modelId stringlengths 4 111 | lastModified stringlengths 24 24 | tags list | pipeline_tag stringlengths 5 30 ⌀ | author stringlengths 2 34 ⌀ | config null | securityStatus null | id stringlengths 4 111 | likes int64 0 9.53k | downloads int64 2 73.6M | library_name stringlengths 2 84 ⌀ | created timestamp[us] | card stringlengths 101 901k | card_len int64 101 901k | embeddings list |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
TheBloke/gpt4-alpaca-lora_mlp-65B-HF | 2023-06-05T00:10:08.000Z | [
"transformers",
"pytorch",
"llama",
"text-generation",
"text2text-generation",
"dataset:c-s-ale/alpaca-gpt4-data",
"license:other",
"endpoints_compatible",
"has_space",
"text-generation-inference",
"region:us"
] | text2text-generation | TheBloke | null | null | TheBloke/gpt4-alpaca-lora_mlp-65B-HF | 7 | 5,979 | transformers | 2023-05-12T22:28:37 | ---
license: other
datasets:
- c-s-ale/alpaca-gpt4-data
pipeline_tag: text2text-generation
---
<!-- header start -->
<div style="width: 100%;">
<img src="https://i.imgur.com/EBdldam.jpg" alt="TheBlokeAI" style="width: 100%; min-width: 400px; display: block; margin: auto;">
</div>
<div style="display: flex; justify-content: space-between; width: 100%;">
<div style="display: flex; flex-direction: column; align-items: flex-start;">
<p><a href="https://discord.gg/Jq4vkcDakD">Chat & support: my new Discord server</a></p>
</div>
<div style="display: flex; flex-direction: column; align-items: flex-end;">
<p><a href="https://www.patreon.com/TheBlokeAI">Want to contribute? TheBloke's Patreon page</a></p>
</div>
</div>
<!-- header end -->
## GPT4-Alpaca-LoRA_MLP-65B GPTQ
These files are the result of merging the [LoRA weights of chtan's gpt4-alpaca-lora_mlp-65B](https://huggingface.co/chtan/gpt4-alpaca-lora_mlp-65b) with the original Llama 65B model.
## Repositories available
* [4bit GPTQ models for GPU inference](https://huggingface.co/TheBloke/gpt4-alpaca-lora_mlp-65B-GPTQ)
* [4bit and 5bit GGML models for CPU inference in llama.cpp](https://huggingface.co/TheBloke/gpt4-alpaca-lora_mlp-65B-GGML)
* [float16 unquantised model for GPU inference and further conversions](https://huggingface.co/TheBloke/gpt4-alpaca-lora_mlp-65B-HF)
<!-- footer start -->
## Discord
For further support, and discussions on these models and AI in general, join us at:
[TheBloke AI's Discord server](https://discord.gg/Jq4vkcDakD)
## Thanks, and how to contribute.
Thanks to the [chirper.ai](https://chirper.ai) team!
I've had a lot of people ask if they can contribute. I enjoy providing models and helping people, and would love to be able to spend even more time doing it, as well as expanding into new projects like fine tuning/training.
If you're able and willing to contribute it will be most gratefully received and will help me to keep providing more models, and to start work on new AI projects.
Donaters will get priority support on any and all AI/LLM/model questions and requests, access to a private Discord room, plus other benefits.
* Patreon: https://patreon.com/TheBlokeAI
* Ko-Fi: https://ko-fi.com/TheBlokeAI
**Patreon special mentions**: Aemon Algiz, Dmitriy Samsonov, Nathan LeClaire, Trenton Dambrowitz, Mano Prime, David Flickinger, vamX, Nikolai Manek, senxiiz, Khalefa Al-Ahmad, Illia Dulskyi, Jonathan Leane, Talal Aujan, V. Lukas, Joseph William Delisle, Pyrater, Oscar Rangel, Lone Striker, Luke Pendergrass, Eugene Pentland, Sebastain Graf, Johann-Peter Hartman.
