daryl149 commited on
Commit
f93e847
1 Parent(s): d173210

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +5 -11
README.md CHANGED
@@ -47,6 +47,7 @@ You are a helpful, respectful and honest assistant. Always answer as helpfully a
47
 
48
  {prompt} [/INST] {model_reply} [INST] {prompt} [/INST]
49
  ```
 
50
 
51
  ### Example usage
52
  An example question you can ask:
@@ -89,7 +90,6 @@ Don't be evil.
89
 
90
  <!-- This section is meant to convey both technical and sociotechnical limitations. -->
91
  - It's not quite perfect Dutch yet, but a very promising start.
92
- - The model rarely generates EOS tokens and goes on a ramble. A quick and dirty way to mitigate this is to system prompt it to generate an `</s>` after 5 sentences.
93
 
94
  ### Recommendations
95
 
@@ -98,7 +98,8 @@ Don't be evil.
98
 
99
  ## How to Get Started with the Model
100
 
101
- If you already have a pipeline running llama 2 7B chat in huggingface format, just call this one instead.
 
102
 
103
  ## Training Details
104
 
@@ -124,6 +125,7 @@ If you already have a pipeline running llama 2 7B chat in huggingface format, ju
124
  #### Speeds, Sizes, Times [optional]
125
 
126
  <!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
 
127
 
128
  [More Information Needed]
129
 
@@ -169,15 +171,7 @@ If you already have a pipeline running llama 2 7B chat in huggingface format, ju
169
 
170
  <!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
171
 
172
- Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
173
-
174
- - **Hardware Type:** [More Information Needed]
175
- - **Hours used:** [More Information Needed]
176
- - **Cloud Provider:** [More Information Needed]
177
- - **Compute Region:** [More Information Needed]
178
- - **Carbon Emitted:** [More Information Needed]
179
-
180
- ## Technical Specifications [optional]
181
 
182
  ### Model Architecture and Objective
183
 
 
47
 
48
  {prompt} [/INST] {model_reply} [INST] {prompt} [/INST]
49
  ```
50
+ **N.B.** Set your pad_token_id=18610 in your generator, otherwise it returns gibberish.
51
 
52
  ### Example usage
53
  An example question you can ask:
 
90
 
91
  <!-- This section is meant to convey both technical and sociotechnical limitations. -->
92
  - It's not quite perfect Dutch yet, but a very promising start.
 
93
 
94
  ### Recommendations
95
 
 
98
 
99
  ## How to Get Started with the Model
100
 
101
+ If you already have a pipeline running llama 2 7B chat in huggingface format, just call this one instead.
102
+ **N.B.** Set pad_token_id=18610 in your generator, otherwise it returns gibberish.
103
 
104
  ## Training Details
105
 
 
125
  #### Speeds, Sizes, Times [optional]
126
 
127
  <!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
128
+ We reached 32 tokens/second on a V100S without trying anything fancy.
129
 
130
  [More Information Needed]
131
 
 
171
 
172
  <!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
173
 
174
+ Yes.
 
 
 
 
 
 
 
 
175
 
176
  ### Model Architecture and Objective
177