backup e16/S2-s-brier (Pod B)

e16/S2-s-brier/ckpt-16000/adapter/README.md

@@ -0,0 +1,206 @@
+---
+base_model: /workspace/jev_kt/models/hf/JEV-9B
+library_name: peft
+tags:
+- base_model:adapter:/workspace/jev_kt/models/hf/JEV-9B
+- 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 0.21.2

e16/S2-s-brier/ckpt-16000/adapter/adapter_config.json

@@ -0,0 +1,57 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "Qwen3_5ForCausalLM",
+    "parent_library": "transformers.models.qwen3_5.modeling_qwen3_5"
+  },
+  "base_model_name_or_path": "/workspace/jev_kt/models/hf/JEV-9B",
+  "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": 32,
+  "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": 16,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "in_proj_z",
+    "in_proj_qkv",
+    "gate_proj",
+    "out_proj",
+    "q_proj",
+    "up_proj",
+    "v_proj",
+    "o_proj",
+    "k_proj",
+    "down_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_bdlora": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false,
+  "velora_config": null
+}

e16/S2-s-brier/ckpt-16000/adapter/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6c20e47fe06a704370102fb5eee6b15385f3f6f7484c31696129b480104729de
+size 160486456

e16/S2-s-brier/ckpt-16000/adapter_vllm/adapter_config.json

@@ -0,0 +1,58 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "Qwen3_5ForCausalLM",
+    "parent_library": "transformers.models.qwen3_5.modeling_qwen3_5"
+  },
+  "base_model_name_or_path": null,
+  "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": 64.0,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "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": [
+    "down_proj",
+    "gate_proj",
+    "in_proj_qkv",
+    "in_proj_z",
+    "k_proj",
+    "lm_head",
+    "o_proj",
+    "out_proj",
+    "q_proj",
+    "up_proj",
+    "v_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_bdlora": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false,
+  "velora_config": null
+}

e16/S2-s-brier/ckpt-16000/adapter_vllm/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:11887838d97546bb43da5700dc973b4bf82e3cd010f5094c3e3d2867cf4d6e62
+size 176641832

e16/S2-s-brier/ckpt-16000/adapter_vllm/decision_head.json

@@ -0,0 +1,99 @@
+{
+ "bias": [
+  0.000260660657659173,
+  -0.000260907196206972,
+  -0.0025878134183585644,
+  -0.0012615364976227283,
+  -0.00025007399381138384,
+  0.0019558551721274853,
+  0.0013559526996687055,
+  -0.0023348911199718714,
+  -0.0004082925443071872,
+  -0.0002078723773593083,
+  -1.2957353646925185e-05,
+  0.0001676026004133746,
+  0.0005569425993598998,
+  -0.001412136247381568,
+  -0.0028663526754826307,
+  -0.005061559844762087,
+  -0.00593328895047307,
+  -0.004319756291806698,
+  -0.004922541789710522,
+  -0.0049067651852965355,
+  -0.003943523392081261,
+  -0.003319827374070883,
+  -0.005581587553024292,
+  -0.0034480984322726727
+ ],
+ "verbalizer_ids": [
+  3721,
+  1802,
+  15,
+  16,
+  17,
+  18,
+  19,
+  20,
+  32,
+  33,
+  34,
+  35,
+  36,
+  37,
+  38,
+  39,
+  40,
+  41,
+  42,
+  43,
+  44,
+  45,
+  46,
+  47
+ ],
+ "slots": {
+  "num_slots": 24,
+  "ranges": {
+   "noul": [
+    0,
+    2
+   ],
+   "score": [
+    2,
+    8
+   ],
+   "choice": [
+    8,
+    24
+   ]
+  },
+  "verbalizers": [
+   "false",
+   "true",
+   "0",
+   "1",
+   "2",
+   "3",
+   "4",
+   "5",
+   "A",
+   "B",
+   "C",
+   "D",
+   "E",
+   "F",
+   "G",
+   "H",
+   "I",
+   "J",
+   "K",
+   "L",
+   "M",
+   "N",
+   "O",
+   "P"
+  ],
+  "template_version": "bare-v1"
+ },
+ "note": "decision logits = lm_head-LoRA logprobs of verbalizer ids + bias, then per-kind temperature"
+}

e16/S2-s-brier/ckpt-16000/calibration.json

@@ -0,0 +1,17 @@
+{
+ "version": 1,
+ "per_kind": {
+  "noul": 1.1541319454861183,
+  "choice": 1.0,
+  "score": 1.0
+ },
+ "fit": {
+  "noul": {
+   "n": 2000,
+   "T": 1.1541319454861183,
+   "ce_before": 0.5325164198875427,
+   "ce_after": 0.5300415754318237
+  }
+ },
+ "source": "models/e16/S2-s-brier/ckpt-16000"
+}