Thank you to all my generous patrons and donaters!
<!-- footer end -->
# Original model card
This repo provides the training checkpoint of LLaMA on the alpaca_data_gpt4 dataset via LoRA [MLP] on 8xA100(80G).
He et al. 2022 gave an insight that FFN can better utilize modification at larger capacities.
The codes is provided by [tloen/alpaca-lora: Instruct-tune LLaMA on consumer hardware (github.com)](https://github.com/tloen/alpaca-lora).
We modify the running scripts to
```bash
torchrun --nproc_per_node=8 finetune.py \
--base_model '/cache1/chtan/large_models/llama-hf/llama-65b' \
--data_path './alpaca_data_gpt4.json' \
--output_dir './gpt4-alpaca-lora_mlp-65b' \
--batch_size 128 \
--micro_batch_size 2 \
--num_epochs 10 \
--learning_rate 1e-4 \
--cutoff_len 512 \
--val_set_size 2000 \
--lora_r 16 \
--lora_alpha 16 \
--lora_dropout 0.05 \
--lora_target_modules '[gate_proj,down_proj,up_proj]' \
--train_on_inputs \
--group_by_length
```
### Examples
**Instruction**: Tell me about alpacas.
**gpt4-alpaca-lora_mlp-65b**:
Alpacas are small, domesticated mammals that are closely related to llamas. They are native to the Andes Mountains of South America, primarily in Peru, Bolivia, and Chile. These animals have been domesticated for thousands of years and were used by the Incas for their fleece, meat, and as pack animals.
Alpacas are known for their soft and warm fleece, which comes in over 22 natural colors. Their fleece is highly valued and is used in the textile industry to make clothing and other products. Alpacas are also known for their friendly and gentle nature, making them popular pets and great additions to farms.
These animals are social and live in herds, typically with one dominant male. They communicate through a series of hums, clucks, and other vocalizations. Alpacas are also known for their distinctive "humming" sound, which they make when they are content or curious.
In terms of physical appearance, alpacas have a small, stocky build, with an average height of around 36 inches at the withers. They have large, expressive eyes, and their ears are perked forward. Alpacas are herbivores and primarily eat grasses and other vegetation. They are also known for their efficient digestive system, which allows them to thrive on a diet of low-quality forage.
Overall, alpacas are unique and fascinating animals that have been an important part of Andean culture for centuries. They continue to be valued for their fleece, meat, and as pets, and their popularity continues to grow around the world.
**Alpaca-LoRA**: Alpacas are members of the camelid family and are native to the Andes Mountains of South America. They are known for their soft, luxurious fleece, which is used to make clothing, blankets, and other items. Alpacas are herbivores and graze on grasses and other plants. They are social animals and live in herds of up to 20 individuals.
**Stanford Alpaca**: Alpacas are small, fluffy animals related to camels and llamas. They are native to Peru and Bolivia, and were first domesticated around 5,000 years ago. They are kept mainly for their fine, soft fleece, which is used to make knitwear and other garments. Alpacas are herd animals and live in small family groups, led by an older male. They are highly social animals and can form strong bonds with their owners.
**text-davinci-003**: Alpacas are a domesticated species of South American camelid, similar to llamas. They are native to the Andes Mountains in South America and are kept as livestock for their soft, luxurious wool. Alpacas are social animals and live in herds of up to 20 individuals. They are typically kept in small herds of two to five animals, and are relatively easy to care for. Alpacas are herbivores and feed on grass, hay, and other vegetation. They are also known for their gentle and friendly nature, making them popular as pets.
---
**Instruction**: Write a Python program that prints the first 10 Fibonacci numbers.
**gpt4-alpaca-lora_mlp-65b**:
Here is a Python program that prints the first 10 Fibonacci numbers:
```python
# initialize variables
a = 0
b = 1
# loop to print the first 10 Fibonacci numbers
for i in range(10):
print(a, end=" ")
a, b = b, a + b
```
In this program, we initialize two variables `a` and `b` with the first two Fibonacci numbers, which are 0 and 1. Then, we use a for loop to iterate 10 times. In each iteration, we print the current value of `a` and update the values of `a` and `b` using the formula `a = b` and `b = a + b`. This formula is the definition of the Fibonacci sequence.
