Upload folder using huggingface_hub

epoch_1/README.md

@@ -0,0 +1,207 @@
+---
+base_model: ./pruned_75_0304_033137_10_rexmoe_natural_phi_mini_moe_R3
+library_name: peft
+pipeline_tag: text-generation
+tags:
+- base_model:adapter:./pruned_75_0304_033137_10_rexmoe_natural_phi_mini_moe_R3
+- 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 0.18.1

epoch_1/adapter_config.json

@@ -0,0 +1,46 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": null,
+  "base_model_name_or_path": "./pruned_75_0304_033137_10_rexmoe_natural_phi_mini_moe_R3",
+  "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,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 32,
+  "lora_bias": false,
+  "lora_dropout": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 16,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "q_proj",
+    "w2",
+    "v_proj",
+    "k_proj",
+    "o_proj",
+    "w3",
+    "w1"
+  ],
+  "target_parameters": null,
+  "task_type": "CAUSAL_LM",
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}

epoch_1/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bf182932a5d8839857fa2f19df686dfcd73b516e70cfaef4a8ae6df4afae5059
+size 179902200

epoch_1/chat_template.jinja

@@ -0,0 +1,4 @@
+{% for message in messages %}{{'<|' + message['role'] + '|>' + '
+' + message['content'] + '<|end|>
+' }}{% endfor %}{% if add_generation_prompt %}{{ '<|assistant|>
+' }}{% else %}{{ eos_token }}{% endif %}

epoch_1/rexmoe_routers.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:97ed82476c09051f842db77d3c1675ea63a48588510368c86402da2b8e07f642
+size 12619762

epoch_1/special_tokens_map.json

@@ -0,0 +1,30 @@
+{
+  "bos_token": {
+    "content": "<s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

epoch_1/tokenizer_config.json

@@ -0,0 +1,131 @@
+{
+  "add_bos_token": false,
+  "add_eos_token": false,
+  "add_prefix_space": null,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": true,
+      "single_word": false,
+      "special": false
+    },
+    "32000": {
+      "content": "<|endoftext|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "32001": {
+      "content": "<|assistant|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": true,
+      "single_word": false,
+      "special": true
+    },
+    "32002": {
+      "content": "<|placeholder1|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": true,
+      "single_word": false,
+      "special": true
+    },
+    "32003": {
+      "content": "<|placeholder2|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": true,
+      "single_word": false,
+      "special": true
+    },
+    "32004": {
+      "content": "<|placeholder3|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": true,
+      "single_word": false,
+      "special": true
+    },
+    "32005": {
+      "content": "<|placeholder4|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": true,
+      "single_word": false,
+      "special": true
+    },
+    "32006": {
+      "content": "<|system|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": true,
+      "single_word": false,
+      "special": true
+    },
+    "32007": {
+      "content": "<|end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": true,
+      "single_word": false,
+      "special": true
+    },
+    "32008": {
+      "content": "<|placeholder5|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": true,
+      "single_word": false,
+      "special": true
+    },
+    "32009": {
+      "content": "<|placeholder6|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": true,
+      "single_word": false,
+      "special": true
+    },
+    "32010": {
+      "content": "<|user|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": true,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "bos_token": "<s>",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|endoftext|>",
+  "extra_special_tokens": {},
+  "legacy": false,
+  "model_max_length": 4096,
+  "pad_token": "<|endoftext|>",
+  "padding_side": "left",
+  "sp_model_kwargs": {},
+  "tokenizer_class": "LlamaTokenizerFast",
+  "unk_token": "<unk>",
+  "use_default_system_prompt": false
+}

epoch_2/README.md

@@ -0,0 +1,207 @@
+---
+base_model: ./pruned_75_0304_033137_10_rexmoe_natural_phi_mini_moe_R3
+library_name: peft
+pipeline_tag: text-generation
+tags:
+- base_model:adapter:./pruned_75_0304_033137_10_rexmoe_natural_phi_mini_moe_R3
+- 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 0.18.1