e16/S2-s-brier/ckpt-16000/head.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ed97c9ae60a6a9d36739cd16d6fd2cb895d6ec746d059e7af384f9e8831b29a7
+size 393464

e16/S2-s-brier/ckpt-16000/judge_config.json

@@ -0,0 +1,160 @@
+{
+ "base_model_path": "/root/models/Qwen3.5-9B",
+ "hidden_size": 4096,
+ "slots": {
+  "num_slots": 24,
+  "ranges": {
+   "noul": [
+    0,
+    2
+   ],
+   "score": [
+    2,
+    8
+   ],
+   "choice": [
+    8,
+    24
+   ]
+  },
+  "verbalizers": [
+   "false",
+   "true",
+   "0",
+   "1",
+   "2",
+   "3",
+   "4",
+   "5",
+   "A",
+   "B",
+   "C",
+   "D",
+   "E",
+   "F",
+   "G",
+   "H",
+   "I",
+   "J",
+   "K",
+   "L",
+   "M",
+   "N",
+   "O",
+   "P"
+  ],
+  "template_version": "bare-v1"
+ },
+ "verbalizer_ids": [
+  3721,
+  1802,
+  15,
+  16,
+  17,
+  18,
+  19,
+  20,
+  32,
+  33,
+  34,
+  35,
+  36,
+  37,
+  38,
+  39,
+  40,
+  41,
+  42,
+  43,
+  44,
+  45,
+  46,
+  47
+ ],
+ "kinds": [
+  "noul",
+  "choice",
+  "score"
+ ],
+ "model_name": "jev-judge-qwen35-9b",
+ "model_version": "0.8.0",
+ "lm_head_included": true,
+ "weights_mode": "unmerged",
+ "adapter_subfolder": "adapter",
+ "lora": {
+  "r": 16,
+  "lora_alpha": 32,
+  "lora_dropout": 0.05,
+  "target_modules": [
+   "down_proj",
+   "gate_proj",
+   "in_proj_qkv",
+   "in_proj_z",
+   "k_proj",
+   "o_proj",
+   "out_proj",
+   "q_proj",
+   "up_proj",
+   "v_proj"
+  ]
+ },
+ "trained_stage": "e16",
+ "source_checkpoint": "models/e16/S2-s-brier/ckpt-16000",
+ "softcap": null,
+ "e16": {
+  "data": "data/e15/F1-pool",
+  "data_manifest": {
+   "arm": "F1-pool",
+   "name": "F1-pool",
+   "train_sha256": "dad16f94a5659768b254a6c12538c01f9c64cd43d574a3435bcc70c9c1bac130",
+   "n_train": 64000
+  },
+  "init": "released",
+  "objective": "s-brier",
+  "sigma": 0.5,
+  "K": 8,
+  "reward": "brier",
+  "clip": 0.2,
+  "passes": 1,
+  "lr": 0.0001,
+  "lr_head": 0.0002,
+  "rows_per_step": 32,
+  "train_seed": 0,
+  "replay": null,
+  "replay_frac": 0.0,
+  "replay_sha256": null,
+  "max_len": 3072,
+  "schedule": "warmup 0.03 + cosine to x0.05 over 500 steps",
+  "total_rows": 16000,
+  "lora": {
+   "r": 16,
+   "lora_alpha": 32,
+   "lora_dropout": 0.05
+  },
+  "trainer_sha256": "bc7c08e836bd99bb2665b02903cd41404ea029ecfa9b68a9ec1a7cda96958ba4",
+  "event": "ckpt",
+  "tag": "ckpt-16000",
+  "rows_seen": 16000,
+  "replay_seen": 0,
+  "step": 500,
+  "tokens": 10544036,
+  "tok_per_s": 803.3155856789108,
+  "rows_per_s": 1.218987621624183,
+  "val": {
+   "n": 2000,
+   "dropped_long": 0,
+   "noul_ce": 0.5325164198875427,
+   "noul_T": 1.1541319454861183,
+   "noul_auc": 0.7755146700071789,
+   "noul_acc": 0.737,
+   "noul_f1": 0.8050407709414381,
+   "noul_nll": 0.5325164518207312,
+   "noul_ece": 0.04705242341756822,
+   "noul_brier": 0.17783741743174705,
+   "noul_sharp": 0.2612763047218323,
+   "entropy": 0.4965769648551941
+  },
+  "dropped_long": 0,
+  "peak_gib": 18.236003398895264
+ }
+}

e16/S2-s-brier/ckpt-4000/adapter/README.md

@@ -0,0 +1,206 @@
+---
+base_model: /workspace/jev_kt/models/hf/JEV-9B
+library_name: peft
+tags:
+- base_model:adapter:/workspace/jev_kt/models/hf/JEV-9B
+- 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 0.21.2

