Crystalcareai commited on
Commit
01bac5f
1 Parent(s): b0ba53b

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +89 -68
README.md CHANGED
@@ -1,105 +1,126 @@
1
  ---
2
- base_model: arcee-ai/Arcee-Scribe
3
  license: apache-2.0
4
- pipeline_tag: text-generation
5
- quantized_by: bartowski
6
  ---
7
 
8
- ## Llamacpp imatrix Quantizations of Arcee-Scribe
 
 
9
 
10
- Using <a href="https://github.com/ggerganov/llama.cpp/">llama.cpp</a> release <a href="https://github.com/ggerganov/llama.cpp/releases/tag/b3389">b3389</a> for quantization.
11
 
12
- Original model: https://huggingface.co/arcee-ai/Arcee-Scribe
13
 
14
- All quants made using imatrix option with dataset from [here](https://gist.github.com/bartowski1182/eb213dccb3571f863da82e99418f81e8)
15
 
16
- ## Prompt format
17
 
18
- ```
19
- <s><|im_start|>system
20
- {system_prompt}<|im_end|>
21
- <|im_start|>user
22
- {prompt}<|im_end|>
23
- <|im_start|>assistant
24
 
25
- ```
26
 
27
- ## Download a file (not the whole branch) from below:
28
 
29
- | Filename | Quant type | File Size | Split | Description |
30
- | -------- | ---------- | --------- | ----- | ----------- |
31
- | [Arcee-Scribe-f32.gguf](https://huggingface.co/bartowski/Arcee-Scribe-GGUF/blob/main/Arcee-Scribe-f32.gguf) | f32 | 30.95GB | false | Full F32 weights. |
32
- | [Arcee-Scribe-Q8_0.gguf](https://huggingface.co/bartowski/Arcee-Scribe-GGUF/blob/main/Arcee-Scribe-Q8_0.gguf) | Q8_0 | 8.22GB | false | Extremely high quality, generally unneeded but max available quant. |
33
- | [Arcee-Scribe-Q6_K_L.gguf](https://huggingface.co/bartowski/Arcee-Scribe-GGUF/blob/main/Arcee-Scribe-Q6_K_L.gguf) | Q6_K_L | 6.53GB | false | Uses Q8_0 for embed and output weights. Very high quality, near perfect, *recommended*. |
34
- | [Arcee-Scribe-Q6_K.gguf](https://huggingface.co/bartowski/Arcee-Scribe-GGUF/blob/main/Arcee-Scribe-Q6_K.gguf) | Q6_K | 6.35GB | false | Very high quality, near perfect, *recommended*. |
35
- | [Arcee-Scribe-Q5_K_L.gguf](https://huggingface.co/bartowski/Arcee-Scribe-GGUF/blob/main/Arcee-Scribe-Q5_K_L.gguf) | Q5_K_L | 5.74GB | false | Uses Q8_0 for embed and output weights. High quality, *recommended*. |
36
- | [Arcee-Scribe-Q5_K_M.gguf](https://huggingface.co/bartowski/Arcee-Scribe-GGUF/blob/main/Arcee-Scribe-Q5_K_M.gguf) | Q5_K_M | 5.51GB | false | High quality, *recommended*. |
37
- | [Arcee-Scribe-Q5_K_S.gguf](https://huggingface.co/bartowski/Arcee-Scribe-GGUF/blob/main/Arcee-Scribe-Q5_K_S.gguf) | Q5_K_S | 5.37GB | false | High quality, *recommended*. |
38
- | [Arcee-Scribe-Q4_K_L.gguf](https://huggingface.co/bartowski/Arcee-Scribe-GGUF/blob/main/Arcee-Scribe-Q4_K_L.gguf) | Q4_K_L | 4.99GB | false | Uses Q8_0 for embed and output weights. Good quality, *recommended*. |