epoch_2/adapter_config.json

@@ -0,0 +1,46 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": null,
+  "base_model_name_or_path": "./pruned_75_0304_033137_10_rexmoe_natural_phi_mini_moe_R3",
+  "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,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 32,
+  "lora_bias": false,
+  "lora_dropout": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 16,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "q_proj",
+    "w2",
+    "v_proj",
+    "k_proj",
+    "o_proj",
+    "w3",
+    "w1"
+  ],
+  "target_parameters": null,
+  "task_type": "CAUSAL_LM",
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}

epoch_2/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:814ee1e3fe24e3bfb43c350bfbf6eca60f6457148e33ef880a1e5326b5e06e14
+size 179902200

epoch_2/chat_template.jinja

@@ -0,0 +1,4 @@
+{% for message in messages %}{{'<|' + message['role'] + '|>' + '
+' + message['content'] + '<|end|>
+' }}{% endfor %}{% if add_generation_prompt %}{{ '<|assistant|>
+' }}{% else %}{{ eos_token }}{% endif %}

epoch_2/rexmoe_routers.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1fc9311a3ddaa16c1f24cf6386064e55ae143c3b934c679cada605eaaf54eece
+size 12619762

epoch_2/special_tokens_map.json

@@ -0,0 +1,30 @@
+{
+  "bos_token": {
+    "content": "<s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

epoch_2/tokenizer_config.json

@@ -0,0 +1,131 @@
+{
+  "add_bos_token": false,
+  "add_eos_token": false,
+  "add_prefix_space": null,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": true,
+      "single_word": false,
+      "special": false
+    },
+    "32000": {
+      "content": "<|endoftext|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "32001": {
+      "content": "<|assistant|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": true,
+      "single_word": false,
+      "special": true
+    },
+    "32002": {
+      "content": "<|placeholder1|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": true,
+      "single_word": false,
+      "special": true
+    },
+    "32003": {
+      "content": "<|placeholder2|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": true,
+      "single_word": false,
+      "special": true
+    },
+    "32004": {
+      "content": "<|placeholder3|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": true,
+      "single_word": false,
+      "special": true
+    },
+    "32005": {
+      "content": "<|placeholder4|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": true,
+      "single_word": false,
+      "special": true
+    },
+    "32006": {
+      "content": "<|system|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": true,
+      "single_word": false,
+      "special": true
+    },
+    "32007": {
+      "content": "<|end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": true,
+      "single_word": false,
+      "special": true
+    },
+    "32008": {
+      "content": "<|placeholder5|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": true,
+      "single_word": false,
+      "special": true
+    },
+    "32009": {
+      "content": "<|placeholder6|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": true,
+      "single_word": false,
+      "special": true
+    },
+    "32010": {
+      "content": "<|user|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": true,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "bos_token": "<s>",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|endoftext|>",
+  "extra_special_tokens": {},
+  "legacy": false,
+  "model_max_length": 4096,
+  "pad_token": "<|endoftext|>",
+  "padding_side": "left",
+  "sp_model_kwargs": {},
+  "tokenizer_class": "LlamaTokenizerFast",
+  "unk_token": "<unk>",
+  "use_default_system_prompt": false
+}