e16/S2-s-brier/ckpt-4000/adapter/adapter_config.json

@@ -0,0 +1,57 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "Qwen3_5ForCausalLM",
+    "parent_library": "transformers.models.qwen3_5.modeling_qwen3_5"
+  },
+  "base_model_name_or_path": "/workspace/jev_kt/models/hf/JEV-9B",
+  "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": 32,
+  "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": 16,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "in_proj_z",
+    "in_proj_qkv",
+    "gate_proj",
+    "out_proj",
+    "q_proj",
+    "up_proj",
+    "v_proj",
+    "o_proj",
+    "k_proj",
+    "down_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_bdlora": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false,
+  "velora_config": null
+}

e16/S2-s-brier/ckpt-4000/adapter/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9accf0a7adf61548a235f03aa483f79e53bd5f2e5414f4c326ba57b6f1915336
+size 160486456

e16/S2-s-brier/ckpt-4000/adapter_vllm/adapter_config.json

@@ -0,0 +1,58 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "Qwen3_5ForCausalLM",
+    "parent_library": "transformers.models.qwen3_5.modeling_qwen3_5"
+  },
+  "base_model_name_or_path": null,
+  "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": 64.0,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "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": [
+    "down_proj",
+    "gate_proj",
+    "in_proj_qkv",
+    "in_proj_z",
+    "k_proj",
+    "lm_head",
+    "o_proj",
+    "out_proj",
+    "q_proj",
+    "up_proj",
+    "v_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_bdlora": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false,
+  "velora_config": null
+}

e16/S2-s-brier/ckpt-4000/adapter_vllm/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:83a04cc8402ef48c4ca57d51b6eb0de2a380411967a8d1e824a05912287f4cfb
+size 176641832

e16/S2-s-brier/ckpt-4000/adapter_vllm/decision_head.json

@@ -0,0 +1,99 @@
+{
+ "bias": [
+  0.0003804515872616321,
+  -0.0003806980967056006,
+  -0.0025878134183585644,
+  -0.0012615364976227283,
+  -0.00025007399381138384,
+  0.0019558551721274853,
+  0.0013559526996687055,
+  -0.0023348911199718714,
+  -0.0004082925443071872,
+  -0.0002078723773593083,
+  -1.2957353646925185e-05,
+  0.0001676026004133746,
+  0.0005569425993598998,
+  -0.001412136247381568,
+  -0.0028663526754826307,
+  -0.005061559844762087,
+  -0.00593328895047307,
+  -0.004319756291806698,
+  -0.004922541789710522,
+  -0.0049067651852965355,
+  -0.003943523392081261,
+  -0.003319827374070883,
+  -0.005581587553024292,
+  -0.0034480984322726727
+ ],
+ "verbalizer_ids": [
+  3721,
+  1802,
+  15,
+  16,
+  17,
+  18,
+  19,
+  20,
+  32,
+  33,
+  34,
+  35,
+  36,
+  37,
+  38,
+  39,
+  40,
+  41,
+  42,
+  43,
+  44,
+  45,
+  46,
+  47
+ ],
+ "slots": {
+  "num_slots": 24,
+  "ranges": {
+   "noul": [
+    0,
+    2
+   ],
+   "score": [
+    2,
+    8
+   ],
+   "choice": [
+    8,
+    24
+   ]
+  },
+  "verbalizers": [
+   "false",
+   "true",
+   "0",
+   "1",
+   "2",
+   "3",
+   "4",
+   "5",
+   "A",
+   "B",
+   "C",
+   "D",
+   "E",
+   "F",
+   "G",
+   "H",
+   "I",
+   "J",
+   "K",
+   "L",
+   "M",
+   "N",
+   "O",
+   "P"
+  ],
+  "template_version": "bare-v1"
+ },
+ "note": "decision logits = lm_head-LoRA logprobs of verbalizer ids + bias, then per-kind temperature"
+}

e16/S2-s-brier/ckpt-4000/calibration.json

@@ -0,0 +1,17 @@
+{
+ "version": 1,
+ "per_kind": {
+  "noul": 0.8232693926701515,
+  "choice": 1.0,
+  "score": 1.0
+ },
+ "fit": {
+  "noul": {
+   "n": 2000,
+   "T": 0.8232693926701515,
+   "ce_before": 0.551962673664093,
+   "ce_after": 0.5483428835868835
+  }
+ },
+ "source": "models/e16/S2-s-brier/ckpt-4000"
+}

e16/S2-s-brier/ckpt-4000/head.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:70f5d6f75f2eca98ac986a555686d843ecd7fb4ecdc5f6810f75cf6a5f56aa8b
+size 393464

