_invocation_tokens": null,
3
+ "alpha_pattern": {},
4
+ "arrow_config": null,
5
+ "auto_mapping": null,
6
+ "base_model_name_or_path": "Qwen/Qwen3.5-2B",
7
+ "bias": "none",
8
+ "corda_config": null,
9
+ "ensure_weight_tying": false,
10
+ "eva_config": null,
11
+ "exclude_modules": null,
12
+ "fan_in_fan_out": false,
13
+ "inference_mode": true,
14
+ "init_lora_weights": true,
15
+ "kasa_config": null,
16
+ "layer_replication": null,
17
+ "layers_pattern": null,
18
+ "layers_to_transform": null,
19
+ "loftq_config": {},
20
+ "lora_alpha": 64,
21
+ "lora_bias": false,
22
+ "lora_dropout": 0.05,
23
+ "lora_ga_config": null,
24
+ "megatron_config": null,
25
+ "megatron_core": "megatron.core",
26
+ "modules_to_save": null,
27
+ "monteclora_config": null,
28
+ "peft_type": "LORA",
29
+ "peft_version": "0.21.0",
30
+ "qalora_group_size": 16,
31
+ "r": 32,
32
+ "rank_pattern": {},
33
+ "revision": null,
34
+ "target_modules": [
35
+ "down_proj",
36
+ "o_proj",
37
+ "out_proj",
38
+ "v_proj",
39
+ "up_proj",
40
+ "in_proj_qkv",
41
+ "in_proj_b",
42
+ "in_proj_z",
43
+ "k_proj",
44
+ "q_proj",
45
+ "in_proj_a",
46
+ "gate_proj"
47
+ ],
48
+ "target_parameters": null,
49
+ "task_type": "CAUSAL_LM",
50
+ "trainable_token_indices": null,
51
+ "use_bdlora": null,
52
+ "use_dora": false,
53
+ "use_qalora": false,
54
+ "use_rslora": false,
55
+ "velora_config": null
56
+ }
qwen3.5-2b-50pct-masked-weighted/checkpoint-1000/adapter_model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5c355b9758d25da25cd198145fefdcbabf96c8381dc3e00789d628ff12d935b2
3
+ size 134604152
qwen3.5-2b-50pct-masked-weighted/checkpoint-1000/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:87a42ad7610cdce89a745a148c3b9f478ca80c2d6be4bbec707af20112091abe
3
+ size 269426431
qwen3.5-2b-50pct-masked-weighted/checkpoint-1000/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8372e63a426c4c2855a22732d06e8275ff2b5b5b4accbf12ba6a522510c9ee7d
3
+ size 14645
qwen3.5-2b-50pct-masked-weighted/checkpoint-1000/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:34532159d3083a204b1ddf71a34daca30b5d985bb7d15860a5e34b3afa0bfec2
3
+ size 1465
qwen3.5-2b-50pct-masked-weighted/checkpoint-1000/trainer_state.json ADDED
@@ -0,0 +1,734 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": null,
3
+ "best_metric": null,
4
+ "best_model_checkpoint": null,
5
+ "epoch": 0.17946474639388024,
6
+ "eval_steps": 500,
7
+ "global_step": 1000,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.0017946474639388024,
14
+ "grad_norm": 3.4698801040649414,
15
+ "learning_rate": 1.6129032258064516e-06,
16
+ "loss": 1.4609528541564942,
17
+ "step": 10
18
+ },
19
+ {
20
+ "epoch": 0.003589294927877605,
21
+ "grad_norm": 2.894972085952759,
22
+ "learning_rate": 3.405017921146954e-06,
23
+ "loss": 1.5060235977172851,
24
+ "step": 20
25
+ },
26
+ {
27
+ "epoch": 0.0053839423918164075,
28
+ "grad_norm": 2.3209266662597656,
29
+ "learning_rate": 5.197132616487455e-06,
30
+ "loss": 1.4531757354736328,
31
+ "step": 30
32
+ },
33
+ {
34
+ "epoch": 0.00717858985575521,
35
+ "grad_norm": 2.2354202270507812,
36
+ "learning_rate": 6.989247311827957e-06,
37
+ "loss": 1.5934924125671386,
38
+ "step": 40
39
+ },
40
+ {
41
+ "epoch": 0.008973237319694013,
42
+ "grad_norm": 6.718564033508301,
43
+ "learning_rate": 8.781362007168458e-06,
44
+ "loss": 1.6818513870239258,
45
+ "step": 50
46
+ },
47
+ {
48
+ "epoch": 0.010767884783632815,
49
+ "grad_norm": 1.1820789575576782,
50
+ "learning_rate": 1.0573476702508961e-05,
51
+ "loss": 1.2959606170654296,
52
+ "step": 60
53
+ },
54
+ {
55
+ "epoch": 0.012562532247571617,
56
+ "grad_norm": 1.3072574138641357,
57
+ "learning_rate": 1.2365591397849464e-05,
58
+ "loss": 1.1890168190002441,
59
+ "step": 70
60
+ },
61
+ {
62
+ "epoch": 0.01435717971151042,