**Alpaca-LoRA**:
```python
def fibonacci(n):
if n == 0:
return 0
elif n == 1:
return 1
else:
return fibonacci(n-1) + fibonacci(n-2)
print(fibonacci(10))
```
**Stanford Alpaca**:
```python
numbers = [0, 1]
for i in range(2, 11):
numbers.append(numbers[i-2] + numbers[i-1])
print(numbers)
# Output: [0, 1, 1, 2, 3, 5, 8, 13, 21, 34]
```
**text-davinci-003**:
```python
# Fibonacci numbers are the numbers in the following integer sequence, called the Fibonacci sequence, and characterized by the fact that every number after the first two is the sum of the two preceding ones.
# Fibonacci sequence: 0, 1, 1, 2, 3, 5, 8, 13, 21, 34, ...
# Program:
# Initialize the first two numbers of the Fibonacci sequence
a = 0
b = 1
# Print the first 10 numbers of the Fibonacci sequence
print(a)
print(b)
for i in range(2,10):
c = a + b
a = b
b = c
print(c)
```
---
> [1] Junxian He, Chunting Zhou, Xuezhe Ma, Taylor Berg-Kirkpatrick, Graham Neubig: Towards a Unified View of Parameter-Efficient Transfer Learning. ICLR 2022
| 8,350 | [
[
-0.05792236328125,
-0.05718994140625,
0.0178985595703125,
0.02276611328125,
-0.0133209228515625,
-0.01241302490234375,
0.0082550048828125,
-0.060638427734375,
0.049102783203125,
0.0260772705078125,
-0.0472412109375,
-0.0239105224609375,
-0.040679931640625,
0.0102081298828125,
-0.030120849609375,
0.083740234375,
0.00542449951171875,
-0.01506805419921875,
0.036834716796875,
-0.031646728515625,
-0.03314208984375,
-0.022247314453125,
-0.05694580078125,
-0.032806396484375,
0.043487548828125,
-0.00217437744140625,
0.05279541015625,
0.06329345703125,
0.02734375,
0.038238525390625,
-0.01023101806640625,
0.018524169921875,
-0.029815673828125,
-0.02294921875,
0.020538330078125,
-0.02484130859375,
-0.039398193359375,
0.00444793701171875,
0.02423095703125,
0.04339599609375,
-0.01500701904296875,
0.01617431640625,
-0.0060272216796875,
0.0489501953125,
-0.02911376953125,
0.02630615234375,
-0.03436279296875,
-0.002490997314453125,
-0.01168060302734375,
-0.005451202392578125,
-0.0005164146423339844,
-0.05194091796875,
-0.00861358642578125,
-0.07427978515625,
0.02301025390625,
-0.0189208984375,
0.08197021484375,
0.026031494140625,
-0.024139404296875,
-0.05126953125,
-0.0399169921875,
0.04833984375,
-0.048736572265625,
0.010223388671875,
0.035400390625,
0.00676727294921875,
-0.03125,
-0.043121337890625,
-0.056304931640625,
-0.01381683349609375,
-0.004398345947265625,
0.01064300537109375,
-0.0249786376953125,
-0.029083251953125,
0.01091766357421875,
0.044525146484375,
-0.02093505859375,
-0.0003864765167236328,
-0.045135498046875,
0.00215911865234375,
0.039398193359375,
-0.0043182373046875,
0.0161895751953125,
0.017120361328125,
-0.029876708984375,
-0.0306396484375,
-0.045135498046875,
0.02557373046875,
0.0307769775390625,
0.01690673828125,
-0.049530029296875,
0.040985107421875,
-0.0235595703125,
0.042877197265625,
0.01395416259765625,
-0.029510498046875,
0.0252532958984375,
-0.0285797119140625,
-0.01386260986328125,
-0.0216217041015625,
0.0618896484375,
0.02447509765625,
-0.0004467964172363281,
0.01654052734375,
0.005718231201171875,