e16/S2-s-brier/ckpt-4000/judge_config.json

@@ -0,0 +1,160 @@
+{
+ "base_model_path": "/root/models/Qwen3.5-9B",
+ "hidden_size": 4096,
+ "slots": {
+  "num_slots": 24,
+  "ranges": {
+   "noul": [
+    0,
+    2
+   ],
+   "score": [
+    2,
+    8
+   ],
+   "choice": [
+    8,
+    24
+   ]
+  },
+  "verbalizers": [
+   "false",
+   "true",
+   "0",
+   "1",
+   "2",
+   "3",
+   "4",
+   "5",
+   "A",
+   "B",
+   "C",
+   "D",
+   "E",
+   "F",
+   "G",
+   "H",
+   "I",
+   "J",
+   "K",
+   "L",
+   "M",
+   "N",
+   "O",
+   "P"
+  ],
+  "template_version": "bare-v1"
+ },
+ "verbalizer_ids": [
+  3721,
+  1802,
+  15,
+  16,
+  17,
+  18,
+  19,
+  20,
+  32,
+  33,
+  34,
+  35,
+  36,
+  37,
+  38,
+  39,
+  40,
+  41,
+  42,
+  43,
+  44,
+  45,
+  46,
+  47
+ ],
+ "kinds": [
+  "noul",
+  "choice",
+  "score"
+ ],
+ "model_name": "jev-judge-qwen35-9b",
+ "model_version": "0.8.0",
+ "lm_head_included": true,
+ "weights_mode": "unmerged",
+ "adapter_subfolder": "adapter",
+ "lora": {
+  "r": 16,
+  "lora_alpha": 32,
+  "lora_dropout": 0.05,
+  "target_modules": [
+   "down_proj",
+   "gate_proj",
+   "in_proj_qkv",
+   "in_proj_z",
+   "k_proj",
+   "o_proj",
+   "out_proj",
+   "q_proj",
+   "up_proj",
+   "v_proj"
+  ]
+ },
+ "trained_stage": "e16",
+ "source_checkpoint": "models/e16/S2-s-brier/ckpt-4000",
+ "softcap": null,
+ "e16": {
+  "data": "data/e15/F1-pool",
+  "data_manifest": {
+   "arm": "F1-pool",
+   "name": "F1-pool",
+   "train_sha256": "dad16f94a5659768b254a6c12538c01f9c64cd43d574a3435bcc70c9c1bac130",
+   "n_train": 64000
+  },
+  "init": "released",
+  "objective": "s-brier",
+  "sigma": 0.5,
+  "K": 8,
+  "reward": "brier",
+  "clip": 0.2,
+  "passes": 1,
+  "lr": 0.0001,
+  "lr_head": 0.0002,
+  "rows_per_step": 32,
+  "train_seed": 0,
+  "replay": null,
+  "replay_frac": 0.0,
+  "replay_sha256": null,
+  "max_len": 3072,
+  "schedule": "warmup 0.03 + cosine to x0.05 over 500 steps",
+  "total_rows": 16000,
+  "lora": {
+   "r": 16,
+   "lora_alpha": 32,
+   "lora_dropout": 0.05
+  },
+  "trainer_sha256": "bc7c08e836bd99bb2665b02903cd41404ea029ecfa9b68a9ec1a7cda96958ba4",
+  "event": "ckpt",
+  "tag": "ckpt-4000",
+  "rows_seen": 4000,
+  "replay_seen": 0,
+  "step": 125,
+  "tokens": 2631181,
+  "tok_per_s": 759.5964960831906,
+  "rows_per_s": 1.1547612955396462,
+  "val": {
+   "n": 2000,
+   "dropped_long": 0,
+   "noul_ce": 0.551962673664093,
+   "noul_T": 0.8232693926701515,
+   "noul_auc": 0.7657923500928355,
+   "noul_acc": 0.7235,
+   "noul_f1": 0.791713747645951,
+   "noul_nll": 0.5519626314677298,
+   "noul_ece": 0.0758846567571163,
+   "noul_brier": 0.1852127751529852,
+   "noul_sharp": 0.20000439882278442,
+   "entropy": 0.5864689946174622
+  },
+  "dropped_long": 0,
+  "peak_gib": 18.23346519470215
+ }
+}