63
+ "grad_norm": 1.6472913026809692,
64
+ "learning_rate": 1.4157706093189965e-05,
65
+ "loss": 1.2888777732849122,
66
+ "step": 80
67
+ },
68
+ {
69
+ "epoch": 0.01615182717544922,
70
+ "grad_norm": 2.2778408527374268,
71
+ "learning_rate": 1.5949820788530468e-05,
72
+ "loss": 1.382105255126953,
73
+ "step": 90
74
+ },
75
+ {
76
+ "epoch": 0.017946474639388026,
77
+ "grad_norm": 6.462679386138916,
78
+ "learning_rate": 1.774193548387097e-05,
79
+ "loss": 1.3651474952697753,
80
+ "step": 100
81
+ },
82
+ {
83
+ "epoch": 0.019741122103326826,
84
+ "grad_norm": 1.3327974081039429,
85
+ "learning_rate": 1.9534050179211473e-05,
86
+ "loss": 1.2019264221191406,
87
+ "step": 110
88
+ },
89
+ {
90
+ "epoch": 0.02153576956726563,
91
+ "grad_norm": 1.625113606452942,
92
+ "learning_rate": 2.132616487455197e-05,
93
+ "loss": 1.1140929222106934,
94
+ "step": 120
95
+ },
96
+ {
97
+ "epoch": 0.023330417031204434,
98
+ "grad_norm": 2.114356279373169,
99
+ "learning_rate": 2.3118279569892472e-05,
100
+ "loss": 1.194014072418213,
101
+ "step": 130
102
+ },
103
+ {
104
+ "epoch": 0.025125064495143234,
105
+ "grad_norm": 2.738748073577881,
106
+ "learning_rate": 2.4910394265232977e-05,
107
+ "loss": 1.310903263092041,
108
+ "step": 140
109
+ },
110
+ {
111
+ "epoch": 0.02691971195908204,
112
+ "grad_norm": 7.28310489654541,
113
+ "learning_rate": 2.670250896057348e-05,
114
+ "loss": 1.341808032989502,
115
+ "step": 150
116
+ },
117
+ {
118
+ "epoch": 0.02871435942302084,
119
+ "grad_norm": 1.4109766483306885,
120
+ "learning_rate": 2.8494623655913982e-05,
121
+ "loss": 1.0624680519104004,
122
+ "step": 160
123
+ },
124
+ {
125
+ "epoch": 0.030509006886959643,
126
+ "grad_norm": 1.8665562868118286,
127
+ "learning_rate": 3.0286738351254483e-05,
128
+ "loss": 1.1256374359130858,
129
+ "step": 170
130
+ },
131
+ {
132
+ "epoch": 0.03230365435089844,
133
+ "grad_norm": 2.103581190109253,
134
+ "learning_rate": 3.207885304659498e-05,
135
+ "loss": 1.1144988059997558,
136
+ "step": 180
137
+ },
138
+ {
139
+ "epoch": 0.03409830181483725,
140
+ "grad_norm": 2.7976624965667725,
141
+ "learning_rate": 3.387096774193548e-05,
142
+ "loss": 1.2549245834350586,
143
+ "step": 190
144
+ },
145
+ {
146
+ "epoch": 0.03589294927877605,
147
+ "grad_norm": 6.750956058502197,
148
+ "learning_rate": 3.566308243727599e-05,
149
+ "loss": 1.2815749168395996,
150
+ "step": 200
151
+ },
152
+ {
153
+ "epoch": 0.03768759674271485,
154
+ "grad_norm": 1.6264805793762207,
155
+ "learning_rate": 3.7455197132616484e-05,
156
+ "loss": 1.0297048568725586,
157
+ "step": 210
158
+ },
159
+ {
160
+ "epoch": 0.03948224420665365,
161
+ "grad_norm": 1.6889760494232178,
162
+ "learning_rate": 3.924731182795699e-05,
163
+ "loss": 1.0584090232849122,
164
+ "step": 220
165
+ },
166
+ {
167
+ "epoch": 0.04127689167059246,
168
+ "grad_norm": 2.0039634704589844,
169
+ "learning_rate": 4.1039426523297493e-05,
170
+ "loss": 1.085252571105957,
171
+ "step": 230
172
+ },
173
+ {
174
+ "epoch": 0.04307153913453126,
175
+ "grad_norm": 2.6522042751312256,
176
+ "learning_rate": 4.283154121863799e-05,
177
+ "loss": 1.2651546478271485,
178
+ "step": 240
179
+ },
180
+ {
181
+ "epoch": 0.04486618659847006,
182
+ "grad_norm": 8.020312309265137,
183
+ "learning_rate": 4.4623655913978496e-05,
184
+ "loss": 1.324000644683838,
185
+ "step": 250
186
+ },
187
+ {
188
+ "epoch": 0.04666083406240887,
189
+ "grad_norm": 1.7270288467407227,
190
+ "learning_rate": 4.6415770609319e-05,
191
+ "loss": 1.0575682640075683,
192
+ "step": 260
193
+ },
194
+ {
195
+ "epoch": 0.04845548152634767,
196
+ "grad_norm": 1.5692397356033325,
197
+ "learning_rate": 4.82078853046595e-05,
198
+ "loss": 1.0243738174438477,
199
+ "step": 270
200
+ },
201
+ {
202
+ "epoch": 0.05025012899028647,
