Text Generation
PEFT
Safetensors
GGUF
English
materialsanalyst-ai-7b
MaterialsAnalyst-AI-7B
materials-science
computational-materials
materials-analysis
chain-of-thought
reasoning-model
property-prediction
materials-discovery
crystal-structure
materials-informatics
scientific-ai
7b
quantized
fine-tuned
lora
json-mode
structured-output
materials-engineering
band-gap-prediction
computational-chemistry
materials-characterization
Update Training/Training_Documentation.txt
Browse files
Training/Training_Documentation.txt
CHANGED
|
@@ -13,9 +13,12 @@ Training Dataset: Custom curated dataset for materials analysis
|
|
| 13 |
Dataset Specifications
|
| 14 |
---------------------
|
| 15 |
|
| 16 |
-
Total Token Count: 6,
|
| 17 |
Total Sample Count: 6,000
|
| 18 |
-
Average Tokens/Sample:
|
|
|
|
|
|
|
|
|
|
| 19 |
Dataset Creation: Generated using DeepSeekV3 API
|
| 20 |
|
| 21 |
Training Configuration
|
|
|
|
| 13 |
Dataset Specifications
|
| 14 |
---------------------
|
| 15 |
|
| 16 |
+
Total Token Count: 6,292,692
|
| 17 |
Total Sample Count: 6,000
|
| 18 |
+
Average Tokens/Sample: 1048.78
|
| 19 |
+
Max Token Count: 1,289
|
| 20 |
+
Min Token Count: 922
|
| 21 |
+
Tokens Counted Using: tiktoken (cl100k_base encoding)
|
| 22 |
Dataset Creation: Generated using DeepSeekV3 API
|
| 23 |
|
| 24 |
Training Configuration
|