e16/S2-s-brier/train_log.jsonl

@@ -0,0 +1,58 @@
+{"time": "12:32:29", "event": "start", "data": "data/e15/F1-pool", "data_manifest": {"arm": "F1-pool", "name": "F1-pool", "train_sha256": "dad16f94a5659768b254a6c12538c01f9c64cd43d574a3435bcc70c9c1bac130", "n_train": 64000}, "init": "released", "objective": "s-brier", "sigma": 0.5, "K": 8, "reward": "brier", "clip": 0.2, "passes": 1, "lr": 0.0001, "lr_head": 0.0002, "rows_per_step": 32, "train_seed": 0, "replay": null, "replay_frac": 0.0, "replay_sha256": null, "max_len": 3072, "schedule": "warmup 0.03 + cosine to x0.05 over 500 steps", "total_rows": 16000, "lora": {"r": 16, "lora_alpha": 32, "lora_dropout": 0.05}, "trainer_sha256": "bc7c08e836bd99bb2665b02903cd41404ea029ecfa9b68a9ec1a7cda96958ba4", "device": "cuda", "load_s": 84.2217948436737}
+{"time": "12:36:27", "event": "step", "step": 10, "rows_seen": 320, "lr": 7.333333333333333e-05, "tok_per_s": 887.7133539400976, "loss": 0.4414525479078293, "entropy": 0.4272823892533779, "sharp": 0.3068898022174835}
+{"time": "12:40:28", "event": "step", "step": 20, "rows_seen": 640, "lr": 9.997508952966197e-05, "tok_per_s": 890.328518703689, "loss": 0.42932236567139626, "entropy": 0.5669348277151585, "sharp": 0.21654935739934444}
+{"time": "12:44:25", "event": "step", "step": 30, "rows_seen": 960, "lr": 9.977596250541417e-05, "tok_per_s": 888.454853783273, "loss": 0.4132920429110527, "entropy": 0.535208135843277, "sharp": 0.244880685582757}
+{"time": "12:48:26", "event": "step", "step": 40, "rows_seen": 1280, "lr": 9.937854366485982e-05, "tok_per_s": 890.405855871193, "loss": 0.4798519182950258, "entropy": 0.44216392282396555, "sharp": 0.3142201863229275}
+{"time": "12:52:24", "event": "step", "step": 50, "rows_seen": 1600, "lr": 9.8784499920729e-05, "tok_per_s": 890.8948830202033, "loss": 0.4949744922341779, "entropy": 0.5572843197733164, "sharp": 0.21984022296965122}
+{"time": "12:56:21", "event": "step", "step": 60, "rows_seen": 1920, "lr": 9.79963228989294e-05, "tok_per_s": 890.7498840052946, "loss": 0.43403394683264196, "entropy": 0.5772904977202415, "sharp": 0.202249925583601}
+{"time": "13:00:16", "event": "step", "step": 70, "rows_seen": 2240, "lr": 9.701731848780134e-05, "tok_per_s": 890.0316027911451, "loss": 0.39701841422356665, "entropy": 0.5190924424678087, "sharp": 0.25132922641932964}
+{"time": "13:04:13", "event": "step", "step": 80, "rows_seen": 2560, "lr": 9.585159297207315e-05, "tok_per_s": 890.3198192435162, "loss": 0.29384247912093997, "entropy": 0.5687758466228843, "sharp": 0.2049234639853239}
+{"time": "13:08:12", "event": "step", "step": 90, "rows_seen": 2880, "lr": 9.450403580967612e-05, "tok_per_s": 889.6719504300371, "loss": 0.3973296588519588, "entropy": 0.4598036725074053, "sharp": 0.28288858011364937}
+{"time": "13:12:08", "event": "step", "step": 100, "rows_seen": 3200, "lr": 9.298029912365855e-05, "tok_per_s": 890.4242985745586, "loss": 0.33592444006353617, "entropy": 0.5535668320953846, "sharp": 0.22449850477278233}
+{"time": "13:13:52", "event": "val", "step": 100, "rows_seen": 3200, "n": 500, "dropped_long": 0, "noul_ce": 0.533730685710907, "noul_T": 0.8768400806344038, "noul_auc": 0.7817060381450079, "noul_acc": 0.714, "noul_f1": 0.7748031496062993, "noul_nll": 0.5337306779772043, "noul_ece": 0.07170516574382782, "noul_brier": 0.17970915210117608, "noul_sharp": 0.2136182188987732, "entropy": 0.55902498960495}
+{"time": "13:17:49", "event": "step", "step": 110, "rows_seen": 3520, "lr": 9.128677399521752e-05, "tok_per_s": 857.4213174279539, "loss": 0.4935409121681005, "entropy": 0.48422031477093697, "sharp": 0.2551578115671873}
+{"time": "13:21:36", "event": "step", "step": 120, "rows_seen": 3840, "lr": 8.943056365728335e-05, "tok_per_s": 860.1997337142207, "loss": 0.46765914710704237, "entropy": 0.5555416755378246, "sharp": 0.21540865488350391}
