Update README.md
Browse files
README.md
CHANGED
@@ -1,22 +1,84 @@
|
|
1 |
---
|
2 |
-
base_model:
|
|
|
3 |
tags:
|
4 |
- text-generation-inference
|
5 |
- transformers
|
6 |
- unsloth
|
7 |
- llama
|
8 |
- gguf
|
|
|
|
|
|
|
|
|
9 |
license: apache-2.0
|
10 |
language:
|
11 |
- en
|
|
|
|
|
|
|
12 |
---
|
13 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
14 |
# Uploaded model
|
15 |
|
16 |
- **Developed by:** Pinkstack
|
17 |
- **License:** apache-2.0
|
18 |
-
- **Finetuned from model :**
|
19 |
-
|
20 |
-
This llama model was trained 2x faster with [Unsloth](https://github.com/unslothai/unsloth) and Huggingface's TRL library.
|
21 |
-
|
22 |
-
[<img src="https://raw.githubusercontent.com/unslothai/unsloth/main/images/unsloth%20made%20with%20love.png" width="200"/>](https://github.com/unslothai/unsloth)
|
|
|
1 |
---
|
2 |
+
base_model:
|
3 |
+
- Pinkstack/Superthoughts-lite-v1
|
4 |
tags:
|
5 |
- text-generation-inference
|
6 |
- transformers
|
7 |
- unsloth
|
8 |
- llama
|
9 |
- gguf
|
10 |
+
- cot
|
11 |
+
- superthoughts
|
12 |
+
- reasoning
|
13 |
+
- grpo
|
14 |
license: apache-2.0
|
15 |
language:
|
16 |
- en
|
17 |
+
datasets:
|
18 |
+
- openai/gsm8k
|
19 |
+
- Pinkstack/intructions-sft-sharegpt
|
20 |
---
|
21 |
|
22 |
+

|
23 |
+
|
24 |
+
# Information
|
25 |
+
Advanced, high-quality and lite reasoning for a tiny size that you can run on your phone.
|
26 |
+
|
27 |
+
Trained similarly to Deepseek R1, we used Smollm2 as a base model, then we've SFT fine tuned in on reasoning & modified the tokenizer slightly, after the SFT fine tuning we used GRPO to further amplify it's mathematics & problem solving abilities.
|
28 |
+
|
29 |
+
# Which quant is right for you?
|
30 |
+
|
31 |
+
***F16***: Least hallucinations, high-quality reasoning yet heavy to run.
|
32 |
+
***Q8_0***: Limited amount of hallucinations high-quality reasoning, recommended
|
33 |
+
***Q6_k***: Hallucinates more, good reasoning but may fail at counting etc. only use if you cannot run Q8_0.
|
34 |
+
***Q4_k_m***: Not recommended, Hallucinates, doesn't always think properly. easier to run though.
|
35 |
+
|
36 |
+
# Format
|
37 |
+
```
|
38 |
+
<|im_start|>user
|
39 |
+
How many R's in strawberry<|im_end|>
|
40 |
+
<|im_start|>assistant
|
41 |
+
<think>
|
42 |
+
Alright, the user has asked how many R's in the word strawberry, that's easy! I just need to count each instance of the letter 'R' in the word 's-t-r-a-w-b-e-r-r-y' and then find out how many R's there are, lets count!
|
43 |
+
S - Not an R,
|
44 |
+
T - Not an R,
|
45 |
+
R - First instance of the letter R! (1),
|
46 |
+
A - Not an R,
|
47 |
+
W - Not an R,
|
48 |
+
B - Not an R,
|
49 |
+
E - Not an R,
|
50 |
+
R - Great! Second instance of the letter R. (2),
|
51 |
+
R - Third instance of the letter R. (3),
|
52 |
+
Y - Not an R.
|
53 |
+
|
54 |
+
So, i've counted all the letters correctly, meaning that I am sure that there are 3 R's in the word Strawberry. I should probably let the user know.
|
55 |
+
</think>
|
56 |
+
<output>3
|
57 |
+
</output><|im_end|>
|
58 |
+
```
|
59 |
+
# system prompt
|
60 |
+
(important to ensure it would always think, output).
|
61 |
+
```
|
62 |
+
respond in the following format:
|
63 |
+
<think>
|
64 |
+
...
|
65 |
+
</think>
|
66 |
+
<output>
|
67 |
+
...
|
68 |
+
</output>
|
69 |
+
```
|
70 |
+
# Examples:
|
71 |
+
all responses below generated with our system prompt and a temperature of 0.7.
|
72 |
+
Generated inside the android application, ChatterUI via GGUF Q8, using the model's prompt format. and our
|
73 |
+
1)
|
74 |
+

|
75 |
+
2)
|
76 |
+

|
77 |
+
3)
|
78 |
+

|
79 |
+
|
80 |
# Uploaded model
|
81 |
|
82 |
- **Developed by:** Pinkstack
|
83 |
- **License:** apache-2.0
|
84 |
+
- **Finetuned from model :** HuggingFaceTB/SmolLM2-1.7B-Instruct
|
|
|
|
|
|
|
|