nunonmg commited on
Commit
53c9fd7
1 Parent(s): 76aea9a

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +6 -28
README.md CHANGED
@@ -21,7 +21,7 @@ pipeline_tag: translation
21
 
22
  ### Model Description
23
 
24
- TowerInstruct is a language model that results from fine-tuning TowerBase on the TowerBlocks supervised fine-tuning dataset. TowerInstruct v0.1 is the first model in the series.
25
  The model is trained to handle several translation-related tasks, such as general machine translation (e.g., sentence- and document-level translation, terminology-aware translation, context-aware translation), automatic post edition, named-entity recognition, gramatical error correction, and paraphrase generation.
26
  We will release more details in the upcoming technical report.
27
 
@@ -29,7 +29,7 @@ We will release more details in the upcoming technical report.
29
  - **Model type:** A 7B parameter model fine-tuned on a mix of publicly available, synthetic datasets on translation-related tasks, as well as conversational datasets and code instructions.
30
  - **Language(s) (NLP):** English, Portuguese, Spanish, French, German, Dutch, Italian, Korean, Chinese, Russian
31
  - **License:** CC-BY-NC-4.0
32
- - **Finetuned from model:** TowerBase
33
 
34
  ## Intended uses & limitations
35
 
@@ -45,7 +45,7 @@ The model was initially fine-tuned on a filtered and preprocessed supervised fin
45
  - Synthetic Chat data
46
  - Code instructions
47
 
48
- You can find the dataset and all data sources of TowerBlocks here.
49
 
50
  Here's how you can run the model using the `pipeline()` function from 🤗 Transformers:
51
 
@@ -95,37 +95,15 @@ TowerInstruct-v0.1 was trained using the ChatML prompt templates without any sys
95
 
96
  ### Supervised tasks
97
 
98
- - Translation
99
- ```
100
- Translate the following text from $SRC_LANG into $TGT_LANG.
101
- $SRC_LANG: $SRC_TEXT
102
- $TGT_LANG: # make sure to add a white space the target placeholder "$TGT_LANG:" for best results
103
- ```
104
- - Automatic Post Edition
105
- ```
106
- Translate the following text from $SRC_LANG into $TGT_LANG.
107
- $SRC_LANG: $SRC_TEXT
108
- $TGT_LANG:
109
- ```
110
- - Machine Translation Evaluation
111
- - Context-aware Translation
112
- - Terminology-aware Translation
113
- - Multi-reference Translation
114
- - Named-entity Recognition
115
- - Paraphrase Generation
116
- - Synthetic Chat data
117
- - Code instructions
118
  [More Information Needed]
119
 
120
  ## Training Details
121
 
122
  ### Training Data
123
 
124
- Link to TowerBlocks.
125
-
126
- ### Training Procedure
127
-
128
- Write sth about Axolotl.
129
 
130
  #### Training Hyperparameters
131
 
 
21
 
22
  ### Model Description
23
 
24
+ TowerInstruct-7B is a language model that results from fine-tuning TowerBase on the TowerBlocks supervised fine-tuning dataset. TowerInstruct-7B-v0.1 is the first model in the series.
25
  The model is trained to handle several translation-related tasks, such as general machine translation (e.g., sentence- and document-level translation, terminology-aware translation, context-aware translation), automatic post edition, named-entity recognition, gramatical error correction, and paraphrase generation.
26
  We will release more details in the upcoming technical report.
27
 
 
29
  - **Model type:** A 7B parameter model fine-tuned on a mix of publicly available, synthetic datasets on translation-related tasks, as well as conversational datasets and code instructions.
30
  - **Language(s) (NLP):** English, Portuguese, Spanish, French, German, Dutch, Italian, Korean, Chinese, Russian
31
  - **License:** CC-BY-NC-4.0
32
+ - **Finetuned from model:** TowerBase [ADD LINK]
33
 
34
  ## Intended uses & limitations
35
 
 
45
  - Synthetic Chat data
46
  - Code instructions
47
 
48
+ You can find the dataset and all data sources of TowerBlocks [ADD LINK] here.
49
 
50
  Here's how you can run the model using the `pipeline()` function from 🤗 Transformers:
51
 
 
95
 
96
  ### Supervised tasks
97
 
98
+ The prompts for all supervised tasks can be found in TowerBlocks [ADD LINK]. We have used multiple prompt templates for each task. While different prompts may offer different outputs, the difference in downstream performance should be very minimal.
99
+
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
100
  [More Information Needed]
101
 
102
  ## Training Details
103
 
104
  ### Training Data
105
 
106
+ Link to TowerBlocks [ADD LINK].
 
 
 
 
107
 
108
  #### Training Hyperparameters
109