Text Generation
Transformers
Safetensors
English
mistral
conversational
Eval Results
Inference Endpoints
text-generation-inference
Felladrin commited on
Commit
aba0753
1 Parent(s): 9722ba3

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +8 -11
README.md CHANGED
@@ -52,7 +52,7 @@ widget:
52
  - role: system
53
  content: You are a highly knowledgeable assistant. Help the user as much as you can.
54
  - role: user
55
- content: How I can become a healthier person?
56
  - messages:
57
  - role: system
58
  content: You are a helpful assistant who gives creative responses.
@@ -72,8 +72,8 @@ inference:
72
  parameters:
73
  max_new_tokens: 250
74
  do_sample: true
75
- temperature: 0.7
76
- top_p: 0.5
77
  top_k: 35
78
  repetition_penalty: 1.176
79
  ---
@@ -98,6 +98,7 @@ inference:
98
  - [[ChatML](https://huggingface.co/datasets/Felladrin/ChatML-ultrafeedback-binarized-preferences)] [argilla/ultrafeedback-binarized-preferences](https://huggingface.co/datasets/argilla/ultrafeedback-binarized-preferences)
99
  - [[ChatML](https://huggingface.co/datasets/Felladrin/ChatML-oasst2_dpo_pairs_en)] [sablo/oasst2_dpo_pairs_en](https://huggingface.co/datasets/sablo/oasst2_dpo_pairs_en)
100
  - [[ChatML](https://huggingface.co/datasets/Felladrin/ChatML-Neural-DPO)] [NeuralNovel/Neural-DPO](https://huggingface.co/datasets/NeuralNovel/Neural-DPO)
 
101
  - Availability in other ML formats:
102
  - GGUF: [Felladrin/gguf-Minueza-32M-Chat](https://huggingface.co/Felladrin/gguf-Minueza-32M-Chat)
103
  - ONNX: [Felladrin/onnx-Minueza-32M-Chat](https://huggingface.co/Felladrin/onnx-Minueza-32M-Chat)
@@ -116,8 +117,8 @@ inference:
116
 
117
  ```yml
118
  do_sample: true
119
- temperature: 0.7
120
- top_p: 0.5
121
  top_k: 35
122
  repetition_penalty: 1.176
123
  ```
@@ -146,15 +147,11 @@ output = generate(
146
  prompt,
147
  max_new_tokens=256,
148
  do_sample=True,
149
- temperature=0.7,
150
  top_k=35,
151
- top_p=0.5,
152
  repetition_penalty=1.176,
153
  )
154
 
155
  print(output[0]["generated_text"])
156
  ```
157
-
158
- ## License
159
-
160
- This model is licensed under the [Apache License 2.0](https://huggingface.co/Felladrin/Minueza-32M-Chat/resolve/main/license.txt).
 
52
  - role: system
53
  content: You are a highly knowledgeable assistant. Help the user as much as you can.
54
  - role: user
55
+ content: How can I become a healthier person?
56
  - messages:
57
  - role: system
58
  content: You are a helpful assistant who gives creative responses.
 
72
  parameters:
73
  max_new_tokens: 250
74
  do_sample: true
75
+ temperature: 0.65
76
+ top_p: 0.55
77
  top_k: 35
78
  repetition_penalty: 1.176
79
  ---
 
98
  - [[ChatML](https://huggingface.co/datasets/Felladrin/ChatML-ultrafeedback-binarized-preferences)] [argilla/ultrafeedback-binarized-preferences](https://huggingface.co/datasets/argilla/ultrafeedback-binarized-preferences)
99
  - [[ChatML](https://huggingface.co/datasets/Felladrin/ChatML-oasst2_dpo_pairs_en)] [sablo/oasst2_dpo_pairs_en](https://huggingface.co/datasets/sablo/oasst2_dpo_pairs_en)
100
  - [[ChatML](https://huggingface.co/datasets/Felladrin/ChatML-Neural-DPO)] [NeuralNovel/Neural-DPO](https://huggingface.co/datasets/NeuralNovel/Neural-DPO)
101
+ - License: [Apache License 2.0](https://huggingface.co/Felladrin/Minueza-32M-Chat/resolve/main/license.txt)
102
  - Availability in other ML formats:
103
  - GGUF: [Felladrin/gguf-Minueza-32M-Chat](https://huggingface.co/Felladrin/gguf-Minueza-32M-Chat)
104
  - ONNX: [Felladrin/onnx-Minueza-32M-Chat](https://huggingface.co/Felladrin/onnx-Minueza-32M-Chat)
 
117
 
118
  ```yml
119
  do_sample: true
120
+ temperature: 0.65
121
+ top_p: 0.55
122
  top_k: 35
123
  repetition_penalty: 1.176
124
  ```
 
147
  prompt,
148
  max_new_tokens=256,
149
  do_sample=True,
150
+ temperature=0.65,
151
  top_k=35,
152
+ top_p=0.55,
153
  repetition_penalty=1.176,
154
  )
155
 
156
  print(output[0]["generated_text"])
157
  ```