logs/rexmoe_training_0504_182818.log

@@ -0,0 +1,116 @@
+2026-04-05 18:28:18 - ReXMoE - INFO - ================================================================================
+2026-04-05 18:28:18 - ReXMoE - INFO - ReXMoE Training Log - 0504_182818
+2026-04-05 18:28:18 - ReXMoE - INFO - Log file: ./rexmoe_recovered_75_pruned_75_0304_033137_10_rexmoe_natural_phi_mini_moe_R3/logs/rexmoe_training_0504_182818.log
+2026-04-05 18:28:18 - ReXMoE - INFO - ================================================================================
+2026-04-05 18:28:18 - ReXMoE - INFO - ================================================================================
+2026-04-05 18:28:18 - ReXMoE - INFO - ReXMoE Recovery Training (Fine-tuning Pruned Model)
+2026-04-05 18:28:18 - ReXMoE - INFO - ================================================================================
+2026-04-05 18:28:18 - ReXMoE - INFO - Checkpoint Path: ./pruned_75_0304_033137_10_rexmoe_natural_phi_mini_moe_R3
+2026-04-05 18:28:18 - ReXMoE - INFO - Dataset Path: ../dataset/alpaca_data_cleaned.json
+2026-04-05 18:28:18 - ReXMoE - INFO - Dataset Mode: IF
+2026-04-05 18:28:18 - ReXMoE - INFO - Num Samples: 5000
+2026-04-05 18:28:18 - ReXMoE - INFO - Num Epochs: 1
+2026-04-05 18:28:18 - ReXMoE - INFO - Batch Size: 1
+2026-04-05 18:28:18 - ReXMoE - INFO - Max Seq Length: 512
+2026-04-05 18:28:18 - ReXMoE - INFO - Learning Rate: 0.0001
+2026-04-05 18:28:18 - ReXMoE - INFO - Save Path: ./rexmoe_recovered_75_pruned_75_0304_033137_10_rexmoe_natural_phi_mini_moe_R3
+2026-04-05 18:28:18 - ReXMoE - INFO - Gradient Checkpointing: False
+2026-04-05 18:28:18 - ReXMoE - INFO - Full LoRA (q,k,v,o): True
+2026-04-05 18:28:18 - ReXMoE - INFO - Use Scheduler: True
+2026-04-05 18:28:18 - ReXMoE - INFO - Evaluation Steps: 5000
+2026-04-05 18:28:18 - ReXMoE - INFO - Gradient Checkpointing: False
+2026-04-05 18:28:18 - ReXMoE - INFO - Using device: cuda
+2026-04-05 18:28:18 - ReXMoE - INFO - 
+[1/5] Loading pruned model from: ./pruned_75_0304_033137_10_rexmoe_natural_phi_mini_moe_R3
+2026-04-05 18:35:31 - ReXMoE - INFO - Loading dense model for KD from microsoft/Phi-mini-MoE-instruct
+2026-04-05 18:35:35 - ReXMoE - INFO - 
+[2/5] Setting up LoRA for fine-tuning...
+2026-04-05 18:35:36 - ReXMoE - INFO - 
+  Attempting to load and remap pretrained LoRA from: ./0304_033137_10_rexmoe_natural_phi_mini_moe_R3
+2026-04-05 18:35:36 - ReXMoE - INFO -   ✓ Loaded remapped LoRA weights.
+2026-04-05 18:35:36 - ReXMoE - INFO -     - Remapped expert weights: 0
+2026-04-05 18:35:36 - ReXMoE - INFO -     - Skipped pruned weights: 0
+2026-04-05 18:35:36 - ReXMoE - INFO - Total parameters: 3,178,713,408
+2026-04-05 18:35:36 - ReXMoE - INFO - Trainable parameters: 44,938,240 (1.4137%)
+2026-04-05 18:35:40 - ReXMoE - INFO - 
+Prior Evaluation:
+2026-04-05 18:35:40 - ReXMoE - INFO - 
+Evaluating model with 3 sample prompts...
+2026-04-05 18:36:02 - ReXMoE - INFO - 
+--- Prompt 1/3 ---
+2026-04-05 18:36:02 - ReXMoE - INFO - Instruction: What is the capital of France?
+2026-04-05 18:36:02 - ReXMoE - INFO - Input: None
+2026-04-05 18:36:02 - ReXMoE - INFO - Generated completion (len 100): aaaaaaaaaaaa.aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa wasaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaallaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa
+2026-04-05 18:36:23 - ReXMoE - INFO - 
+--- Prompt 2/3 ---
+2026-04-05 18:36:23 - ReXMoE - INFO - Instruction: High-pressure systems stop air from rising into the colder regions of the atmosphere where water can condense. What will most likely result if a high-pressure system remains in an area for a long period of time?