+{"time": "13:30:15", "event": "ckpt", "tag": "ckpt-4000", "rows_seen": 4000, "replay_seen": 0, "step": 125, "tokens": 2631181, "tok_per_s": 759.5964960831906, "rows_per_s": 1.1547612955396462, "val": {"n": 2000, "dropped_long": 0, "noul_ce": 0.551962673664093, "noul_T": 0.8232693926701515, "noul_auc": 0.7657923500928355, "noul_acc": 0.7235, "noul_f1": 0.791713747645951, "noul_nll": 0.5519626314677298, "noul_ece": 0.0758846567571163, "noul_brier": 0.1852127751529852, "noul_sharp": 0.20000439882278442, "entropy": 0.5864689946174622}, "dropped_long": 0, "peak_gib": 18.23346519470215}
+{"time": "13:32:14", "event": "step", "step": 130, "rows_seen": 4160, "lr": 8.741945370109202e-05, "tok_per_s": 764.1075830215013, "loss": 0.43129769805818796, "entropy": 0.5826413333415985, "sharp": 0.2060139775276184}
+{"time": "13:36:10", "event": "step", "step": 140, "rows_seen": 4480, "lr": 8.526187942070997e-05, "tok_per_s": 771.9737266882441, "loss": 0.422426208620891, "entropy": 0.3841750221326947, "sharp": 0.34726870246231556}
+{"time": "13:40:07", "event": "step", "step": 150, "rows_seen": 4800, "lr": 8.296689043247915e-05, "tok_per_s": 779.1484628620079, "loss": 0.3209194848604966, "entropy": 0.467545410618186, "sharp": 0.2662421576678753}
+{"time": "13:44:06", "event": "step", "step": 160, "rows_seen": 5120, "lr": 8.054411271778132e-05, "tok_per_s": 785.812795050819, "loss": 0.2992526153102517, "entropy": 0.623009730130434, "sharp": 0.16614416521042585}
+{"time": "13:48:03", "event": "step", "step": 170, "rows_seen": 5440, "lr": 7.800370824832674e-05, "tok_per_s": 791.4086313683398, "loss": 0.2998864183027763, "entropy": 0.40607616771012545, "sharp": 0.3329711798578501}
+{"time": "13:51:51", "event": "step", "step": 180, "rows_seen": 5760, "lr": 7.535633236331324e-05, "tok_per_s": 796.3523645284316, "loss": 0.32280976604670286, "entropy": 0.5591683089733124, "sharp": 0.22262093611061573}
+{"time": "13:55:44", "event": "step", "step": 190, "rows_seen": 6080, "lr": 7.26130890772308e-05, "tok_per_s": 801.1754499072064, "loss": 0.41964696906507015, "entropy": 0.5653395541012287, "sharp": 0.2128626275807619}
+{"time": "13:59:49", "event": "step", "step": 200, "rows_seen": 6400, "lr": 6.978548450576568e-05, "tok_per_s": 805.5023526710341, "loss": 0.4207197832874954, "entropy": 0.54399748519063, "sharp": 0.2068555559962988}
+{"time": "14:01:33", "event": "val", "step": 200, "rows_seen": 6400, "n": 500, "dropped_long": 0, "noul_ce": 0.5379700660705566, "noul_T": 1.1070102993985047, "noul_auc": 0.7747560390337547, "noul_acc": 0.736, "noul_f1": 0.8181818181818182, "noul_nll": 0.5379700822010637, "noul_ece": 0.08263294816017151, "noul_brier": 0.18064423958647796, "noul_sharp": 0.2315235584974289, "entropy": 0.5166136622428894}
+{"time": "14:05:30", "event": "step", "step": 210, "rows_seen": 6720, "lr": 6.688537860515266e-05, "tok_per_s": 794.2829110030689, "loss": 0.34419260872527957, "entropy": 0.5198629926890135, "sharp": 0.2350244987756014}
+{"time": "14:09:17", "event": "step", "step": 220, "rows_seen": 7040, "lr": 6.392493542739706e-05, "tok_per_s": 798.4607991604337, "loss": 0.37870556442067027, "entropy": 0.4694436024874449, "sharp": 0.29127724282443523}
+{"time": "14:13:14", "event": "step", "step": 230, "rows_seen": 7360, "lr": 6.091657210001309e-05, "tok_per_s": 802.6036865730832, "loss": 0.4055229537189007, "entropy": 0.5255146455019712, "sharp": 0.24926473200321198}
+{"time": "14:17:01", "event": "step", "step": 240, "rows_seen": 7680, "lr": 5.7872906744276015e-05, "tok_per_s": 806.0497691838353, "loss": 0.32633032114245, "entropy": 0.5132154701277614, "sharp": 0.24441024102270603}