203
+ "grad_norm": 2.029719591140747,
204
+ "learning_rate": 5e-05,
205
+ "loss": 1.0656013488769531,
206
+ "step": 280
207
+ },
208
+ {
209
+ "epoch": 0.052044776454225276,
210
+ "grad_norm": 2.4044599533081055,
211
+ "learning_rate": 5.17921146953405e-05,
212
+ "loss": 1.1308915138244628,
213
+ "step": 290
214
+ },
215
+ {
216
+ "epoch": 0.05383942391816408,
217
+ "grad_norm": 7.324768543243408,
218
+ "learning_rate": 5.3584229390681005e-05,
219
+ "loss": 1.2482977867126466,
220
+ "step": 300
221
+ },
222
+ {
223
+ "epoch": 0.05563407138210288,
224
+ "grad_norm": 1.4643646478652954,
225
+ "learning_rate": 5.53763440860215e-05,
226
+ "loss": 0.984095287322998,
227
+ "step": 310
228
+ },
229
+ {
230
+ "epoch": 0.05742871884604168,
231
+ "grad_norm": 1.6681671142578125,
232
+ "learning_rate": 5.7168458781362014e-05,
233
+ "loss": 1.011151123046875,
234
+ "step": 320
235
+ },
236
+ {
237
+ "epoch": 0.059223366309980485,
238
+ "grad_norm": 1.8012951612472534,
239
+ "learning_rate": 5.8960573476702505e-05,
240
+ "loss": 1.0619875907897949,
241
+ "step": 330
242
+ },
243
+ {
244
+ "epoch": 0.061018013773919286,
245
+ "grad_norm": 2.7174854278564453,
246
+ "learning_rate": 6.0752688172043016e-05,
247
+ "loss": 1.1800737380981445,
248
+ "step": 340
249
+ },
250
+ {
251
+ "epoch": 0.06281266123785809,
252
+ "grad_norm": 7.283389568328857,
253
+ "learning_rate": 6.254480286738351e-05,
254
+ "loss": 1.2749577522277833,
255
+ "step": 350
256
+ },
257
+ {
258
+ "epoch": 0.06460730870179689,
259
+ "grad_norm": 1.4635604619979858,
260
+ "learning_rate": 6.433691756272402e-05,
261
+ "loss": 1.0051876068115235,
262
+ "step": 360
263
+ },
264
+ {
265
+ "epoch": 0.06640195616573569,
266
+ "grad_norm": 1.4372541904449463,
267
+ "learning_rate": 6.612903225806452e-05,
268
+ "loss": 1.0501286506652832,
269
+ "step": 370
270
+ },
271
+ {
272
+ "epoch": 0.0681966036296745,
273
+ "grad_norm": 1.5717954635620117,
274
+ "learning_rate": 6.792114695340503e-05,
275
+ "loss": 1.0269421577453612,
276
+ "step": 380
277
+ },
278
+ {
279
+ "epoch": 0.0699912510936133,
280
+ "grad_norm": 2.4193785190582275,
281
+ "learning_rate": 6.971326164874552e-05,
282
+ "loss": 1.018425464630127,
283
+ "step": 390
284
+ },
285
+ {
286
+ "epoch": 0.0717858985575521,
287
+ "grad_norm": 5.167453289031982,
288
+ "learning_rate": 7.150537634408602e-05,
289
+ "loss": 1.2043580055236816,
290
+ "step": 400
291
+ },
292
+ {
293
+ "epoch": 0.0735805460214909,
294
+ "grad_norm": 1.1322799921035767,
295
+ "learning_rate": 7.329749103942653e-05,
296
+ "loss": 0.954345417022705,
297
+ "step": 410
298
+ },
299
+ {
300
+ "epoch": 0.0753751934854297,
301
+ "grad_norm": 1.3257076740264893,
302
+ "learning_rate": 7.508960573476703e-05,
303
+ "loss": 0.9895898818969726,
304
+ "step": 420
305
+ },
306
+ {
307
+ "epoch": 0.0771698409493685,
308
+ "grad_norm": 1.317622184753418,
309
+ "learning_rate": 7.688172043010752e-05,
310
+ "loss": 0.9647544860839844,
311
+ "step": 430
312
+ },
313
+ {
314
+ "epoch": 0.0789644884133073,
315
+ "grad_norm": 1.8667112588882446,
316
+ "learning_rate": 7.867383512544804e-05,
317
+ "loss": 1.139955997467041,
318
+ "step": 440
319
+ },
320
+ {
321
+ "epoch": 0.08075913587724612,
322
+ "grad_norm": 5.289592266082764,
323
+ "learning_rate": 8.046594982078853e-05,
324
+ "loss": 1.1954397201538085,
325
+ "step": 450
326
+ },
327
+ {
328
+ "epoch": 0.08255378334118492,
329
+ "grad_norm": 1.0380669832229614,
330
+ "learning_rate": 8.225806451612904e-05,
331
+ "loss": 0.9445084571838379,
332
+ "step": 460
333
+ },
334
+ {
335
+ "epoch": 0.08434843080512372,
336
+ "grad_norm": 1.3310823440551758,
337
+ "learning_rate": 8.405017921146954e-05,
338
+ "loss": 0.9562747955322266,
339
+ "step": 470
340
+ },
341
+ {
342
+ "epoch": 0.08614307826906252,