39
- | [Arcee-Scribe-Q4_K_M.gguf](https://huggingface.co/bartowski/Arcee-Scribe-GGUF/blob/main/Arcee-Scribe-Q4_K_M.gguf) | Q4_K_M | 4.71GB | false | Good quality, default size for must use cases, *recommended*. |
40
- | [Arcee-Scribe-Q4_K_S.gguf](https://huggingface.co/bartowski/Arcee-Scribe-GGUF/blob/main/Arcee-Scribe-Q4_K_S.gguf) | Q4_K_S | 4.48GB | false | Slightly lower quality with more space savings, *recommended*. |
41
- | [Arcee-Scribe-Q3_K_XL.gguf](https://huggingface.co/bartowski/Arcee-Scribe-GGUF/blob/main/Arcee-Scribe-Q3_K_XL.gguf) | Q3_K_XL | 4.47GB | false | Uses Q8_0 for embed and output weights. Lower quality but usable, good for low RAM availability. |
42
- | [Arcee-Scribe-IQ4_XS.gguf](https://huggingface.co/bartowski/Arcee-Scribe-GGUF/blob/main/Arcee-Scribe-IQ4_XS.gguf) | IQ4_XS | 4.24GB | false | Decent quality, smaller than Q4_K_S with similar performance, *recommended*. |
43
- | [Arcee-Scribe-Q3_K_L.gguf](https://huggingface.co/bartowski/Arcee-Scribe-GGUF/blob/main/Arcee-Scribe-Q3_K_L.gguf) | Q3_K_L | 4.13GB | false | Lower quality but usable, good for low RAM availability. |
44
- | [Arcee-Scribe-Q3_K_M.gguf](https://huggingface.co/bartowski/Arcee-Scribe-GGUF/blob/main/Arcee-Scribe-Q3_K_M.gguf) | Q3_K_M | 3.83GB | false | Low quality. |
45
- | [Arcee-Scribe-IQ3_M.gguf](https://huggingface.co/bartowski/Arcee-Scribe-GGUF/blob/main/Arcee-Scribe-IQ3_M.gguf) | IQ3_M | 3.60GB | false | Medium-low quality, new method with decent performance comparable to Q3_K_M. |
46
- | [Arcee-Scribe-Q3_K_S.gguf](https://huggingface.co/bartowski/Arcee-Scribe-GGUF/blob/main/Arcee-Scribe-Q3_K_S.gguf) | Q3_K_S | 3.48GB | false | Low quality, not recommended. |
47
- | [Arcee-Scribe-Q2_K_L.gguf](https://huggingface.co/bartowski/Arcee-Scribe-GGUF/blob/main/Arcee-Scribe-Q2_K_L.gguf) | Q2_K_L | 3.38GB | false | Uses Q8_0 for embed and output weights. Very low quality but surprisingly usable. |
48
- | [Arcee-Scribe-IQ3_XS.gguf](https://huggingface.co/bartowski/Arcee-Scribe-GGUF/blob/main/Arcee-Scribe-IQ3_XS.gguf) | IQ3_XS | 3.33GB | false | Lower quality, new method with decent performance, slightly better than Q3_K_S. |
49
- | [Arcee-Scribe-Q2_K.gguf](https://huggingface.co/bartowski/Arcee-Scribe-GGUF/blob/main/Arcee-Scribe-Q2_K.gguf) | Q2_K | 3.01GB | false | Very low quality but surprisingly usable. |
50
- | [Arcee-Scribe-IQ2_M.gguf](https://huggingface.co/bartowski/Arcee-Scribe-GGUF/blob/main/Arcee-Scribe-IQ2_M.gguf) | IQ2_M | 2.78GB | false | Relatively low quality, uses SOTA techniques to be surprisingly usable. |
 
 
 
 
 
 
 
 
 
 
 
 
51
 
52
- ## Credits
53
 
54
- Thank you kalomaze and Dampf for assistance in creating the imatrix calibration dataset
 
