Text Generation
GGUF
English
ggml
quantized
q2_k
q3_k_m
q4_k_m
q5_k_m
q6_k
q8_0
afrideva commited on
Commit
4bcbc66
1 Parent(s): 6e8333b

Upload README.md with huggingface_hub

Browse files
Files changed (1) hide show
  1. README.md +143 -0
README.md ADDED
@@ -0,0 +1,143 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ base_model: Felladrin/Llama-68M-Chat-v1
3
+ datasets:
4
+ - THUDM/webglm-qa
5
+ - databricks/databricks-dolly-15k
6
+ - cognitivecomputations/wizard_vicuna_70k_unfiltered
7
+ - totally-not-an-llm/EverythingLM-data-V3
8
+ - Amod/mental_health_counseling_conversations
9
+ - sablo/oasst2_curated
10
+ - starfishmedical/webGPT_x_dolly
11
+ - Open-Orca/OpenOrca
12
+ - mlabonne/chatml_dpo_pairs
13
+ inference: false
14
+ language:
15
+ - en
16
+ license: apache-2.0
17
+ model_creator: Felladrin
18
+ model_name: Llama-68M-Chat-v1
19
+ pipeline_tag: text-generation
20
+ quantized_by: afrideva
21
+ tags:
22
+ - text-generation
23
+ - gguf
24
+ - ggml
25
+ - quantized
26
+ - q2_k
27
+ - q3_k_m
28
+ - q4_k_m
29
+ - q5_k_m
30
+ - q6_k
31
+ - q8_0
32
+ widget:
33
+ - text: '<|im_start|>system
34
+
35
+ You are a knowledgeable assistant. Help the user as much as you can.<|im_end|>
36
+
37
+ <|im_start|>user
38
+
39
+ How to become healthier?<|im_end|>
40
+
41
+ <|im_start|>assistant'
42
+ - text: '<|im_start|>system
43
+
44
+ You are a career counselor. The user will provide you with an individual looking
45
+ for guidance in their professional life, and your task is to assist them in determining
46
+ what careers they are most suited for based on their skills, interests, and experience.
47
+ You should also conduct research into the various options available, explain the
48
+ job market trends in different industries, and advice on which qualifications
49
+ would be beneficial for pursuing particular fields.<|im_end|>
50
+
51
+ <|im_start|>user
52
+
53
+ Heya!<|im_end|>
54
+
55
+ <|im_start|>assistant
56
+
57
+ Hi! How may I help you?<|im_end|>
58
+
59
+ <|im_start|>user
60
+
61
+ I am interested in developing a career in software engineering. What would you
62
+ recommend me to do?<|im_end|>
63
+
64
+ <|im_start|>assistant'
65
+ - text: '<|im_start|>system
66
+
67
+ You are a helpful assistant who provides concise responses.<|im_end|>
68
+
69
+ <|im_start|>user
70
+
71
+ Hi!<|im_end|>
72
+
73
+ <|im_start|>assistant
74
+
75
+ Hello there! How may I help you?<|im_end|>
76
+
77
+ <|im_start|>user
78
+
79
+ I need to build a simple website. Where should I start learning about web development?<|im_end|>
80
+
81
+ <|im_start|>assistant'
82
+ - text: '<|im_start|>system
83
+
84
+ You are a very creative assistant. User will give you a task, which you should
85
+ complete with all your knowledge.<|im_end|>
86
+
87
+ <|im_start|>user
88
+
89
+ Write the background story of an RPG game about wizards and dragons in a sci-fi
90
+ world.<|im_end|>
91
+
92
+ <|im_start|>assistant'
93
+ ---
94
+ # Felladrin/Llama-68M-Chat-v1-GGUF