343
+ "grad_norm": 1.417316198348999,
344
+ "learning_rate": 8.584229390681004e-05,
345
+ "loss": 0.9948525428771973,
346
+ "step": 480
347
+ },
348
+ {
349
+ "epoch": 0.08793772573300132,
350
+ "grad_norm": 1.8197181224822998,
351
+ "learning_rate": 8.763440860215054e-05,
352
+ "loss": 1.0975610733032226,
353
+ "step": 490
354
+ },
355
+ {
356
+ "epoch": 0.08973237319694012,
357
+ "grad_norm": 4.941095352172852,
358
+ "learning_rate": 8.942652329749104e-05,
359
+ "loss": 1.2107452392578124,
360
+ "step": 500
361
+ },
362
+ {
363
+ "epoch": 0.09152702066087894,
364
+ "grad_norm": 0.9819939732551575,
365
+ "learning_rate": 9.121863799283154e-05,
366
+ "loss": 0.9593785285949707,
367
+ "step": 510
368
+ },
369
+ {
370
+ "epoch": 0.09332166812481774,
371
+ "grad_norm": 1.1892472505569458,
372
+ "learning_rate": 9.301075268817204e-05,
373
+ "loss": 0.9651047706604003,
374
+ "step": 520
375
+ },
376
+ {
377
+ "epoch": 0.09511631558875654,
378
+ "grad_norm": 1.4324865341186523,
379
+ "learning_rate": 9.480286738351255e-05,
380
+ "loss": 0.9062895774841309,
381
+ "step": 530
382
+ },
383
+ {
384
+ "epoch": 0.09691096305269534,
385
+ "grad_norm": 1.6410809755325317,
386
+ "learning_rate": 9.659498207885304e-05,
387
+ "loss": 1.0246500968933105,
388
+ "step": 540
389
+ },
390
+ {
391
+ "epoch": 0.09870561051663414,
392
+ "grad_norm": 5.0956501960754395,
393
+ "learning_rate": 9.838709677419355e-05,
394
+ "loss": 1.1241219520568848,
395
+ "step": 550
396
+ },
397
+ {
398
+ "epoch": 0.10050025798057294,
399
+ "grad_norm": 0.951851487159729,
400
+ "learning_rate": 9.999055534567435e-05,
401
+ "loss": 0.9464516639709473,
402
+ "step": 560
403
+ },
404
+ {
405
+ "epoch": 0.10229490544451174,
406
+ "grad_norm": 1.1737598180770874,
407
+ "learning_rate": 9.989610880241783e-05,
408
+ "loss": 0.907046890258789,
409
+ "step": 570
410
+ },
411
+ {
412
+ "epoch": 0.10408955290845055,
413
+ "grad_norm": 1.3924814462661743,
414
+ "learning_rate": 9.980166225916132e-05,
415
+ "loss": 0.9977164268493652,
416
+ "step": 580
417
+ },
418
+ {
419
+ "epoch": 0.10588420037238935,
420
+ "grad_norm": 1.6580984592437744,
421
+ "learning_rate": 9.97072157159048e-05,
422
+ "loss": 1.0950872421264648,
423
+ "step": 590
424
+ },
425
+ {
426
+ "epoch": 0.10767884783632815,
427
+ "grad_norm": 4.493084907531738,
428
+ "learning_rate": 9.961276917264828e-05,
429
+ "loss": 1.211144256591797,
430
+ "step": 600
431
+ },
432
+ {
433
+ "epoch": 0.10947349530026695,
434
+ "grad_norm": 0.8613011240959167,
435
+ "learning_rate": 9.951832262939177e-05,
436
+ "loss": 0.9501323699951172,
437
+ "step": 610
438
+ },
439
+ {
440
+ "epoch": 0.11126814276420575,
441
+ "grad_norm": 1.1255444288253784,
442
+ "learning_rate": 9.942387608613525e-05,
443
+ "loss": 0.903736686706543,
444
+ "step": 620
445
+ },
446
+ {
447
+ "epoch": 0.11306279022814456,
448
+ "grad_norm": 1.4082947969436646,
449
+ "learning_rate": 9.932942954287873e-05,
450
+ "loss": 1.0084752082824706,
451
+ "step": 630
452
+ },
453
+ {
454
+ "epoch": 0.11485743769208336,
455
+ "grad_norm": 1.652286171913147,
456
+ "learning_rate": 9.923498299962222e-05,
457
+ "loss": 0.930430793762207,
458
+ "step": 640
459
+ },
460
+ {
461
+ "epoch": 0.11665208515602217,
462
+ "grad_norm": 5.117873668670654,
463
+ "learning_rate": 9.91405364563657e-05,
464
+ "loss": 1.091403865814209,
465
+ "step": 650
466
+ },
467
+ {
468
+ "epoch": 0.11844673261996097,
469
+ "grad_norm": 1.0188069343566895,
470
+ "learning_rate": 9.904608991310918e-05,
471
+ "loss": 0.883120059967041,
472
+ "step": 660
473
+ },
474
+ {
475
+ "epoch": 0.12024138008389977,
476
+ "grad_norm": 1.1442457437515259,
477
+ "learning_rate": 9.895164336985267e-05,
478
+ "loss": 0.9172245979309082,
479
+ "step": 670
480
+ },
481
+ {
482
