mlabonne commited on
Commit
0d42a2e
1 Parent(s): b4a443f

Create README.md

Browse files
Files changed (1) hide show
  1. README.md +41 -0
README.md ADDED
@@ -0,0 +1,41 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: transformers
3
+ tags: []
4
+ extra_gated_heading: "Access Gemma on Hugging Face"
5
+ extra_gated_prompt: "To access Gemma on Hugging Face, you’re required to review and agree to Google’s usage license. To do this, please ensure you’re logged-in to Hugging Face and click below. Requests are processed immediately."
6
+ extra_gated_button_content: "Acknowledge license"
7
+ license: other
8
+ license_name: gemma-terms-of-use
9
+ license_link: https://ai.google.dev/gemma/terms
10
+ ---
11
+
12
+ # Gemma-2B GGUF
13
+
14
+ This is a quantized version of the [google/gemma-2b](https://huggingface.co/google/gemma-2b) model using [llama.cpp](https://github.com/ggerganov/llama.cpp).
15
+
16
+ This model card corresponds to the 2B base version of the Gemma model. You can also visit the model card of the [7B base model](https://huggingface.co/google/gemma-7b), [7B instruct model](https://huggingface.co/google/gemma-7b-it), and [2B instruct model](https://huggingface.co/google/gemma-2b-it).
17
+
18
+ **Model Page**: [Gemma](https://ai.google.dev/gemma/docs)
19
+
20
+ **Terms of Use**: [Terms](https://www.kaggle.com/models/google/gemma/license/consent)
21
+
22
+ ## ⚡ Quants
23
+
24
+ * `q2_k`: Uses Q4_K for the attention.vw and feed_forward.w2 tensors, Q2_K for the other tensors.
25
+ * `q3_k_l`: Uses Q5_K for the attention.wv, attention.wo, and feed_forward.w2 tensors, else Q3_K
26
+ * `q3_k_m`: Uses Q4_K for the attention.wv, attention.wo, and feed_forward.w2 tensors, else Q3_K
27
+ * `q3_k_s`: Uses Q3_K for all tensors
28
+ * `q4_0`: Original quant method, 4-bit.
29
+ * `q4_1`: Higher accuracy than q4_0 but not as high as q5_0. However has quicker inference than q5 models.
30
+ * `q4_k_m`: Uses Q6_K for half of the attention.wv and feed_forward.w2 tensors, else Q4_K
31
+ * `q4_k_s`: Uses Q4_K for all tensors
32
+ * `q5_0`: Higher accuracy, higher resource usage and slower inference.
33
+ * `q5_1`: Even higher accuracy, resource usage and slower inference.
34
+ * `q5_k_m`: Uses Q6_K for half of the attention.wv and feed_forward.w2 tensors, else Q5_K
35
+ * `q5_k_s`: Uses Q5_K for all tensors
36
+ * `q6_k`: Uses Q8_K for all tensors
37
+ * `q8_0`: Almost indistinguishable from float16. High resource use and slow. Not recommended for most users.
38
+
39
+ ## 💻 Usage
40
+
41
+ This model can be used with the latest version of llama.cpp but is not necessarily compatible with other tools yet.