0.0005803108215332031,
-0.0180206298828125,
-0.0479736328125,
-0.0128173828125,
0.01319122314453125,
-0.048553466796875,
-0.0294647216796875,
-0.004245758056640625,
-0.0621337890625,
-0.01383209228515625,
0.00653076171875,
0.038543701171875,
-0.028045654296875,
0.0002486705780029297,
0.0278472900390625,
-0.023651123046875,
0.043670654296875,
0.037445068359375,
-0.0770263671875,
0.0234222412109375,
0.042449951171875,
0.05548095703125,
0.042449951171875,
-0.0369873046875,
-0.0281524658203125,
0.0131988525390625,
-0.0181121826171875,
0.051361083984375,
-0.01227569580078125,
-0.0304412841796875,
-0.0257568359375,
0.031402587890625,
0.01128387451171875,
-0.02679443359375,
0.022735595703125,
-0.0355224609375,
0.00732421875,
-0.046112060546875,
-0.0239715576171875,
-0.029052734375,
0.0035343170166015625,
-0.07232666015625,
0.048858642578125,
0.01377105712890625,
-0.0521240234375,
0.004924774169921875,
-0.057647705078125,
-0.02362060546875,
-0.0074005126953125,
0.008453369140625,
-0.0111541748046875,
-0.0148468017578125,
0.01213836669921875,
0.0115509033203125,
-0.0311431884765625,
-0.031646728515625,
-0.0216522216796875,
-0.031341552734375,
0.0233612060546875,
-0.024078369140625,
0.09027099609375,
0.00920867919921875,
-0.031646728515625,
-0.0190277099609375,
-0.05999755859375,
-0.016632080078125,
0.039794921875,
-0.044952392578125,
0.0085296630859375,
-0.00012159347534179688,
-0.031341552734375,
-0.00585174560546875,
0.03924560546875,
-0.032928466796875,
0.03399658203125,
-0.01105499267578125,
0.06103515625,
0.0618896484375,
-0.0025005340576171875,
0.0182342529296875,
-0.052398681640625,
0.01519012451171875,
-0.0255279541015625,
0.03265380859375,
-0.01256561279296875,
-0.0625,
-0.08233642578125,
-0.020721435546875,
0.00799560546875,
0.046478271484375,
-0.0265655517578125,
0.050262451171875,
0.01058197021484375,
-0.0538330078125,
-0.03271484375,
0.01422882080078125,
0.0004143714904785156,
0.037445068359375,
0.0222015380859375,
-0.0280303955078125,
-0.0286712646484375,
-0.0496826171875,
0.0271148681640625,
-0.03594970703125,
0.01015472412109375,
0.01788330078125,
0.05126953125,
-0.02960205078125,
0.03729248046875,
-0.04815673828125,
-0.036895751953125,
0.001422882080078125,
-0.0108184814453125,
0.041656494140625,
0.050933837890625,
0.06268310546875,
-0.031585693359375,
-0.00691986083984375,
0.00591278076171875,
-0.04827880859375,
0.00249481201171875,
0.0045166015625,
-0.0198516845703125,
0.00972747802734375,
-0.004302978515625,
-0.0631103515625,
0.039215087890625,
0.043365478515625,
-0.0282745361328125,
0.01306915283203125,
-0.0178375244140625,
0.01213836669921875,
-0.045135498046875,
0.0007824897766113281,
-0.0205078125,
0.005313873291015625,
-0.024017333984375,
0.0020618438720703125,
-0.034423828125,
0.0038814544677734375,
-0.0303802490234375,
0.04595947265625,
-0.031524658203125,
-0.00861358642578125,
-0.00659942626953125,
-0.00943756103515625,
0.029815673828125,
0.043731689453125,
-0.015472412109375,
0.04290771484375,
0.0271148681640625,
-0.043487548828125,
0.0498046875,
0.039764404296875,
-0.017181396484375,