+ "epoch": 0.12203602754783857,
483
+ "grad_norm": 1.3536146879196167,
484
+ "learning_rate": 9.885719682659615e-05,
485
+ "loss": 0.9197479248046875,
486
+ "step": 680
487
+ },
488
+ {
489
+ "epoch": 0.12383067501177737,
490
+ "grad_norm": 1.8869231939315796,
491
+ "learning_rate": 9.876275028333963e-05,
492
+ "loss": 1.0764457702636718,
493
+ "step": 690
494
+ },
495
+ {
496
+ "epoch": 0.12562532247571617,
497
+ "grad_norm": 3.6568493843078613,
498
+ "learning_rate": 9.866830374008312e-05,
499
+ "loss": 1.0616437911987304,
500
+ "step": 700
501
+ },
502
+ {
503
+ "epoch": 0.127419969939655,
504
+ "grad_norm": 0.8427063822746277,
505
+ "learning_rate": 9.85738571968266e-05,
506
+ "loss": 0.8856773376464844,
507
+ "step": 710
508
+ },
509
+ {
510
+ "epoch": 0.12921461740359377,
511
+ "grad_norm": 1.0265027284622192,
512
+ "learning_rate": 9.847941065357009e-05,
513
+ "loss": 0.907376766204834,
514
+ "step": 720
515
+ },
516
+ {
517
+ "epoch": 0.1310092648675326,
518
+ "grad_norm": 1.3209590911865234,
519
+ "learning_rate": 9.838496411031357e-05,
520
+ "loss": 1.0000056266784667,
521
+ "step": 730
522
+ },
523
+ {
524
+ "epoch": 0.13280391233147137,
525
+ "grad_norm": 1.7275080680847168,
526
+ "learning_rate": 9.829051756705705e-05,
527
+ "loss": 1.1050853729248047,
528
+ "step": 740
529
+ },
530
+ {
531
+ "epoch": 0.1345985597954102,
532
+ "grad_norm": 4.58080530166626,
533
+ "learning_rate": 9.819607102380054e-05,
534
+ "loss": 1.1597484588623046,
535
+ "step": 750
536
+ },
537
+ {
538
+ "epoch": 0.136393207259349,
539
+ "grad_norm": 0.8872320055961609,
540
+ "learning_rate": 9.810162448054402e-05,
541
+ "loss": 0.917822265625,
542
+ "step": 760
543
+ },
544
+ {
545
+ "epoch": 0.1381878547232878,
546
+ "grad_norm": 0.9287213087081909,
547
+ "learning_rate": 9.80071779372875e-05,
548
+ "loss": 0.8946282386779785,
549
+ "step": 770
550
+ },
551
+ {
552
+ "epoch": 0.1399825021872266,
553
+ "grad_norm": 1.183867335319519,
554
+ "learning_rate": 9.791273139403099e-05,
555
+ "loss": 0.9173077583312989,
556
+ "step": 780
557
+ },
558
+ {
559
+ "epoch": 0.1417771496511654,
560
+ "grad_norm": 1.6546437740325928,
561
+ "learning_rate": 9.781828485077447e-05,
562
+ "loss": 0.962445068359375,
563
+ "step": 790
564
+ },
565
+ {
566
+ "epoch": 0.1435717971151042,
567
+ "grad_norm": 4.307088375091553,
568
+ "learning_rate": 9.772383830751794e-05,
569
+ "loss": 1.1235834121704102,
570
+ "step": 800
571
+ },
572
+ {
573
+ "epoch": 0.145366444579043,
574
+ "grad_norm": 0.8552832007408142,
575
+ "learning_rate": 9.762939176426144e-05,
576
+ "loss": 0.8700121879577637,
577
+ "step": 810
578
+ },
579
+ {
580
+ "epoch": 0.1471610920429818,
581
+ "grad_norm": 0.9782463908195496,
582
+ "learning_rate": 9.753494522100492e-05,
583
+ "loss": 0.8827217102050782,
584
+ "step": 820
585
+ },
586
+ {
587
+ "epoch": 0.14895573950692062,
588
+ "grad_norm": 1.2572635412216187,
589
+ "learning_rate": 9.74404986777484e-05,
590
+ "loss": 0.9147042274475098,
591
+ "step": 830
592
+ },
593
+ {
594
+ "epoch": 0.1507503869708594,
595
+ "grad_norm": 1.6980787515640259,
596
+ "learning_rate": 9.734605213449189e-05,
597
+ "loss": 1.0121471405029296,
598
+ "step": 840
599
+ },
600
+ {
601
+ "epoch": 0.15254503443479822,
602
+ "grad_norm": 4.741994380950928,
603
+ "learning_rate": 9.725160559123536e-05,
604
+ "loss": 1.070129680633545,
605
+ "step": 850
606
+ },
607
+ {
608
+ "epoch": 0.154339681898737,
609
+ "grad_norm": 0.7885052561759949,
610
+ "learning_rate": 9.715715904797885e-05,
611
+ "loss": 0.9009537696838379,
612
+ "step": 860
613
+ },
614
+ {
615
+ "epoch": 0.15613432936267582,
616
+ "grad_norm": 0.9653502106666565,
617
+ "learning_rate": 9.706271250472234e-05,
618
+ "loss": 0.8701554298400879,
619
+ "step": 870
620
+ },
621
+ {
622