+A. fog
+B. rain
+C. drought
+D. tornado
+Answer:
+2026-04-05 18:36:23 - ReXMoE - INFO - Input: None
+2026-04-05 18:36:23 - ReXMoE - INFO - Generated completion (len 100): aaaaaaaaaaaaaallaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa asaaaaaaaaaallaaaaaallaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaallaaaaaaaaaaaa
+aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa0aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa
+2026-04-05 18:36:45 - ReXMoE - INFO - 
+--- Prompt 3/3 ---
+2026-04-05 18:36:45 - ReXMoE - INFO - Instruction: Given the fact: predators eat prey
+Question: Predators eat
+A. lions
+B. humans
+C. bunnies
+D. grass
+Answer:
+2026-04-05 18:36:45 - ReXMoE - INFO - Input: None
+2026-04-05 18:36:45 - ReXMoE - INFO - Generated completion (len 100): aall
+aaaaaaaaaaaaaaaaaaaaaaaaaaaallllaaaaaaaaaallaaaa
+aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa
+2026-04-05 18:36:45 - ReXMoE - INFO - Evaluation of all 3 prompts complete.
+2026-04-05 18:36:45 - ReXMoE - INFO - 
+Epoch 1/1
+2026-04-05 18:51:04 - ReXMoE - INFO - Step 500/5000 | Loss: 3.2554 | KD Loss: 0.9844
+2026-04-05 19:06:08 - ReXMoE - INFO - Step 1000/5000 | Loss: 2.0955 | KD Loss: 1.5000
+2026-04-05 19:21:21 - ReXMoE - INFO - Step 1500/5000 | Loss: 2.2041 | KD Loss: 1.0781
+2026-04-05 19:36:38 - ReXMoE - INFO - Step 2000/5000 | Loss: 3.4660 | KD Loss: 1.4219
+2026-04-05 19:51:48 - ReXMoE - INFO - Step 2500/5000 | Loss: 1.8320 | KD Loss: 0.8945
+2026-04-05 20:07:07 - ReXMoE - INFO - Step 3000/5000 | Loss: 2.4157 | KD Loss: 1.4219
+2026-04-05 20:22:22 - ReXMoE - INFO - Step 3500/5000 | Loss: 1.9184 | KD Loss: 1.3516
+2026-04-05 20:37:35 - ReXMoE - INFO - Step 4000/5000 | Loss: 1.4622 | KD Loss: 0.7266
+2026-04-05 20:52:49 - ReXMoE - INFO - Step 4500/5000 | Loss: 1.9962 | KD Loss: 1.2344
+2026-04-05 21:08:04 - ReXMoE - INFO - Step 5000/5000 | Loss: 1.7498 | KD Loss: 1.0000
+2026-04-05 21:08:04 - ReXMoE - INFO - 
+Evaluation at step 5000:
+2026-04-05 21:08:04 - ReXMoE - INFO - 
+Evaluating model with 3 sample prompts...
+2026-04-05 21:08:06 - ReXMoE - INFO - 
+--- Prompt 1/3 ---
+2026-04-05 21:08:06 - ReXMoE - INFO - Instruction: What is the capital of France?
+2026-04-05 21:08:06 - ReXMoE - INFO - Input: None
+2026-04-05 21:08:06 - ReXMoE - INFO - Generated completion (len 9): The capital of France is Paris.
+2026-04-05 21:08:12 - ReXMoE - INFO - 
+--- Prompt 2/3 ---
+2026-04-05 21:08:12 - ReXMoE - INFO - Instruction: High-pressure systems stop air from rising into the colder regions of the atmosphere where water can condense. What will most likely result if a high-pressure system remains in an area for a long period of time?
+A. fog
+B. rain
+C. drought
+D. tornado
+Answer:
+2026-04-05 21:08:12 - ReXMoE - INFO - Input: None
+2026-04-05 21:08:12 - ReXMoE - INFO - Generated completion (len 25): The most likely result of a high-pressure system working up for an extended period of time is a fog.
+2026-04-05 21:08:17 - ReXMoE - INFO - 
+--- Prompt 3/3 ---
+2026-04-05 21:08:17 - ReXMoE - INFO - Instruction: Given the fact: predators eat prey
+Question: Predators eat
+A. lions
+B. humans
+C. bunnies
+D. grass
+Answer:
+2026-04-05 21:08:17 - ReXMoE - INFO - Input: None
+2026-04-05 21:08:17 - ReXMoE - INFO - Generated completion (len 21): A. lions
+B. humans
+C. bunnies
+D. grass.
+2026-04-05 21:08:17 - ReXMoE - INFO - Evaluation of all 3 prompts complete.
+2026-04-05 21:08:17 - ReXMoE - INFO - Epoch 1 completed. Average Loss: nan
+2026-04-05 21:08:21 - ReXMoE - INFO - Training complete.