 
55
 
56
- Thank you ZeroWw for the inspiration to experiment with embed/output
57
 
58
- ## Downloading using huggingface-cli
59
 
60
- First, make sure you have hugginface-cli installed:
61
 
62
- ```
63
- pip install -U "huggingface_hub[cli]"
64
- ```
65
 
66
- Then, you can target the specific file you want:
67
 
68
- ```
69
- huggingface-cli download bartowski/Arcee-Scribe-GGUF --include "Arcee-Scribe-Q4_K_M.gguf" --local-dir ./
70
- ```
71
 
72
- If the model is bigger than 50GB, it will have been split into multiple files. In order to download them all to a local folder, run:
73
 
74
- ```
75
- huggingface-cli download bartowski/Arcee-Scribe-GGUF --include "Arcee-Scribe-Q8_0.gguf/*" --local-dir Arcee-Scribe-Q8_0
76
- ```
77
 
78
- You can either specify a new local-dir (Arcee-Scribe-Q8_0) or download them all in place (./)
79
 
80
- ## Which file should I choose?
81
 
82
- A great write up with charts showing various performances is provided by Artefact2 [here](https://gist.github.com/Artefact2/b5f810600771265fc1e39442288e8ec9)
83
 
84
- The first thing to figure out is how big a model you can run. To do this, you'll need to figure out how much RAM and/or VRAM you have.
85
 
86
- If you want your model running as FAST as possible, you'll want to fit the whole thing on your GPU's VRAM. Aim for a quant with a file size 1-2GB smaller than your GPU's total VRAM.
 
87
 
88
- If you want the absolute maximum quality, add both your system RAM and your GPU's VRAM together, then similarly grab a quant with a file size 1-2GB Smaller than that total.
89
 
90
- Next, you'll need to decide if you want to use an 'I-quant' or a 'K-quant'.
91
 
92
- If you don't want to think too much, grab one of the K-quants. These are in format 'QX_K_X', like Q5_K_M.
 
 
 
 
93
 
94
- If you want to get more into the weeds, you can check out this extremely useful feature chart:
95
 
96
- [llama.cpp feature matrix](https://github.com/ggerganov/llama.cpp/wiki/Feature-matrix)
 
 
97
 
98
- But basically, if you're aiming for below Q4, and you're running cuBLAS (Nvidia) or rocBLAS (AMD), you should look towards the I-quants. These are in format IQX_X, like IQ3_M. These are newer and offer better performance for their size.
 
 
99
 
100
- These I-quants can also be used on CPU and Apple Metal, but will be slower than their K-quant equivalent, so speed vs performance is a tradeoff you'll have to decide.
 
 
101
 
102
- The I-quants are *not* compatible with Vulcan, which is also AMD, so if you have an AMD card double check if you're using the rocBLAS build or the Vulcan build. At the time of writing this, LM Studio has a preview with ROCm support, and other inference engines have specific builds for ROCm.
103
 
104
- Want to support my work? Visit my ko-fi page here: https://ko-fi.com/bartowski
105
 
 
 
 
 
 
 
 
 
1
  ---
 
2
  license: apache-2.0
 
 
3
  ---
4
 
5
+ <div align="center">
6
+ <img src="https://i.ibb.co/HpgNssZ/arcee-scribe.webp" alt="Arcee Scribe" style="border-radius: 10px; box-shadow: 0 4px 8px 0 rgba(0, 0, 0, 0.2), 0 6px 20px 0 rgba(0, 0, 0, 0.19); max-width: 100%; height: auto;">
7
+ </div>
8
 
 
9
 
10
+ ## Overview
11
 
12
+ Arcee-Scribe is a versatile chat model that excels across a wide range of benchmarks. While it didn't quite meet our exacting standards for Arcee-Spark V2, we found it too delightful to not release. It's our strongest model to date in terms of reasoning on the Big-Bench Hard (BBH) dataset, and we discovered it to be particularly adept at creative writing tasks. The model is a merge of InternLM-2.5-chat with a custom InternLM finetune, incorporating both general and writing-specific data.
13
 
 
14
 
15
+ ## Capabilities and Use Cases
 
 
 
 
 