0.004604339599609375,
-0.06585693359375,
0.007534027099609375,
0.00238800048828125,
0.0517578125,
-0.05340576171875,
-0.0091552734375,
0.045654296875,
-0.033203125,
0.005157470703125,
0.0018329620361328125,
-0.02288818359375,
-0.040557861328125,
-0.0305633544921875,
0.04083251953125,
0.050079345703125,
-0.04034423828125,
0.039093017578125,
0.032073974609375,
0.0024127960205078125,
-0.0511474609375,
-0.04315185546875,
-0.0237884521484375,
-0.0194549560546875,
-0.04461669921875,
0.017486572265625,
-0.0244293212890625,
-0.021697998046875,
-0.00707244873046875,
0.01438140869140625,
0.002826690673828125,
0.0004413127899169922,
0.0297698974609375,
0.0263671875,
-0.0185394287109375,
-0.013671875,
-0.01551055908203125,
0.0017404556274414062,
-0.0115966796875,
0.0012874603271484375,
0.06829833984375,
-0.056732177734375,
-0.025390625,
-0.055450439453125,
0.03594970703125,
0.036834716796875,
-0.012359619140625,
0.056549072265625,
0.04534912109375,
-0.027496337890625,
0.01306915283203125,
-0.0511474609375,
-0.000629425048828125,
-0.040679931640625,
-0.007076263427734375,
-0.0236053466796875,
-0.040313720703125,
0.0626220703125,
0.028167724609375,
0.01284027099609375,
0.060028076171875,
0.040557861328125,
-0.015716552734375,
0.051544189453125,
0.05584716796875,
-0.0185699462890625,
0.04022216796875,
-0.070068359375,
-0.008544921875,
-0.043121337890625,
-0.0418701171875,
-0.057342529296875,
-0.0273895263671875,
-0.046844482421875,
-0.035797119140625,
0.01015472412109375,
0.01247406005859375,
-0.044769287109375,
0.029388427734375,
-0.04571533203125,
0.036712646484375,
0.0312347412109375,
0.01198577880859375,
0.01898193359375,
0.0004017353057861328,
0.0290374755859375,
0.035797119140625,
-0.03839111328125,
-0.035888671875,
0.058563232421875,
0.038848876953125,
0.05572509765625,
0.00891876220703125,
0.0538330078125,
0.0303802490234375,
0.025909423828125,
-0.061004638671875,
0.037567138671875,
-0.020172119140625,
-0.035400390625,
-0.02349853515625,
-0.00920867919921875,
-0.07745361328125,
0.01343536376953125,
-0.00646209716796875,
-0.0518798828125,
0.04034423828125,
-0.00539398193359375,
-0.044464111328125,
0.0242767333984375,
-0.0303192138671875,
0.049713134765625,
-0.0236053466796875,
-0.0138702392578125,
-0.005626678466796875,
-0.04644775390625,
0.026702880859375,
0.00449371337890625,
0.02667236328125,
-0.0335693359375,
-0.01074981689453125,
0.072265625,
-0.0709228515625,
0.0733642578125,
0.0007343292236328125,
-0.0222625732421875,
0.056488037109375,
-0.0011320114135742188,
0.038604736328125,
-0.00213623046875,
-0.0223236083984375,
0.03326416015625,
0.00911712646484375,
-0.0413818359375,
-0.006103515625,
0.060516357421875,
-0.08978271484375,
-0.035797119140625,
-0.036651611328125,
-0.0338134765625,
0.0188751220703125,
0.01076507568359375,
0.032867431640625,
0.03302001953125,
-0.013458251953125,
0.01427459716796875,
0.0170745849609375,
-0.0210113525390625,
0.049591064453125,
0.0257720947265625,
-0.003299713134765625,
-0.052276611328125,
0.057769775390625,
0.007442474365234375,
0.0026874542236328125,
0.0164794921875,
