Hugging Face's logo

RichardErkhov
/
phanerozoic_-_Tiny-Knight-1.1b-v0.1-gguf

GGUF
Inference Endpoints
conversational
Model card Files Community
RichardErkhov commited on
Commit
8f7399d
•
1 Parent(s): 24bc5b0

uploaded readme

Browse files
Files changed (1) hide show
  1. README.md +105 -0
README.md ADDED
@@ -0,0 +1,105 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ Quantization made by Richard Erkhov.
2
+
3
+ [Github](https://github.com/RichardErkhov)
4
+
5
+ [Discord](https://discord.gg/pvy7H8DZMG)
6
+
7
+ [Request more models](https://github.com/RichardErkhov/quant_request)
8
+
9
+
10
+ Tiny-Knight-1.1b-v0.1 - GGUF
11
+ - Model creator: https://huggingface.co/phanerozoic/
12
+ - Original model: https://huggingface.co/phanerozoic/Tiny-Knight-1.1b-v0.1/
13
+
14
+
15
+ | Name | Quant method | Size |
16
+ | ---- | ---- | ---- |
17
+ | [Tiny-Knight-1.1b-v0.1.Q2_K.gguf](https://huggingface.co/RichardErkhov/phanerozoic_-_Tiny-Knight-1.1b-v0.1-gguf/blob/main/Tiny-Knight-1.1b-v0.1.Q2_K.gguf) | Q2_K | 0.4GB |
18
+ | [Tiny-Knight-1.1b-v0.1.IQ3_XS.gguf](https://huggingface.co/RichardErkhov/phanerozoic_-_Tiny-Knight-1.1b-v0.1-gguf/blob/main/Tiny-Knight-1.1b-v0.1.IQ3_XS.gguf) | IQ3_XS | 0.44GB |
19
+ | [Tiny-Knight-1.1b-v0.1.IQ3_S.gguf](https://huggingface.co/RichardErkhov/phanerozoic_-_Tiny-Knight-1.1b-v0.1-gguf/blob/main/Tiny-Knight-1.1b-v0.1.IQ3_S.gguf) | IQ3_S | 0.47GB |
20
+ | [Tiny-Knight-1.1b-v0.1.Q3_K_S.gguf](https://huggingface.co/RichardErkhov/phanerozoic_-_Tiny-Knight-1.1b-v0.1-gguf/blob/main/Tiny-Knight-1.1b-v0.1.Q3_K_S.gguf) | Q3_K_S | 0.47GB |
21
+ | [Tiny-Knight-1.1b-v0.1.IQ3_M.gguf](https://huggingface.co/RichardErkhov/phanerozoic_-_Tiny-Knight-1.1b-v0.1-gguf/blob/main/Tiny-Knight-1.1b-v0.1.IQ3_M.gguf) | IQ3_M | 0.48GB |
22
+ | [Tiny-Knight-1.1b-v0.1.Q3_K.gguf](https://huggingface.co/RichardErkhov/phanerozoic_-_Tiny-Knight-1.1b-v0.1-gguf/blob/main/Tiny-Knight-1.1b-v0.1.Q3_K.gguf) | Q3_K | 0.51GB |
23
+ | [Tiny-Knight-1.1b-v0.1.Q3_K_M.gguf](https://huggingface.co/RichardErkhov/phanerozoic_-_Tiny-Knight-1.1b-v0.1-gguf/blob/main/Tiny-Knight-1.1b-v0.1.Q3_K_M.gguf) | Q3_K_M | 0.51GB |
24
+ | [Tiny-Knight-1.1b-v0.1.Q3_K_L.gguf](https://huggingface.co/RichardErkhov/phanerozoic_-_Tiny-Knight-1.1b-v0.1-gguf/blob/main/Tiny-Knight-1.1b-v0.1.Q3_K_L.gguf) | Q3_K_L | 0.55GB |
25
+ | [Tiny-Knight-1.1b-v0.1.IQ4_XS.gguf](https://huggingface.co/RichardErkhov/phanerozoic_-_Tiny-Knight-1.1b-v0.1-gguf/blob/main/Tiny-Knight-1.1b-v0.1.IQ4_XS.gguf) | IQ4_XS | 0.57GB |
26