16
 
17
+ Arcee-Scribe functions as a writing assistant and can handle a variety of tasks. Here are some examples of its capabilities:
18
 
19
+ 1. **Challenging Dialogue**: Crafting nuanced conversations with subtext and complex emotions.
20
 
21
+ Example:
22
+ ```markdown
23
+ Character A (an optimist who is usually the first one to bring up solutions, even in difficult situations):
24
+
25
+ "Maybe it's time we should figure this out, y'know? Like, how we both really feel about what we want from our future. It's been... quite a ride, hasn't it?"
26
+
27
+ Character B (a reserved person who prefers to let things unfold, but is deeply affected by what's said):
28
+
29
+ "Yeah, it's... it's been interesting. You've got such a way of making everything sound like it's gonna be all right. But I... I'm not sure it's that simple."
30
+
31
+ Character A (trying to see the lighter side, attempting to lighten the mood without making things awkward):
32
+
33
+ "You mean, I'm like a magician's rabbit? Poof, and everything's gonna be just fine? But hey, I like the sound of that. We could use a bit of magic, right?"
34
+
35
+ Character B (acknowledging the tension, but avoiding the conversation directly):
36
+
37
+ "I think what you mean is that I'm trying to see the silver lining in every cloud. And you're right, it's... it's a bit like that."
38
+
39
+ Character A (feeling the need to reiterate, hoping to gently steer the conversation toward a positive direction):
40
+
41
+ "Well, I mean, we both know it's been challenging. But what I'm trying to say is, maybe we could find a way to make things better. Because, y'know, that's what we've always done. We've always found a way to navigate through the rough waters, haven't we?"
42
+
43
+ Character B (feeling the weight of their unspoken emotions, hinting at a resignation to their fate):
44
+
45
+ "That's... that's true. But sometimes, no matter how hard we try, we can't make the wind change. Maybe it's time to look at what's right in front of us, instead of always looking towards what's beyond the horizon."
46
+
47
+ Character A (realizing the gravity of the unspoken words, attempting to bridge the gap without acknowledging the breakup directly):
48
+
49
+ "You know, life's been a series of chapters, and I think we've had our fair share of adventures. Maybe it's time for us to start writing a new story—one where we find a way to walk side by side, even if it's on different paths."
50
+
51
+ Character B (acknowledging the poignant conversation, but still avoiding the explicit mention of a breakup):
52
+
53
+ "Yeah, chapters and stories. You always have such a way with words, making everything sound so... so beautiful. But sometimes, the most beautiful stories are the ones we write by ourselves."
54
+ ```
55
 
56
+ 2. **Integrating Technical Concepts**: Explaining complex ideas using creative analogies.
57
 
58
+ Example:
59
+ ```markdown
60
+ In the quaint heart of a bustling city, there was a bakery renowned not only for its delectable pastries but for its mysterious, enigmatic nature. The bakery, run by the eccentric Mr. Quantum, was a place where the ordinary merged with the extraordinary in a dance of flours and creams.
61
 
62
+ In this bakery, there were two magical eggs, one white and one blue. These eggs were intertwined in a bond so strong and peculiar that when one was cracked open, the contents inside mirrored the other's contents in an instant. Just as a baker might watch with wonder as dough miraculously doubled in size, the bakers at the Quantum Bakery would stand in awe as the contents of these eggs shared the same fate, no matter the distance separating them.
63
 
64
+ These eggs represented quantum entanglement, a phenomenon where two or more particles become interconnected in such a way that the state of one instantly influences the state of the other, no matter how far apart they may be. Just as the contents of the magical eggs were indistinguishably alike, the particles in a quantum entanglement remain intertwined, their states dependent on each other's, transcending the limitations of space and time.
65
 
66
+ The Quantum Bakery was a place where bakers and patrons alike could witness this extraordinary connection firsthand. It was a testament to the boundless possibilities of the quantum world, where the unseen forces at work could create wonders that defied the logic of the everyday.
67
 
68
+ Through the metaphor of the magical eggs, we see that quantum entanglement is not merely a scientific concept, but a dance of particles that defies the conventions of the physical world, much like the artistry of baking at the Quantum Bakery. It is a reminder that the universe, with its infinite mysteries, continues to astound and inspire, challenging us to look beyond the surface and discover the marvels that lie within.
69
+ ```
 