+ "epoch": 0.1579289768266146,
623
+ "grad_norm": 1.2959074974060059,
624
+ "learning_rate": 9.69682659614658e-05,
625
+ "loss": 0.9180045127868652,
626
+ "step": 880
627
+ },
628
+ {
629
+ "epoch": 0.15972362429055342,
630
+ "grad_norm": 1.5041567087173462,
631
+ "learning_rate": 9.68738194182093e-05,
632
+ "loss": 0.9027626037597656,
633
+ "step": 890
634
+ },
635
+ {
636
+ "epoch": 0.16151827175449224,
637
+ "grad_norm": 5.474041938781738,
638
+ "learning_rate": 9.677937287495277e-05,
639
+ "loss": 1.0739646911621095,
640
+ "step": 900
641
+ },
642
+ {
643
+ "epoch": 0.16331291921843102,
644
+ "grad_norm": 0.933229923248291,
645
+ "learning_rate": 9.668492633169627e-05,
646
+ "loss": 0.881598949432373,
647
+ "step": 910
648
+ },
649
+ {
650
+ "epoch": 0.16510756668236984,
651
+ "grad_norm": 0.997329592704773,
652
+ "learning_rate": 9.659047978843975e-05,
653
+ "loss": 0.8712461471557618,
654
+ "step": 920
655
+ },
656
+ {
657
+ "epoch": 0.16690221414630863,
658
+ "grad_norm": 1.319835901260376,
659
+ "learning_rate": 9.649603324518322e-05,
660
+ "loss": 0.8690865516662598,
661
+ "step": 930
662
+ },
663
+ {
664
+ "epoch": 0.16869686161024744,
665
+ "grad_norm": 1.8019057512283325,
666
+ "learning_rate": 9.640158670192672e-05,
667
+ "loss": 1.1501221656799316,
668
+ "step": 940
669
+ },
670
+ {
671
+ "epoch": 0.17049150907418623,
672
+ "grad_norm": 3.8484811782836914,
673
+ "learning_rate": 9.630714015867019e-05,
674
+ "loss": 1.1916112899780273,
675
+ "step": 950
676
+ },
677
+ {
678
+ "epoch": 0.17228615653812504,
679
+ "grad_norm": 0.8143914937973022,
680
+ "learning_rate": 9.621269361541369e-05,
681
+ "loss": 0.91885404586792,
682
+ "step": 960
683
+ },
684
+ {
685
+ "epoch": 0.17408080400206385,
686
+ "grad_norm": 1.0034821033477783,
687
+ "learning_rate": 9.611824707215716e-05,
688
+ "loss": 0.8618650436401367,
689
+ "step": 970
690
+ },
691
+ {
692
+ "epoch": 0.17587545146600264,
693
+ "grad_norm": 1.3009910583496094,
694
+ "learning_rate": 9.602380052890064e-05,
695
+ "loss": 0.8944579124450683,
696
+ "step": 980
697
+ },
698
+ {
699
+ "epoch": 0.17767009892994146,
700
+ "grad_norm": 1.4715969562530518,
701
+ "learning_rate": 9.592935398564414e-05,
702
+ "loss": 1.0547027587890625,
703
+ "step": 990
704
+ },
705
+ {
706
+ "epoch": 0.17946474639388024,
707
+ "grad_norm": 4.630854606628418,
708
+ "learning_rate": 9.583490744238761e-05,
709
+ "loss": 1.1706391334533692,
710
+ "step": 1000
711
+ }
712
+ ],
713
+ "logging_steps": 10,
714
+ "max_steps": 11146,
715
+ "num_input_tokens_seen": 0,
716
+ "num_train_epochs": 2,
717
+ "save_steps": 1000,
718
+ "stateful_callbacks": {
719
+ "TrainerControl": {
720
+ "args": {
721
+ "should_epoch_stop": false,
722
+ "should_evaluate": false,
723
+ "should_log": false,
724
+ "should_save": true,
725
+ "should_training_stop": false
726
+ },
727
+ "attributes": {}
728
+ }
729
+ },
730
+ "total_flos": 1.014956564093929e+17,
731
+ "train_batch_size": 2,
732
+ "trial_name": null,
733
+ "trial_params": null
734
+ }
qwen3.5-2b-50pct-masked-weighted/checkpoint-1000/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:62edea7eb1e0274b5104b0e4cbcf850d6b18e77a951e843e93b693eaf58caa36
3
+ size 5265
qwen3.5-2b-50pct-masked-weighted/config.yaml ADDED
@@ -0,0 +1,61 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ model_id: Qwen/Qwen3.5-2B
2
+ dataset_id: PedramR/ASR_Post-processing-dataset
3
+ data_config: configs/data/default.yaml
4
+ train_split: train
5
+ eval_split: null
6
+ eval_fraction: 0.1
7
+ train_fraction: 0.5
8
+ input_column: text_whisper
9
+ target_column: text
10
+ system_prompt: null
11
+ include_punctuation: false
12
+ output_dir: outputs/qwen-mask-weighted/qwen3.5-2b-50pct-masked-weighted
13
+ max_length: 2048
14
+ on_long_example: drop
15