+ | [Tiny-Knight-1.1b-v0.1.Q4_0.gguf](https://huggingface.co/RichardErkhov/phanerozoic_-_Tiny-Knight-1.1b-v0.1-gguf/blob/main/Tiny-Knight-1.1b-v0.1.Q4_0.gguf) | Q4_0 | 0.59GB |
27
+ | [Tiny-Knight-1.1b-v0.1.IQ4_NL.gguf](https://huggingface.co/RichardErkhov/phanerozoic_-_Tiny-Knight-1.1b-v0.1-gguf/blob/main/Tiny-Knight-1.1b-v0.1.IQ4_NL.gguf) | IQ4_NL | 0.6GB |
28
+ | [Tiny-Knight-1.1b-v0.1.Q4_K_S.gguf](https://huggingface.co/RichardErkhov/phanerozoic_-_Tiny-Knight-1.1b-v0.1-gguf/blob/main/Tiny-Knight-1.1b-v0.1.Q4_K_S.gguf) | Q4_K_S | 0.6GB |
29
+ | [Tiny-Knight-1.1b-v0.1.Q4_K.gguf](https://huggingface.co/RichardErkhov/phanerozoic_-_Tiny-Knight-1.1b-v0.1-gguf/blob/main/Tiny-Knight-1.1b-v0.1.Q4_K.gguf) | Q4_K | 0.62GB |
30
+ | [Tiny-Knight-1.1b-v0.1.Q4_K_M.gguf](https://huggingface.co/RichardErkhov/phanerozoic_-_Tiny-Knight-1.1b-v0.1-gguf/blob/main/Tiny-Knight-1.1b-v0.1.Q4_K_M.gguf) | Q4_K_M | 0.62GB |
31
+ | [Tiny-Knight-1.1b-v0.1.Q4_1.gguf](https://huggingface.co/RichardErkhov/phanerozoic_-_Tiny-Knight-1.1b-v0.1-gguf/blob/main/Tiny-Knight-1.1b-v0.1.Q4_1.gguf) | Q4_1 | 0.65GB |
32
+ | [Tiny-Knight-1.1b-v0.1.Q5_0.gguf](https://huggingface.co/RichardErkhov/phanerozoic_-_Tiny-Knight-1.1b-v0.1-gguf/blob/main/Tiny-Knight-1.1b-v0.1.Q5_0.gguf) | Q5_0 | 0.71GB |
33
+ | [Tiny-Knight-1.1b-v0.1.Q5_K_S.gguf](https://huggingface.co/RichardErkhov/phanerozoic_-_Tiny-Knight-1.1b-v0.1-gguf/blob/main/Tiny-Knight-1.1b-v0.1.Q5_K_S.gguf) | Q5_K_S | 0.71GB |
34
+ | [Tiny-Knight-1.1b-v0.1.Q5_K.gguf](https://huggingface.co/RichardErkhov/phanerozoic_-_Tiny-Knight-1.1b-v0.1-gguf/blob/main/Tiny-Knight-1.1b-v0.1.Q5_K.gguf) | Q5_K | 0.73GB |
35
+ | [Tiny-Knight-1.1b-v0.1.Q5_K_M.gguf](https://huggingface.co/RichardErkhov/phanerozoic_-_Tiny-Knight-1.1b-v0.1-gguf/blob/main/Tiny-Knight-1.1b-v0.1.Q5_K_M.gguf) | Q5_K_M | 0.73GB |
36
+ | [Tiny-Knight-1.1b-v0.1.Q5_1.gguf](https://huggingface.co/RichardErkhov/phanerozoic_-_Tiny-Knight-1.1b-v0.1-gguf/blob/main/Tiny-Knight-1.1b-v0.1.Q5_1.gguf) | Q5_1 | 0.77GB |
37
+ | [Tiny-Knight-1.1b-v0.1.Q6_K.gguf](https://huggingface.co/RichardErkhov/phanerozoic_-_Tiny-Knight-1.1b-v0.1-gguf/blob/main/Tiny-Knight-1.1b-v0.1.Q6_K.gguf) | Q6_K | 0.84GB |
38
+ | [Tiny-Knight-1.1b-v0.1.Q8_0.gguf](https://huggingface.co/RichardErkhov/phanerozoic_-_Tiny-Knight-1.1b-v0.1-gguf/blob/main/Tiny-Knight-1.1b-v0.1.Q8_0.gguf) | Q8_0 | 1.09GB |
39
+
40
+
41
+
42
+
43
+ Original model description:
44
+ ---
45
+ license: cc-by-nc-4.0
46
+ language:
47
+ - en
48
+ widget:
49
+ - text: |
50
+ Hail and well met! Pray, what kind of food do ye enjoy supping upon?