70
 
71
+ 3. **Complex Narrative Structures**: Creating stories with non-linear narratives or unique perspectives.
72
 
73
+ Example:
74
+ ```markdown
75
+ Arcee-Scribe: "The Unraveling"
76
 
77
+ The old man closed his eyes, a peaceful smile on his face. The tattered photograph slipped from his fingers.
78
 
79
+ Moments earlier, he had gazed at the image - a young couple, laughing by a lakeside. His wrinkled hand traced her face.
 
 
80
 
81
+ That morning, he'd found the photo while clearing out his attic, buried in a dusty box he hadn't opened in decades.
82
 
83
+ Fifty years before, he'd shoved that box into the attic's darkest corner, determined to forget.
84
 
85
+ Fifty years and one day earlier, he'd received the telegram. "We regret to inform you..."
86
 
87
+ A week before that, she'd waved goodbye, boarding a train in her crisp nurse's uniform. "I'll be back before you know it," she'd said.
88
 
89
+ And at the very beginning, on a sunny day by the lake, a young man asked a question. She said yes, and they laughed, unaware of the camera capturing their joy.
90
+ ```
91
 
92
+ ## Business Applications
93
 
94
+ Arcee-Scribe can be applied to various business tasks:
95
 
96
+ - **Content Creation**: Develop blog posts, articles, and social media content.
97
+ - **Product Descriptions**: Generate clear and engaging product descriptions.
98
+ - **Customer Communication**: Draft responses to customer inquiries or complaints.
99
+ - **Training Materials**: Create scenarios for employee training programs.
100
+ - **Brainstorming**: Generate ideas for new products or marketing campaigns.
101
 
102
+ ## Evaluations
103
 
104
+ <div align="center">
105
+ <img src="https://i.ibb.co/7Gk2Q9B/Screenshot-2024-07-14-at-10-18-11-AM.png" alt="Arcee Scribe" style="border-radius: 10px; box-shadow: 0 4px 8px 0 rgba(0, 0, 0, 0.2), 0 6px 20px 0 rgba(0, 0, 0, 0.19); max-width: 100%; height: auto;">
106
+ </div>
107
 
108
+ <div align="center">
109
+ <img src="https://i.ibb.co/7GnHg33/Screenshot-2024-07-14-at-10-13-28-AM.png" alt="Arcee Scribe" style="border-radius: 10px; box-shadow: 0 4px 8px 0 rgba(0, 0, 0, 0.2), 0 6px 20px 0 rgba(0, 0, 0, 0.19); max-width: 100%; height: auto;">
110
+ </div>
111
 
112
+ <div align="center">
113
+ <img src="https://i.ibb.co/sFQt7L3/Screenshot-2024-07-14-at-12-20-11-PM.png" alt="Arcee Scribe" style="border-radius: 10px; box-shadow: 0 4px 8px 0 rgba(0, 0, 0, 0.2), 0 6px 20px 0 rgba(0, 0, 0, 0.19); max-width: 100%; height: auto;">
114
+ </div>
115
 
116
+ ## Acknowledgments
117
 
118
+ We owe a great deal of credit to the wonderful pre/post-training work done by the Internlm team. Their incredible work has enabled us to deliver models of this caliber. If you wish to use Arcee-Scribe for commercial purposes, InternLM allows free commercial usage - simply fill out this form: https://wj.qq.com/s2/12727483/5dba/
119
 
120
+ ## Future Directions
121
+
122
+ We look forward to seeing how users will utilize Arcee-Scribe in their creative and professional endeavors. The model aims to assist businesses in their content creation processes and provide a tool for exploring new ideas in writing and communication.
123
+
124
+ ---
125
+
126
+ *Note: This README was written in large part by Arcee-Scribe.*