+ mask_low_signal_corrections: true
16
+ mask_min_similarity: 0.75
17
+ mask_max_common_freq: 1
18
+ recoverable_correction_weight: 3.0
19
+ num_train_epochs: 2.0
20
+ per_device_train_batch_size: 2
21
+ per_device_eval_batch_size: 2
22
+ gradient_accumulation_steps: 8
23
+ learning_rate: 0.0001
24
+ warmup_ratio: 0.05
25
+ logging_steps: 10
26
+ eval_steps: 1000
27
+ save_steps: 1000
28
+ save_total_limit: 3
29
+ seed: 42
30
+ lora_r: 32
31
+ lora_alpha: 64
32
+ lora_dropout: 0.05
33
+ load_in_4bit: false
34
+ use_unsloth: false
35
+ push_to_hub: true
36
+ hub_repo_id: PedramR/ASR_Post-processing
37
+ hub_repo_folder: qwen3.5-2b-50pct-masked-weighted
38
+ hub_private: false
39
+ tensorboard_logging_dir: outputs/qwen-mask-weighted/qwen3.5-2b-50pct-masked-weighted/tb
40
+ resume_from_checkpoint: false
41
+ lora_init_mode: fresh
42
+ init_lora_from: null
43
+ init_lora_from_subfolder: null
44
+ test_dataset_id: ErfanRou/callcc-test-1k
45
+ test_split: test
46
+ test_input_column: text_whisper
47
+ test_target_column: text
48
+ test_max_new_tokens: 1024
49
+ test_batch_size: 16
50
+ test_max_examples: null
51
+ test_checkpoint_max_examples: null
52
+ test_checkpoint_eval_main: false
53
+ test_hallucination_overlap_floor: 50.0
54
+ test_repetition_penalty: 1.0
55
+ test_no_repeat_ngram_size: 0
56
+ test_baseline: true
57
+ test_entity_dataset_id: PedramR/ASR_Post-processing-eval
58
+ test_typo_dataset_id: PedramR/ASR_Post-processing-eval
59
+ test_entity_row_type: entity
60
+ test_typo_row_type: typo
61
+ test_fix_weight: 0.5
qwen3.5-2b-50pct-masked-weighted/resolved_config.yaml ADDED
@@ -0,0 +1,54 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ model_id: Qwen/Qwen3.5-2B
2
+ data_config: configs/data/default.yaml
3
+ train_split: train
4
+ eval_split: null
5
+ eval_fraction: 0.1
6
+ train_fraction: 0.5
7
+ input_column: text_whisper
8
+ target_column: text
9
+ system_prompt: null
10
+ output_dir: outputs/qwen-mask-weighted/qwen3.5-2b-50pct-masked-weighted
11
+ max_length: 2048
12
+ on_long_example: drop
13
+ mask_low_signal_corrections: true
14
+ mask_min_similarity: 0.75
15
+ mask_max_common_freq: 1
16
+ recoverable_correction_weight: 3.0
17
+ training:
18
+ num_train_epochs: 2.0
19
+ per_device_train_batch_size: 2
20
+ per_device_eval_batch_size: 2
21
+ gradient_accumulation_steps: 8
22
+ learning_rate: 0.0001
23
+ warmup_ratio: 0.05
24
+ logging_steps: 10
25
+ eval_steps: 1000
26
+ save_steps: 1000
27
+ save_total_limit: 3
28
+ seed: 42
29
+ lora:
30
+ r: 32
31
+ alpha: 64
32
+ dropout: 0.05
33
+ quantization:
34
+ load_in_4bit: false
35
+ hub:
36
+ push_to_hub: true
37
+ repo_id: PedramR/ASR_Post-processing
38
+ folder: qwen3.5-2b-50pct-masked-weighted
39
+ private: false
40
+ tensorboard:
41
+ logging_dir: outputs/qwen-mask-weighted/qwen3.5-2b-50pct-masked-weighted/tb
42
+ test:
43
+ dataset_id: ErfanRou/callcc-test-1k
44
+ split: test
45
+ input_column: text_whisper
46
+ target_column: text
47
+ max_new_tokens: 1024
48
+ batch_size: 16
49
+ max_examples: null
50
+ checkpoint_max_examples: null
51
+ fix_weight: 0.5
52
+ checkpoint_eval_main: false
53
+ resume_from_checkpoint: false
54
+ use_unsloth: false
qwen3.5-2b-50pct-masked-weighted/tb/events.out.tfevents.1790426135.b0aacb224392.4711.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:85f106682ee3b33940d0fc6758d0dcb3a97c07d7e3b123ff12be14b8095c3acf
3
+ size 30303
qwen3.5-2b-50pct-masked-weighted/test_eval_baseline/metrics.json ADDED
@@ -0,0 +1,15 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "wer": 0.2669905854225542,
3
+ "wer_zwnj_normalized": 0.25631724206316414,
4
+ "exact_match": 0.003033367037411527,
5
+ "fix_rate": 0.028245578305891937,
6
+ "preservation_rate": 0.9054879169577001,
7
+ "targeted_score": 0.46686674763179603,
