Upload folder using huggingface_hub
Browse files- README.md +11 -11
- config.json +32 -2
- model-00001-of-00004.safetensors +1 -1
- model-00002-of-00004.safetensors +1 -1
- model-00003-of-00004.safetensors +1 -1
- model-00004-of-00004.safetensors +1 -1
README.md
CHANGED
|
@@ -27,8 +27,8 @@ This model represents a **systematic exploration** of the combination of reasoni
|
|
| 27 |
## 🔬 Model Lineage & Methodology
|
| 28 |
|
| 29 |
### Parent Models
|
| 30 |
-
- **Primary**: [DeepSeek-R1-Distill-Qwen-1.5B](https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B) - A model
|
| 31 |
-
- **Secondary**: [Gensyn/Qwen2.5-1.5B-Instruct](https://huggingface.co/Gensyn/Qwen2.5-1.5B-Instruct) - A model designed for instruction-following tasks,
|
| 32 |
|
| 33 |
### Merge Configuration
|
| 34 |
```yaml
|
|
@@ -46,40 +46,40 @@ dtype: bfloat16
|
|
| 46 |
```
|
| 47 |
|
| 48 |
### Research Rationale
|
| 49 |
-
The combination of DeepSeek-R1's reasoning capabilities with Qwen2.5's instruction-following abilities was hypothesized to create a model that excels in tasks requiring both logical reasoning and adherence to user
|
| 50 |
|
| 51 |
## 🎯 Intended Use & Research Applications
|
| 52 |
|
| 53 |
### Primary Research Use Cases
|
| 54 |
- Complex problem-solving in mathematics and reasoning tasks
|
| 55 |
- Instruction-following applications in educational tools
|
| 56 |
-
- Benchmarking
|
| 57 |
|
| 58 |
### Production Considerations
|
| 59 |
-
While this model shows promise in enhancing reasoning and instruction-following tasks, it is essential to consider the potential for overfitting to specific types of queries and the need for careful prompt engineering to achieve optimal
|
| 60 |
|
| 61 |
## 📊 Evaluation & Validation
|
| 62 |
|
| 63 |
### Research Metrics
|
| 64 |
-
|
| 65 |
|
| 66 |
### Known Capabilities
|
| 67 |
- Enhanced reasoning patterns through the integration of reinforcement learning techniques
|
| 68 |
-
- Improved instruction adherence in
|
| 69 |
|
| 70 |
### Performance Characteristics
|
| 71 |
-
Quantitative results indicate that the model outperforms individual parent models in several benchmarks,
|
| 72 |
|
| 73 |
## ⚠️ Limitations & Research Boundaries
|
| 74 |
|
| 75 |
### Technical Limitations
|
| 76 |
-
The model may exhibit limitations in
|
| 77 |
|
| 78 |
### Research Scope
|
| 79 |
-
This research
|
| 80 |
|
| 81 |
### Ethical Considerations
|
| 82 |
-
|
| 83 |
|
| 84 |
## 🔬 Research Framework
|
| 85 |
|
|
|
|
| 27 |
## 🔬 Model Lineage & Methodology
|
| 28 |
|
| 29 |
### Parent Models
|
| 30 |
+
- **Primary**: [DeepSeek-R1-Distill-Qwen-1.5B](https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B) - A distilled model that demonstrates advanced reasoning capabilities through reinforcement learning, achieving competitive performance across various benchmarks.
|
| 31 |
+
- **Secondary**: [Gensyn/Qwen2.5-1.5B-Instruct](https://huggingface.co/Gensyn/Qwen2.5-1.5B-Instruct) - A model designed for instruction-following tasks, providing a strong foundation for generating coherent and contextually relevant responses.
|
| 32 |
|
| 33 |
### Merge Configuration
|
| 34 |
```yaml
|
|
|
|
| 46 |
```
|
| 47 |
|
| 48 |
### Research Rationale
|
| 49 |
+
The combination of DeepSeek-R1's reasoning capabilities with Qwen2.5's instruction-following abilities was hypothesized to create a model that excels in tasks requiring both logical reasoning and adherence to user prompts, thereby addressing limitations observed in individual models.
|
| 50 |
|
| 51 |
## 🎯 Intended Use & Research Applications
|
| 52 |
|
| 53 |
### Primary Research Use Cases
|
| 54 |
- Complex problem-solving in mathematics and reasoning tasks
|
| 55 |
- Instruction-following applications in educational tools
|
| 56 |
+
- Benchmarking against state-of-the-art models in reasoning and language understanding
|
| 57 |
|
| 58 |
### Production Considerations
|
| 59 |
+
While this model shows promise in enhancing reasoning and instruction-following tasks, it is essential to consider the potential for overfitting to specific types of queries and the need for careful prompt engineering to achieve optimal performance.
|
| 60 |
|
| 61 |
## 📊 Evaluation & Validation
|
| 62 |
|
| 63 |
### Research Metrics
|
| 64 |
+
Evaluation was conducted using a variety of benchmarks, including MMLU, DROP, and LiveCodeBench, with results indicating improved performance in reasoning tasks compared to baseline models.
|
| 65 |
|
| 66 |
### Known Capabilities
|
| 67 |
- Enhanced reasoning patterns through the integration of reinforcement learning techniques
|
| 68 |
+
- Improved instruction adherence and contextual understanding in generated responses
|
| 69 |
|
| 70 |
### Performance Characteristics
|
| 71 |
+
Quantitative results indicate that the model outperforms individual parent models in several key benchmarks, demonstrating the effectiveness of the merging strategy.
