Push model using huggingface_hub.
Browse files- README.md +42 -0
- adapter_config.json +20 -0
- adapter_model.bin +3 -0
- pytorch_model.bin +3 -0
- special_tokens_map.json +24 -0
- tokenizer.json +0 -0
- tokenizer_config.json +31 -0
README.md
ADDED
@@ -0,0 +1,42 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
license: apache-2.0
|
3 |
+
tags:
|
4 |
+
- trl
|
5 |
+
- transformers
|
6 |
+
- reinforcement-learning
|
7 |
+
---
|
8 |
+
|
9 |
+
# TRL Model
|
10 |
+
|
11 |
+
This is a [TRL language model](https://github.com/lvwerra/trl) that has been fine-tuned with reinforcement learning to
|
12 |
+
guide the model outputs according to a value, function, or human feedback. The model can be used for text generation.
|
13 |
+
|
14 |
+
## Usage
|
15 |
+
|
16 |
+
To use this model for inference, first install the TRL library:
|
17 |
+
|
18 |
+
```bash
|
19 |
+
python -m pip install trl
|
20 |
+
```
|
21 |
+
|
22 |
+
You can then generate text as follows:
|
23 |
+
|
24 |
+
```python
|
25 |
+
from transformers import pipeline
|
26 |
+
|
27 |
+
generator = pipeline("text-generation", model="renyulin//tmp/tmp6w__m82c/renyulin/llama-7b-es-ppo-adpater")
|
28 |
+
outputs = generator("Hello, my llama is cute")
|
29 |
+
```
|
30 |
+
|
31 |
+
If you want to use the model for training or to obtain the outputs from the value head, load the model as follows:
|
32 |
+
|
33 |
+
```python
|
34 |
+
from transformers import AutoTokenizer
|
35 |
+
from trl import AutoModelForCausalLMWithValueHead
|
36 |
+
|
37 |
+
tokenizer = AutoTokenizer.from_pretrained("renyulin//tmp/tmp6w__m82c/renyulin/llama-7b-es-ppo-adpater")
|
38 |
+
model = AutoModelForCausalLMWithValueHead.from_pretrained("renyulin//tmp/tmp6w__m82c/renyulin/llama-7b-es-ppo-adpater")
|
39 |
+
|
40 |
+
inputs = tokenizer("Hello, my llama is cute", return_tensors="pt")
|
41 |
+
outputs = model(**inputs, labels=inputs["input_ids"])
|
42 |
+
```
|
adapter_config.json
ADDED
@@ -0,0 +1,20 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"base_model_name_or_path": "renyulin/llama-7b-se-sft-merged",
|
3 |
+
"bias": "none",
|
4 |
+
"fan_in_fan_out": false,
|
5 |
+
"inference_mode": true,
|
6 |
+
"init_lora_weights": true,
|
7 |
+
"layers_pattern": null,
|
8 |
+
"layers_to_transform": null,
|
9 |
+
"lora_alpha": 32,
|
10 |
+
"lora_dropout": 0.05,
|
11 |
+
"modules_to_save": null,
|
12 |
+
"peft_type": "LORA",
|
13 |
+
"r": 4,
|
14 |
+
"revision": null,
|
15 |
+
"target_modules": [
|
16 |
+
"q_proj",
|
17 |
+
"v_proj"
|
18 |
+
],
|
19 |
+
"task_type": "CAUSAL_LM"
|
20 |
+
}
|
adapter_model.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:4ecb44b47865d5aacb2392611d68340329eed6038a775f29989ef97c3e759553
|
3 |
+
size 8434381
|
pytorch_model.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:01c229488bbde8f3127278101bee565f8f058c08a2060bbc850e7b0cf428b6c8
|
3 |
+
size 17471
|
special_tokens_map.json
ADDED
@@ -0,0 +1,24 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"bos_token": {
|
3 |
+
"content": "<s>",
|
4 |
+
"lstrip": false,
|
5 |
+
"normalized": true,
|
6 |
+
"rstrip": false,
|
7 |
+
"single_word": false
|
8 |
+
},
|
9 |
+
"eos_token": {
|
10 |
+
"content": "</s>",
|
11 |
+
"lstrip": false,
|
12 |
+
"normalized": true,
|
13 |
+
"rstrip": false,
|
14 |
+
"single_word": false
|
15 |
+
},
|
16 |
+
"pad_token": "</s>",
|
17 |
+
"unk_token": {
|
18 |
+
"content": "<unk>",
|
19 |
+
"lstrip": false,
|
20 |
+
"normalized": true,
|
21 |
+
"rstrip": false,
|
22 |
+
"single_word": false
|
23 |
+
}
|
24 |
+
}
|
tokenizer.json
ADDED
The diff for this file is too large to render.
See raw diff
|
|
tokenizer_config.json
ADDED
@@ -0,0 +1,31 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"bos_token": {
|
3 |
+
"__type": "AddedToken",
|
4 |
+
"content": "<s>",
|
5 |
+
"lstrip": false,
|
6 |
+
"normalized": true,
|
7 |
+
"rstrip": false,
|
8 |
+
"single_word": false
|
9 |
+
},
|
10 |
+
"clean_up_tokenization_spaces": false,
|
11 |
+
"eos_token": {
|
12 |
+
"__type": "AddedToken",
|
13 |
+
"content": "</s>",
|
14 |
+
"lstrip": false,
|
15 |
+
"normalized": true,
|
16 |
+
"rstrip": false,
|
17 |
+
"single_word": false
|
18 |
+
},
|
19 |
+
"model_max_length": 2048,
|
20 |
+
"pad_token": null,
|
21 |
+
"sp_model_kwargs": {},
|
22 |
+
"tokenizer_class": "LlamaTokenizer",
|
23 |
+
"unk_token": {
|
24 |
+
"__type": "AddedToken",
|
25 |
+
"content": "<unk>",
|
26 |
+
"lstrip": false,
|
27 |
+
"normalized": true,
|
28 |
+
"rstrip": false,
|
29 |
+
"single_word": false
|
30 |
+
}
|
31 |
+
}
|