Upload model.yml with huggingface_hub

model.yml

@@ -1,16 +1,48 @@
-name: Codestreal 22B
-model: codestral:22B
-version: 1
-stop:
-  - <endofstring>
-  - [/INST]
-top_p: 0.95
-temperature: 0.7
-frequency_penalty: 0
-presence_penalty: 0
-max_tokens: 32000
-stream: true
-ngl: 32
-ctx_len: 32000
-engine: llama-cpp
-prompt_template: "{system_message} [INST] {prompt} [/INST]"
+# BEGIN GENERAL GGUF METADATA
+"id": "codestral"
+"model": "codestral"
+"name": "codestral"
+"version": 1
+
+# END GENERAL GGUF METADATA
+
+# BEGIN INFERENCE PARAMETERS
+# BEGIN REQUIRED
+"stop":
+- "[/INST]"
+# END REQUIRED
+
+# BEGIN OPTIONAL
+"dynatemp_exponent": 1
+"dynatemp_range": 0
+"frequency_penalty": 0
+"ignore_eos": false
+"max_tokens": 4096
+"min_keep": 0
+"min_p": 0.05
+"mirostat": false
+"mirostat_eta": 0.100000001
+"mirostat_tau": 5
+"n_probs": 0
+"penalize_nl": false
+"presence_penalty": 0
+"repeat_last_n": 64
+"repeat_penalty": 1
+"seed": -1
+"stream": true
+"temperature": 0.7
+"tfs_z": 1
+"top_k": 40
+"top_p": 0.9
+"typ_p": 1
+# END OPTIONAL
+# END INFERENCE PARAMETERS
+
+# BEGIN MODEL LOAD PARAMETERS
+# BEGIN REQUIRED
+"ctx_len": 4096
+"engine": "llama-cpp"
+"ngl": 34
+"prompt_template": "[INST] {system_message}\\n\\n{prompt}[/INST]"
+# END OPTIONAL
+# END MODEL LOAD PARAMETERS