+{"time": "14:20:56", "event": "step", "step": 250, "rows_seen": 8000, "lr": 5.48067055504363e-05, "tok_per_s": 809.2647450747849, "loss": 0.23868821212090552, "entropy": 0.5686109773814678, "sharp": 0.20598986186087132}
+{"time": "14:24:43", "event": "step", "step": 260, "rows_seen": 8320, "lr": 5.173082923188119e-05, "tok_per_s": 812.2141030477625, "loss": 0.3346852413378656, "entropy": 0.5494738649576902, "sharp": 0.21925760433077812}
+{"time": "14:28:38", "event": "step", "step": 270, "rows_seen": 8640, "lr": 4.865817908283331e-05, "tok_per_s": 814.8538700969561, "loss": 0.3514334708452225, "entropy": 0.49151576682925224, "sharp": 0.2555016726255417}
+{"time": "14:32:34", "event": "step", "step": 280, "rows_seen": 8960, "lr": 4.560164286583922e-05, "tok_per_s": 817.6596014404188, "loss": 0.33947887597605586, "entropy": 0.39401742350310087, "sharp": 0.32053098268806934}
+{"time": "14:36:33", "event": "step", "step": 290, "rows_seen": 9280, "lr": 4.257404075601443e-05, "tok_per_s": 820.2975867603864, "loss": 0.2645078123314306, "entropy": 0.49356093630194664, "sharp": 0.24763088580220938}
+{"time": "14:40:27", "event": "step", "step": 300, "rows_seen": 9600, "lr": 3.958807156877371e-05, "tok_per_s": 822.4269946458418, "loss": 0.3668871977715753, "entropy": 0.5365531910210848, "sharp": 0.21875404752790928}
+{"time": "14:42:12", "event": "val", "step": 300, "rows_seen": 9600, "n": 500, "dropped_long": 0, "noul_ce": 0.542827308177948, "noul_T": 1.2433840580699096, "noul_auc": 0.7695568709006558, "noul_acc": 0.74, "noul_f1": 0.8115942028985508, "noul_nll": 0.5428273191135377, "noul_ece": 0.05970746827125549, "noul_brier": 0.1801440434078972, "noul_sharp": 0.251962810754776, "entropy": 0.4912777841091156}
+{"time": "14:46:14", "event": "step", "step": 310, "rows_seen": 9920, "lr": 3.665625949658603e-05, "tok_per_s": 813.79572822501, "loss": 0.49841352086514235, "entropy": 0.49018238484859467, "sharp": 0.2688685143366456}
+{"time": "14:50:08", "event": "step", "step": 320, "rows_seen": 10240, "lr": 3.379090157815887e-05, "tok_per_s": 816.1038646626967, "loss": 0.2856982562225312, "entropy": 0.474627198651433, "sharp": 0.2914400789886713}
+{"time": "14:54:02", "event": "step", "step": 330, "rows_seen": 10560, "lr": 3.10040161203844e-05, "tok_per_s": 818.1714411319052, "loss": 0.4961167573928833, "entropy": 0.5218038335442543, "sharp": 0.23648016154766083}
+{"time": "14:58:02", "event": "step", "step": 340, "rows_seen": 10880, "lr": 2.8307292289383776e-05, "tok_per_s": 820.4238152938354, "loss": 0.43578588659875095, "entropy": 0.5656474810093641, "sharp": 0.21665999665856361}
+{"time": "15:01:58", "event": "step", "step": 350, "rows_seen": 11200, "lr": 2.571204108208194e-05, "tok_per_s": 822.5026160823783, "loss": 0.4133405309403315, "entropy": 0.5317072132602334, "sharp": 0.2343736905604601}
+{"time": "15:05:49", "event": "step", "step": 360, "rows_seen": 11520, "lr": 2.322914788395595e-05, "tok_per_s": 824.3117917170531, "loss": 0.2804553834721446, "entropy": 0.5401455946266651, "sharp": 0.23471914790570736}
+{"time": "15:09:42", "event": "step", "step": 370, "rows_seen": 11840, "lr": 2.0869026811945157e-05, "tok_per_s": 826.1760062007451, "loss": 0.341657642275095, "entropy": 0.5392973851412535, "sharp": 0.23795277997851372}
+{"time": "15:13:37", "event": "step", "step": 380, "rows_seen": 12160, "lr": 1.8641577034025627e-05, "tok_per_s": 827.9427502804693, "loss": 0.377601636108011, "entropy": 0.5328305251896381, "sharp": 0.23643611185252666}
+{"time": "15:17:32", "event": "step", "step": 390, "rows_seen": 12480, "lr": 1.6556141248659472e-05, "tok_per_s": 829.5761050604551, "loss": 0.2509818379767239, "entropy": 0.48075176402926445, "sharp": 0.2709543723613024}