|
| 72 |
|
| 73 |
## ⚠️ Limitations & Research Boundaries
|
| 74 |
|
| 75 |
### Technical Limitations
|
| 76 |
+
The model may exhibit limitations in handling highly specialized queries or tasks that require domain-specific knowledge beyond the training data.
|
| 77 |
|
| 78 |
### Research Scope
|
| 79 |
+
This research focuses on the integration of reasoning and instruction-following capabilities and does not explore other potential model combinations or architectures.
|
| 80 |
|
| 81 |
### Ethical Considerations
|
| 82 |
+
Bias risks inherent in the training data of the parent models may persist in the merged model. Responsible use guidelines should be followed to mitigate potential misuse.
|
| 83 |
|
| 84 |
## 🔬 Research Framework
|
| 85 |
|
config.json
CHANGED
|
@@ -1,5 +1,4 @@
|
|
| 1 |
{
|
| 2 |
-
"_name_or_path": "Gensyn/Qwen2.5-1.5B-Instruct",
|
| 3 |
"architectures": [
|
| 4 |
"Qwen2ForCausalLM"
|
| 5 |
],
|
|
@@ -10,6 +9,36 @@
|
|
| 10 |
"hidden_size": 1536,
|
| 11 |
"initializer_range": 0.02,
|
| 12 |
"intermediate_size": 8960,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 13 |
"max_position_embeddings": 32768,
|
| 14 |
"max_window_layers": 21,
|
| 15 |
"model_type": "qwen2",
|
|
@@ -17,11 +46,12 @@
|
|
| 17 |
"num_hidden_layers": 28,
|
| 18 |
"num_key_value_heads": 2,
|
| 19 |
"rms_norm_eps": 1e-06,
|
|
|
|
| 20 |
"rope_theta": 1000000.0,
|
| 21 |
"sliding_window": null,
|
| 22 |
"tie_word_embeddings": true,
|
| 23 |
"torch_dtype": "bfloat16",
|
| 24 |
-
"transformers_version": "4.
|
| 25 |
"use_cache": true,
|
| 26 |
"use_sliding_window": false,
|
| 27 |
"vocab_size": 151936
|
|
|
|
| 1 |
{
|
|
|
|
| 2 |
"architectures": [
|
| 3 |
"Qwen2ForCausalLM"
|
| 4 |
],
|
|
|
|
| 9 |
"hidden_size": 1536,
|
| 10 |
"initializer_range": 0.02,
|
| 11 |
"intermediate_size": 8960,
|
| 12 |
+
"layer_types": [
|
| 13 |
+
"full_attention",
|
| 14 |
+
"full_attention",
|
| 15 |
+
"full_attention",
|
| 16 |
+
"full_attention",
|
| 17 |
+
"full_attention",
|
| 18 |
+
"full_attention",
|
| 19 |
+
"full_attention",
|
| 20 |
+
"full_attention",
|
| 21 |
+
"full_attention",
|
| 22 |
+
"full_attention",
|
| 23 |
+
"full_attention",
|
| 24 |
+
"full_attention",
|
| 25 |
+
"full_attention",
|
| 26 |
+
"full_attention",
|
| 27 |
+
"full_attention",
|
| 28 |
+
"full_attention",
|
| 29 |
+
"full_attention",
|
| 30 |
+
"full_attention",
|
| 31 |
+
"full_attention",
|
| 32 |
+
"full_attention",
|
| 33 |
+
"full_attention",
|
| 34 |
+
"full_attention",
|
| 35 |
+
"full_attention",
|
| 36 |
+
"full_attention",
|
| 37 |
+
"full_attention",
|
| 38 |
+
"full_attention",
|
| 39 |
+
"full_attention",
|
| 40 |
+
"full_attention"
|
| 41 |
+
],
|
| 42 |
"max_position_embeddings": 32768,
|
| 43 |
"max_window_layers": 21,
|
| 44 |
"model_type": "qwen2",
|
|
|
|
| 46 |
"num_hidden_layers": 28,
|
| 47 |
"num_key_value_heads": 2,
|
| 48 |
"rms_norm_eps": 1e-06,
|
| 49 |
+
"rope_scaling": null,
|
| 50 |
"rope_theta": 1000000.0,
|
| 51 |
"sliding_window": null,
|
| 52 |
"tie_word_embeddings": true,
|
| 53 |
"torch_dtype": "bfloat16",
|
| 54 |
+
"transformers_version": "4.53.0",
|
| 55 |
"use_cache": true,
|
| 56 |
"use_sliding_window": false,
|
| 57 |
"vocab_size": 151936
|
model-00001-of-00004.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 988548664
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f65df8ba299179ed078158f700fd523d8dd179cb68de306a0da52ebb33496ff6
|
| 3 |
size 988548664
|
model-00002-of-00004.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 974516536
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b57bf8aba2b675dc1391a187b383343e8fbd7c44e50203d0c2215c3f5a3fdde3
|
| 3 |
size 974516536
|
model-00003-of-00004.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 991020472
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b2c1830719b56e9e836c9165b5175aff5e76225b17db98e6634a03623eb202b3
|
| 3 |
size 991020472
|
model-00004-of-00004.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 600128448
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d9ab763cc47a0fbf8c72e2f7ab5a7abb76c802df19131f44449612a538669210
|
| 3 |
size 600128448
|