Update README.md
Browse files
README.md
CHANGED
|
@@ -1,3 +1,48 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
# Command-R 35B — SFT (Supervised Fine-Tuning on Synthetic QA)
|
| 2 |
|
| 3 |
**Model type:** Causal Language Model
|
|
|
|
| 1 |
+
---
|
| 2 |
+
{
|
| 3 |
+
"language": ["en"],
|
| 4 |
+
"license": "apache-2.0",
|
| 5 |
+
"tags": [
|
| 6 |
+
"text-generation",
|
| 7 |
+
"causal-lm",
|
| 8 |
+
"instruction-tuning",
|
| 9 |
+
"supervised-fine-tuning",
|
| 10 |
+
"synthetic-qa",
|
| 11 |
+
"lora",
|
| 12 |
+
"axolotl",
|
| 13 |
+
"deepspeed",
|
| 14 |
+
"transformers",
|
| 15 |
+
"commandr",
|
| 16 |
+
"cohere",
|
| 17 |
+
"eu-hpc"
|
| 18 |
+
],
|
| 19 |
+
"datasets": [
|
| 20 |
+
"axolotl_deduplicated_synthetic_qa"
|
| 21 |
+
],
|
| 22 |
+
"metrics": [
|
| 23 |
+
"loss"
|
| 24 |
+
],
|
| 25 |
+
"library_name": "transformers",
|
| 26 |
+
"framework": "pytorch",
|
| 27 |
+
"base_model": "CohereLabs/c4ai-command-r-v01",
|
| 28 |
+
"model_name": "commandr-35b-sft",
|
| 29 |
+
"pipeline_tag": "text-generation",
|
| 30 |
+
"task_categories": ["text-generation", "instruction-following"],
|
| 31 |
+
"model_type": "AutoModelForCausalLM",
|
| 32 |
+
"inference": {
|
| 33 |
+
"parameters": {
|
| 34 |
+
"max_new_tokens": 512,
|
| 35 |
+
"temperature": 0.7,
|
| 36 |
+
"top_p": 0.9
|
| 37 |
+
}
|
| 38 |
+
},
|
| 39 |
+
"trained_on": [
|
| 40 |
+
"Leonardo EuroHPC"
|
| 41 |
+
],
|
| 42 |
+
"description": "Supervised fine-tuning (SFT) of Cohere Command-R 35B on the synthetic QA dataset using LoRA and Axolotl. The model improves conversational reasoning and instruction-following capabilities."
|
| 43 |
+
}
|
| 44 |
+
---
|
| 45 |
+
|
| 46 |
# Command-R 35B — SFT (Supervised Fine-Tuning on Synthetic QA)
|
| 47 |
|
| 48 |
**Model type:** Causal Language Model
|