+{"time": "15:21:29", "event": "step", "step": 400, "rows_seen": 12800, "lr": 1.4621466498270773e-05, "tok_per_s": 831.0003036922243, "loss": 0.18581731512676924, "entropy": 0.5067173345014453, "sharp": 0.25209457986056805}
+{"time": "15:23:14", "event": "val", "step": 400, "rows_seen": 12800, "n": 500, "dropped_long": 0, "noul_ce": 0.5334097146987915, "noul_T": 1.1023531530534962, "noul_auc": 0.7747382640999663, "noul_acc": 0.744, "noul_f1": 0.8202247191011236, "noul_nll": 0.5334097525402903, "noul_ece": 0.057135584354400644, "noul_brier": 0.1783033972145633, "noul_sharp": 0.2438567578792572, "entropy": 0.5111835598945618}
+{"time": "15:27:15", "event": "step", "step": 410, "rows_seen": 13120, "lr": 1.284566748110963e-05, "tok_per_s": 824.0087689723431, "loss": 0.20884068962186575, "entropy": 0.43983631767332554, "sharp": 0.30192769318819046}
+{"time": "15:31:10", "event": "step", "step": 420, "rows_seen": 13440, "lr": 1.1236192515387821e-05, "tok_per_s": 825.7142831726933, "loss": 0.4004824161529541, "entropy": 0.5324106924235821, "sharp": 0.22574284486472607}
+{"time": "15:35:10", "event": "step", "step": 430, "rows_seen": 13760, "lr": 9.799792298443728e-06, "tok_per_s": 827.1407608287284, "loss": 0.38774158619344234, "entropy": 0.5050091594457626, "sharp": 0.2526229238137603}
+{"time": "15:39:01", "event": "step", "step": 440, "rows_seen": 14080, "lr": 8.542491591971554e-06, "tok_per_s": 828.6033797368447, "loss": 0.3097478752024472, "entropy": 0.4575989469885826, "sharp": 0.27676453441381454}
+{"time": "15:42:59", "event": "step", "step": 450, "rows_seen": 14400, "lr": 7.469563952077155e-06, "tok_per_s": 829.946629915199, "loss": 0.3112135879928246, "entropy": 0.5036396309733391, "sharp": 0.2565087042748928}
+{"time": "15:46:53", "event": "step", "step": 460, "rows_seen": 14720, "lr": 6.585509610150972e-06, "tok_per_s": 831.2747730864423, "loss": 0.4279343258822337, "entropy": 0.4059394393116236, "sharp": 0.32562578096985817}
+{"time": "15:50:53", "event": "step", "step": 470, "rows_seen": 15040, "lr": 5.894036597333646e-06, "tok_per_s": 832.6456431699711, "loss": 0.3151024814287666, "entropy": 0.515789058059454, "sharp": 0.22724274266511202}
+{"time": "15:54:47", "event": "step", "step": 480, "rows_seen": 15360, "lr": 5.398045191744809e-06, "tok_per_s": 833.7431282949841, "loss": 0.3808067631907761, "entropy": 0.4804289434105158, "sharp": 0.2723103426396847}
+{"time": "15:58:42", "event": "step", "step": 490, "rows_seen": 15680, "lr": 5.099615753708697e-06, "tok_per_s": 835.0373660501426, "loss": 0.38076886837370694, "entropy": 0.45202462002635, "sharp": 0.285292437300086}
+{"time": "16:02:40", "event": "step", "step": 500, "rows_seen": 16000, "lr": 5e-06, "tok_per_s": 836.1367145354955, "loss": 0.24354271031916142, "entropy": 0.485270693898201, "sharp": 0.26418536901474}
+{"time": "16:04:25", "event": "val", "step": 500, "rows_seen": 16000, "n": 500, "dropped_long": 0, "noul_ce": 0.5328580737113953, "noul_T": 1.1574036459009038, "noul_auc": 0.7761958086706127, "noul_acc": 0.736, "noul_f1": 0.8023952095808383, "noul_nll": 0.5328581131622195, "noul_ece": 0.059142975032329556, "noul_brier": 0.17824585024506803, "noul_sharp": 0.2607003152370453, "entropy": 0.4964542090892792}
+{"time": "16:11:17", "event": "ckpt", "tag": "ckpt-16000", "rows_seen": 16000, "replay_seen": 0, "step": 500, "tokens": 10544036, "tok_per_s": 803.3155856789108, "rows_per_s": 1.218987621624183, "val": {"n": 2000, "dropped_long": 0, "noul_ce": 0.5325164198875427, "noul_T": 1.1541319454861183, "noul_auc": 0.7755146700071789, "noul_acc": 0.737, "noul_f1": 0.8050407709414381, "noul_nll": 0.5325164518207312, "noul_ece": 0.04705242341756822, "noul_brier": 0.17783741743174705, "noul_sharp": 0.2612763047218323, "entropy": 0.4965769648551941}, "dropped_long": 0, "peak_gib": 18.236003398895264}