95
+
96
+ Quantized GGUF model files for [Llama-68M-Chat-v1](https://huggingface.co/Felladrin/Llama-68M-Chat-v1) from [Felladrin](https://huggingface.co/Felladrin)
97
+
98
+
99
+ | Name | Quant method | Size |
100
+ | ---- | ---- | ---- |
101
+ | [llama-68m-chat-v1.fp16.gguf](https://huggingface.co/afrideva/Llama-68M-Chat-v1-GGUF/resolve/main/llama-68m-chat-v1.fp16.gguf) | fp16 | 136.79 MB |
102
+ | [llama-68m-chat-v1.q2_k.gguf](https://huggingface.co/afrideva/Llama-68M-Chat-v1-GGUF/resolve/main/llama-68m-chat-v1.q2_k.gguf) | q2_k | 35.88 MB |
103
+ | [llama-68m-chat-v1.q3_k_m.gguf](https://huggingface.co/afrideva/Llama-68M-Chat-v1-GGUF/resolve/main/llama-68m-chat-v1.q3_k_m.gguf) | q3_k_m | 40.66 MB |
104
+ | [llama-68m-chat-v1.q4_k_m.gguf](https://huggingface.co/afrideva/Llama-68M-Chat-v1-GGUF/resolve/main/llama-68m-chat-v1.q4_k_m.gguf) | q4_k_m | 46.10 MB |
105
+ | [llama-68m-chat-v1.q5_k_m.gguf](https://huggingface.co/afrideva/Llama-68M-Chat-v1-GGUF/resolve/main/llama-68m-chat-v1.q5_k_m.gguf) | q5_k_m | 51.16 MB |
106
+ | [llama-68m-chat-v1.q6_k.gguf](https://huggingface.co/afrideva/Llama-68M-Chat-v1-GGUF/resolve/main/llama-68m-chat-v1.q6_k.gguf) | q6_k | 56.54 MB |
107
+ | [llama-68m-chat-v1.q8_0.gguf](https://huggingface.co/afrideva/Llama-68M-Chat-v1-GGUF/resolve/main/llama-68m-chat-v1.q8_0.gguf) | q8_0 | 73.02 MB |
108
+
109
+
110
+
111
+ ## Original Model Card:
112
+ # A Llama Chat Model of 68M Parameters
113
+
114
+ - Base model: [JackFram/llama-68m](https://huggingface.co/JackFram/llama-68m)
115
+ - Datasets:
116
+ - [THUDM/webglm-qa](https://huggingface.co/datasets/THUDM/webglm-qa)
117
+ - [databricks/databricks-dolly-15k](https://huggingface.co/datasets/databricks/databricks-dolly-15k)
118
+ - [cognitivecomputations/wizard_vicuna_70k_unfiltered](https://huggingface.co/datasets/cognitivecomputations/wizard_vicuna_70k_unfiltered)
119
+ - [totally-not-an-llm/EverythingLM-data-V3](https://huggingface.co/datasets/totally-not-an-llm/EverythingLM-data-V3)
120
+ - [Amod/mental_health_counseling_conversations](https://huggingface.co/datasets/Amod/mental_health_counseling_conversations)
121
+ - [sablo/oasst2_curated](https://huggingface.co/datasets/sablo/oasst2_curated)
122
+ - [starfishmedical/webGPT_x_dolly](https://huggingface.co/datasets/starfishmedical/webGPT_x_dolly)
123
+ - [Open-Orca/OpenOrca](https://huggingface.co/datasets/Open-Orca/OpenOrca)
124
+ - [mlabonne/chatml_dpo_pairs](https://huggingface.co/datasets/mlabonne/chatml_dpo_pairs)
125
+ - Availability in other ML formats:
126
+ - ONNX: [Felladrin/onnx-Llama-68M-Chat-v1](https://huggingface.co/Felladrin/onnx-Llama-68M-Chat-v1)
127
+
128
+ ## Recommended Prompt Format
129
+
130
+ ```
131
+ <|im_start|>system
132
+ {system_message}<|im_end|>
133
+ <|im_start|>user
134
+ {user_message}<|im_end|>
135
+ <|im_start|>assistant
136
+ ```
137
+
138
+ ## Recommended Inference Parameters
139
+
140
+ ```yml
141
+ penalty_alpha: 0.5
142
+ top_k: 4
143
+ ```