0.020111083984375,
-0.0160980224609375,
-0.00789642333984375,
-0.039459228515625,
0.048431396484375,
-0.0626220703125,
-0.0173187255859375,
-0.04571533203125,
-0.0166015625,
-0.028106689453125,
-0.034820556640625,
-0.014739990234375,
-0.038604736328125,
-0.041778564453125,
0.01152801513671875,
0.055938720703125,
0.062255859375,
-0.0233612060546875,
0.0533447265625,
-0.04290771484375,
0.030517578125,
0.043914794921875,
-0.00015616416931152344,
0.0101165771484375,
-0.033905029296875,
0.0038280487060546875,
0.015869140625,
-0.04095458984375,
-0.07427978515625,
0.044464111328125,
0.020172119140625,
0.038604736328125,
0.04779052734375,
-0.01261138916015625,
0.072265625,
-0.0250701904296875,
0.0584716796875,
0.0228271484375,
-0.073486328125,
0.040557861328125,
-0.025970458984375,
-0.00801849365234375,
0.038177490234375,
0.0217132568359375,
-0.017486572265625,
-0.0203399658203125,
-0.055511474609375,
-0.048797607421875,
0.04473876953125,
0.0159454345703125,
0.0057373046875,
-0.0120391845703125,
0.026702880859375,
0.02093505859375,
0.00371551513671875,
-0.07403564453125,
-0.0340576171875,
-0.0399169921875,
0.0106048583984375,
-0.00205230712890625,
0.0181427001953125,
-0.0269927978515625,
-0.042694091796875,
0.07037353515625,
-0.00731658935546875,
0.03948974609375,
0.01226806640625,
0.0201416015625,
-0.017669677734375,
-0.00377655029296875,
0.049713134765625,
0.050994873046875,
-0.0177001953125,
-0.01232147216796875,
0.0222015380859375,
-0.037017822265625,
0.0193328857421875,
0.00569915771484375,
-0.00922393798828125,
-0.006580352783203125,
0.0350341796875,
0.049285888671875,
-0.0167999267578125,
-0.038177490234375,
0.0299072265625,
-0.015106201171875,
0.004177093505859375,
-0.031005859375,
0.02105712890625,
0.0199127197265625,
0.036590576171875,
0.0307159423828125,
-0.0019483566284179688,
0.0017652511596679688,
-0.065673828125,
-0.0120086669921875,
0.039306640625,
0.00910186767578125,
-0.023101806640625,
0.046844482421875,
0.0187225341796875,
-0.036834716796875,
0.044097900390625,
-0.0002503395080566406,
-0.0287322998046875,
0.105224609375,
0.05804443359375,
0.039459228515625,
-0.01241302490234375,
0.0131988525390625,
0.0246429443359375,
0.0251312255859375,
-0.00112152099609375,
0.020172119140625,
-0.01219940185546875,
-0.04449462890625,
0.01059722900390625,
-0.06402587890625,
-0.03521728515625,
0.03118896484375,
-0.041015625,
0.0321044921875,
-0.05816650390625,
-0.0157012939453125,
0.01383209228515625,
0.0207366943359375,
-0.057342529296875,
0.006908416748046875,
0.01007843017578125,
0.07611083984375,
-0.0645751953125,
0.062469482421875,
0.0244293212890625,
-0.036529541015625,
-0.0611572265625,
-0.0248260498046875,
-0.015594482421875,
-0.08489990234375,
0.0303192138671875,
0.020416259765625,
-0.017059326171875,
-0.01629638671875,
-0.05218505859375,
-0.07647705078125,
0.1109619140625,
0.0159454345703125,
-0.02099609375,
0.00925445556640625,
0.0001226663589477539,
0.029541015625,
-0.0419921875,
0.01361083984375,
0.04266357421875,
0.03546142578125,
0.0178375244140625,
-0.046234130859375,
0.0007491111755371094,
-0.03125,