logs/rexmoe_training_0604_042639.log

@@ -0,0 +1,24 @@
+2026-04-06 04:26:39 - ReXMoE - INFO - ================================================================================
+2026-04-06 04:26:39 - ReXMoE - INFO - ReXMoE Training Log - 0604_042639
+2026-04-06 04:26:39 - ReXMoE - INFO - Log file: ./rexmoe_recovered_75_pruned_75_0304_033137_10_rexmoe_natural_phi_mini_moe_R3/logs/rexmoe_training_0604_042639.log
+2026-04-06 04:26:39 - ReXMoE - INFO - ================================================================================
+2026-04-06 04:26:39 - ReXMoE - INFO - ================================================================================
+2026-04-06 04:26:39 - ReXMoE - INFO - ReXMoE Recovery Training (Fine-tuning Pruned Model)
+2026-04-06 04:26:39 - ReXMoE - INFO - ================================================================================
+2026-04-06 04:26:39 - ReXMoE - INFO - Checkpoint Path: ./pruned_75_0304_033137_10_rexmoe_natural_phi_mini_moe_R3
+2026-04-06 04:26:39 - ReXMoE - INFO - Dataset Path: ../dataset/alpaca_data_cleaned.json
+2026-04-06 04:26:39 - ReXMoE - INFO - Dataset Mode: IF
+2026-04-06 04:26:39 - ReXMoE - INFO - Num Samples: 10000
+2026-04-06 04:26:39 - ReXMoE - INFO - Num Epochs: 1
+2026-04-06 04:26:39 - ReXMoE - INFO - Batch Size: 1
+2026-04-06 04:26:39 - ReXMoE - INFO - Max Seq Length: 512
+2026-04-06 04:26:39 - ReXMoE - INFO - Learning Rate: 0.0001
+2026-04-06 04:26:39 - ReXMoE - INFO - Save Path: ./rexmoe_recovered_75_pruned_75_0304_033137_10_rexmoe_natural_phi_mini_moe_R3
+2026-04-06 04:26:39 - ReXMoE - INFO - Gradient Checkpointing: False
+2026-04-06 04:26:39 - ReXMoE - INFO - Full LoRA (q,k,v,o): True
+2026-04-06 04:26:39 - ReXMoE - INFO - Use Scheduler: True
+2026-04-06 04:26:39 - ReXMoE - INFO - Evaluation Steps: 10000
+2026-04-06 04:26:39 - ReXMoE - INFO - Gradient Checkpointing: False
+2026-04-06 04:26:39 - ReXMoE - INFO - Using device: cuda
+2026-04-06 04:26:39 - ReXMoE - INFO - 
+[1/5] Loading pruned model from: ./pruned_75_0304_033137_10_rexmoe_natural_phi_mini_moe_R3