8
+ "fix_rate_lenient": 0.029970652369944093,
9
+ "preservation_rate_lenient": 0.9107829846480783,
10
+ "targeted_score_lenient": 0.4703768185090112,
11
+ "fix_rate_recoverable": 0.030331864316239316,
12
+ "unrecoverable_targets": 1089,
13
+ "hallucination_rate": 0.007077856420626896,
14
+ "n_examples": 1978
15
+ }
qwen3.5-2b-50pct-masked-weighted/test_eval_baseline/predictions.jsonl ADDED
The diff for this file is too large to render. See raw diff
 
qwen3.5-2b-50pct-masked-weighted/test_eval_entity/metrics.json ADDED
@@ -0,0 +1,15 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "wer": 0.7114996637525218,
3
+ "wer_zwnj_normalized": 0.7087723230964657,
4
+ "exact_match": 0.0019801980198019802,
5
+ "fix_rate": 0.09490559539092588,
6
+ "preservation_rate": 0.34338748592568763,
7
+ "targeted_score": 0.21914654065830674,
8
+ "fix_rate_lenient": 0.09133004926108375,
9
+ "preservation_rate_lenient": 0.3451304158166866,
10
+ "targeted_score_lenient": 0.21823023253888515,
11
+ "fix_rate_recoverable": 0.09265880308751295,
12
+ "unrecoverable_targets": 523,
13
+ "hallucination_rate": 0.00891089108910891,
14
+ "n_examples": 1010
15
+ }
qwen3.5-2b-50pct-masked-weighted/test_eval_entity/predictions.jsonl ADDED
The diff for this file is too large to render. See raw diff
 
qwen3.5-2b-50pct-masked-weighted/test_eval_entity_baseline/metrics.json ADDED
@@ -0,0 +1,15 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "wer": 0.25836615749191405,
3
+ "wer_zwnj_normalized": 0.2470725120355248,
4
+ "exact_match": 0.0009900990099009901,
5
+ "fix_rate": 0.028618843347841063,
6
+ "preservation_rate": 0.9102010616052758,
7
+ "targeted_score": 0.4694099524765584,
8
+ "fix_rate_lenient": 0.030607553366174056,
9
+ "preservation_rate_lenient": 0.9158680061070269,
10
+ "targeted_score_lenient": 0.4732377797366005,
11
+ "fix_rate_recoverable": 0.03097537340862766,
12
+ "unrecoverable_targets": 523,
13
+ "hallucination_rate": 0.005940594059405941,
14
+ "n_examples": 1010
15
+ }
qwen3.5-2b-50pct-masked-weighted/test_eval_entity_baseline/predictions.jsonl ADDED
The diff for this file is too large to render. See raw diff
 
qwen3.5-2b-50pct-masked-weighted/test_eval_typo/metrics.json ADDED
@@ -0,0 +1,15 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "wer": 0.5876286027928825,
3
+ "wer_zwnj_normalized": 0.5829312501595568,
4
+ "exact_match": 0.00558659217877095,
5
+ "fix_rate": 0.15878247382706476,
6
+ "preservation_rate": 0.459294975450563,
7
+ "targeted_score": 0.3090387246388139,
8
+ "fix_rate_lenient": 0.1542123783895674,
9
+ "preservation_rate_lenient": 0.46072255266781914,
10
+ "targeted_score_lenient": 0.30746746552869325,
11
+ "fix_rate_recoverable": 0.15657839428331233,
12
+ "unrecoverable_targets": 73,
13
+ "hallucination_rate": 0.00558659217877095,
14
+ "n_examples": 358
15
+ }
qwen3.5-2b-50pct-masked-weighted/test_eval_typo/predictions.jsonl ADDED
The diff for this file is too large to render. See raw diff
 
qwen3.5-2b-50pct-masked-weighted/test_eval_typo_baseline/metrics.json ADDED
@@ -0,0 +1,15 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "wer": 0.19879502693319037,
3
+ "wer_zwnj_normalized": 0.18631130172831942,
4
+ "exact_match": 0.0,
5
+ "fix_rate": 0.04226444358278403,
6
+ "preservation_rate": 0.9286449298797519,
7
+ "targeted_score": 0.48545468673126796,
8
+ "fix_rate_lenient": 0.04533222935210102,
9
+ "preservation_rate_lenient": 0.9356558938146444,
10
+ "targeted_score_lenient": 0.4904940615833727,
11
+ "fix_rate_recoverable": 0.046027742749054225,
12
+ "unrecoverable_targets": 73,
13
+ "hallucination_rate": 0.0111731843575419,
14
+ "n_examples": 358
15
+ }
qwen3.5-2b-50pct-masked-weighted/test_eval_typo_baseline/predictions.jsonl ADDED
The diff for this file is too large to render. See raw diff