51
+ example_title: "The Code of Chivalry"
52
+ ---
53
+ ![tinyknight.png](https://huggingface.co/phanerozoic/Tiny-Knight-1.1b-v0.1/resolve/bbfc665ce3af9d4d73e199b89144b30d668f50aa/tinyknight.png)
54
+ # Tiny Knight-1.1b-v0.1
55
+
56
+ Tiny Knight-1.1b-v0.1 is a specialized language model crafted for generating knight and medieval-themed content. This iteration is built upon the foundations of TinyLlama-1.1B-Chat-v1.0, tailored to operate within environments constrained by computing resources.
57
+
58
+ ### Performance
59
+ While this model excels in creating knight-themed narratives, its specialization, however, limits its effectiveness in broader language tasks, especially those requiring detailed knowledge outside the medieval theme.
60
+
61
+ ### Direct Use
62
+ Tiny Knight-1.1b-v0.1 is particularly suited for generating content within medieval, knightly, or fantasy settings, ideal for storytelling, educational content, and thematic exploration. It is not recommended for general-purpose tasks or technical domains.
63
+
64
+ ### Context Setting and Interaction Guidelines
65
+ Given its specialized nature, Tiny Knight-1.1b-v0.1 benefits significantly from detailed context-setting. Providing a rich thematic backdrop in prompts enhances the model's performance, guiding it to generate more accurate and immersive content.
66
+
67
+ ### Training Data
68
+ Incorporates a dataset focused on knightly tales, medieval history, and literature, derived from the foundational TinyLlama-1.1B model.
69
+
70
+ ### Custom Stopping Strings
71
+ Custom stopping strings were used to refine output quality:
72
+
73
+ - "},"
74
+ - "User:"
75
+ - "You:"
76
+ - "\nUser"
77
+ - "\nUser:"
78
+ - "me:"
79
+ - "user"
80
+ - "\n"
81
+
82
+ ### Training Hyperparameters and Fine-Tuning Details
83
+ - **Base Model Name**: TinyLlama-1.1B-Chat-v1.0
84
+ - **Base Model Class**: LlamaForCausalLM
85
+ - **Projections**: gate, down, up, q, k, v, o
86
+ - **LoRA Rank**: 16
87
+ - **LoRA Alpha**: 32
88
+ - **True Batch Size**: 32
89
+ - **Gradient Accumulation Steps**: 1
90
+ - **Epochs**: 0.18
91
+ - **Learning Rate**: 3e-4
92
+ - **LR Scheduler**: Linear
93
+ - **Step**: 75
94
+ - **Loss**: 1.87
95
+
96
+ ### Limitations
97
+ While adept at producing themed content, Tiny Knight-1.1b-v0.1's applicability is limited outside its specialized domain of knights and medieval themes.
98
+
99
+ ### Summary
100
+ Tiny Knight-1.1b-v0.1 represents a significant advancement in thematic language models, offering a specialized tool for exploring the medieval era. Its emphasis on context for optimal performance and the use of custom stopping strings make it a sophisticated asset for generating historically rich content.
101
+
102
+ ### Acknowledgments
103
+ Special thanks to the TinyLlama-1.1B team, whose pioneering work laid the groundwork for the creation of Tiny Knight-1.1b-v0.1.
104
+
105
+