logs/rexmoe_training_0604_043401.log

@@ -0,0 +1,82 @@
+2026-04-06 04:34:01 - ReXMoE - INFO - ================================================================================
+2026-04-06 04:34:01 - ReXMoE - INFO - ReXMoE Training Log - 0604_043401
+2026-04-06 04:34:01 - ReXMoE - INFO - Log file: ./rexmoe_recovered_75_pruned_75_0304_033137_10_rexmoe_natural_phi_mini_moe_R3/logs/rexmoe_training_0604_043401.log
+2026-04-06 04:34:01 - ReXMoE - INFO - ================================================================================
+2026-04-06 04:34:01 - ReXMoE - INFO - ================================================================================
+2026-04-06 04:34:01 - ReXMoE - INFO - ReXMoE Recovery Training (Fine-tuning Pruned Model)
+2026-04-06 04:34:01 - ReXMoE - INFO - ================================================================================
+2026-04-06 04:34:01 - ReXMoE - INFO - Checkpoint Path: ./pruned_75_0304_033137_10_rexmoe_natural_phi_mini_moe_R3
+2026-04-06 04:34:01 - ReXMoE - INFO - Dataset Path: ../dataset/alpaca_data_cleaned.json
+2026-04-06 04:34:01 - ReXMoE - INFO - Dataset Mode: IF_2
+2026-04-06 04:34:01 - ReXMoE - INFO - Num Samples: 500
+2026-04-06 04:34:01 - ReXMoE - INFO - Num Epochs: 2
+2026-04-06 04:34:01 - ReXMoE - INFO - Batch Size: 1
+2026-04-06 04:34:01 - ReXMoE - INFO - Max Seq Length: 512
+2026-04-06 04:34:01 - ReXMoE - INFO - Learning Rate: 0.0001
+2026-04-06 04:34:01 - ReXMoE - INFO - Save Path: ./rexmoe_recovered_75_pruned_75_0304_033137_10_rexmoe_natural_phi_mini_moe_R3
+2026-04-06 04:34:01 - ReXMoE - INFO - Gradient Checkpointing: False
+2026-04-06 04:34:01 - ReXMoE - INFO - Full LoRA (q,k,v,o): True
+2026-04-06 04:34:01 - ReXMoE - INFO - Use Scheduler: True
+2026-04-06 04:34:01 - ReXMoE - INFO - Evaluation Steps: 10000
+2026-04-06 04:34:01 - ReXMoE - INFO - Gradient Checkpointing: False
+2026-04-06 04:34:01 - ReXMoE - INFO - Using device: cuda
+2026-04-06 04:34:01 - ReXMoE - INFO - 
+[1/5] Loading pruned model from: ./pruned_75_0304_033137_10_rexmoe_natural_phi_mini_moe_R3
+2026-04-06 04:41:17 - ReXMoE - INFO - Loading dense model for KD from microsoft/Phi-mini-MoE-instruct
+2026-04-06 04:41:22 - ReXMoE - INFO - 
+[2/5] Setting up LoRA for fine-tuning...
+2026-04-06 04:41:23 - ReXMoE - INFO - 
+  Attempting to load and remap pretrained LoRA from: ./0304_033137_10_rexmoe_natural_phi_mini_moe_R3
+2026-04-06 04:41:23 - ReXMoE - INFO -   ✓ Loaded remapped LoRA weights.
+2026-04-06 04:41:23 - ReXMoE - INFO -     - Remapped expert weights: 0
+2026-04-06 04:41:23 - ReXMoE - INFO -     - Skipped pruned weights: 0
+2026-04-06 04:41:23 - ReXMoE - INFO - Total parameters: 3,178,713,408
+2026-04-06 04:41:23 - ReXMoE - INFO - Trainable parameters: 44,938,240 (1.4137%)
+2026-04-06 04:41:26 - ReXMoE - INFO - 
+Prior Evaluation:
+2026-04-06 04:41:26 - ReXMoE - INFO - 
+Evaluating model with 3 sample prompts...
+2026-04-06 04:41:51 - ReXMoE - INFO - 
+--- Prompt 1/3 ---
+2026-04-06 04:41:51 - ReXMoE - INFO - Instruction: What is the capital of France?
+2026-04-06 04:41:51 - ReXMoE - INFO - Input: None
+2026-04-06 04:41:51 - ReXMoE - INFO - Generated completion (len 100): aaaaaaaaaaaaaaaaaaaaaaaaaa aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa!aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaallaaaaaaaaaaaa
+2026-04-06 04:42:14 - ReXMoE - INFO - 
+--- Prompt 2/3 ---
+2026-04-06 04:42:14 - ReXMoE - INFO - Instruction: High-pressure systems stop air from rising into the colder regions of the atmosphere where water can condense. What will most likely result if a high-pressure system remains in an area for a long period of time?
+A. fog
+B. rain
+C. drought
+D. tornado
+Answer:
+2026-04-06 04:42:14 - ReXMoE - INFO - Input: None
+2026-04-06 04:42:14 - ReXMoE - INFO - Generated completion (len 100): aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa1aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa
+2026-04-06 04:42:38 - ReXMoE - INFO - 
+--- Prompt 3/3 ---
+2026-04-06 04:42:38 - ReXMoE - INFO - Instruction: Given the fact: predators eat prey
+Question: Predators eat
+A. lions
+B. humans
+C. bunnies
+D. grass
+Answer:
+2026-04-06 04:42:38 - ReXMoE - INFO - Input: None
+2026-04-06 04:42:38 - ReXMoE - INFO - Generated completion (len 100): aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa_aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa
+2026-04-06 04:42:38 - ReXMoE - INFO - Evaluation of all 3 prompts complete.
+2026-04-06 04:42:38 - ReXMoE - INFO - 
+Epoch 1/2
+2026-04-06 04:45:42 - ReXMoE - INFO - Step 100/1000 | Loss: 4.0626 | KD Loss: 1.4453
+2026-04-06 04:49:01 - ReXMoE - INFO - Step 200/1000 | Loss: 4.2264 | KD Loss: 3.7812
+2026-04-06 04:52:20 - ReXMoE - INFO - Step 300/1000 | Loss: 3.6379 | KD Loss: 2.4844
+2026-04-06 04:55:38 - ReXMoE - INFO - Step 400/1000 | Loss: 3.0670 | KD Loss: 1.7109
+2026-04-06 04:58:58 - ReXMoE - INFO - Step 500/1000 | Loss: nan | KD Loss: 2.1875
+2026-04-06 04:58:58 - ReXMoE - INFO - Epoch 1 completed. Average Loss: nan
+2026-04-06 04:59:04 - ReXMoE - INFO - 
+Epoch 2/2
+2026-04-06 05:02:25 - ReXMoE - INFO - Step 600/1000 | Loss: 2.6729 | KD Loss: 1.8906
+2026-04-06 05:05:45 - ReXMoE - INFO - Step 700/1000 | Loss: 1.6074 | KD Loss: 1.2188
+2026-04-06 05:09:07 - ReXMoE - INFO - Step 800/1000 | Loss: 1.8340 | KD Loss: 1.2812
+2026-04-06 05:12:29 - ReXMoE - INFO - Step 900/1000 | Loss: 2.6532 | KD Loss: 1.2891
+2026-04-06 05:15:50 - ReXMoE - INFO - Step 1000/1000 | Loss: 1.7152 | KD Loss: 1.1719
+2026-04-06 05:15:50 - ReXMoE - INFO - Epoch 2 completed. Average Loss: nan
+2026-04-06 05:15:54 - ReXMoE - INFO - Training complete.