-0.00556182861328125,
0.0005273818969726562,
-0.10333251953125,
0.08795166015625,
-0.02532958984375,
0.00698089599609375,
0.03924560546875,
0.057342529296875,
0.037261962890625,
0.0150604248046875,
0.045135498046875,
0.035400390625,
0.061981201171875,
-0.0009379386901855469,
0.067626953125,
-0.026397705078125,
0.0203094482421875,
0.053009033203125,
-0.0161285400390625,
0.052154541015625,
0.0309600830078125,
-0.03607177734375,
0.045135498046875,
0.06365966796875,
-0.0203399658203125,
0.038726806640625,
0.005207061767578125,
-0.0274505615234375,
0.006954193115234375,
-0.014984130859375,
-0.07537841796875,
0.041259765625,
0.03314208984375,
-0.031890869140625,
-0.0023288726806640625,
-0.018798828125,
0.00396728515625,
-0.04425048828125,
-0.02557373046875,
0.0458984375,
0.005077362060546875,
-0.04229736328125,
0.06781005859375,
0.00315093994140625,
0.0494384765625,
-0.061248779296875,
-0.0087432861328125,
-0.032623291015625,
0.0230560302734375,
-0.016143798828125,
-0.044525146484375,
0.01399993896484375,
-0.0104217529296875,
-0.0165252685546875,
0.01641845703125,
0.042510986328125,
0.0014753341674804688,
-0.034027099609375,
0.03497314453125,
0.0184173583984375,
0.037506103515625,
0.037078857421875,
-0.053955078125,
0.0300750732421875,
0.00957489013671875,
-0.0264892578125,
0.016815185546875,
0.021942138671875,
-0.0008087158203125,
0.055877685546875,
0.06280517578125,
0.003932952880859375,
0.01898193359375,
-0.0162811279296875,
0.0670166015625,
-0.051788330078125,
-0.02471923828125,
-0.06329345703125,
0.01149749755859375,
0.02508544921875,
-0.0297088623046875,
0.035675048828125,
0.0285797119140625,
0.0511474609375,
-0.01300048828125,
0.036407470703125,
-0.020233154296875,
0.00733184814453125,
-0.0239715576171875,
0.053466796875,
-0.03363037109375,
0.0120849609375,
-0.01081085205078125,
-0.064697265625,
0.0038623809814453125,
0.0794677734375,
-0.0009765625,
0.0056915283203125,
0.0240325927734375,
0.051544189453125,
0.004772186279296875,
0.01132965087890625,
-0.0115814208984375,
0.0218658447265625,
0.0291595458984375,
0.057159423828125,
0.061981201171875,
-0.06561279296875,
0.04571533203125,
-0.050079345703125,
-0.0260009765625,
-0.00746917724609375,
-0.0699462890625,
-0.032806396484375,
-0.0288848876953125,
-0.026336669921875,
-0.029510498046875,
-0.007190704345703125,
0.054718017578125,
0.0465087890625,
-0.05426025390625,
-0.040863037109375,
0.0191192626953125,
0.0181121826171875,
-0.0218048095703125,
-0.01506805419921875,
0.03363037109375,
0.0225677490234375,
-0.050933837890625,
0.0284271240234375,
0.005435943603515625,
0.04254150390625,
-0.00653839111328125,
-0.01415252685546875,
-0.0243072509765625,
0.0257720947265625,
0.039215087890625,
0.060028076171875,
-0.057952880859375,
-0.01436614990234375,
-0.019561767578125,
-0.00911712646484375,
0.016510009765625,
0.0220489501953125,
-0.0609130859375,
-0.0196533203125,
0.037750244140625,
0.01473236083984375,
0.045745849609375,
0.0020236968994140625,
0.0019235610961914062,
-0.00435638427734375,
0.029205322265625,
-0.00506591796875,
0.054901123046875,
0.012237548828125,
-0.0299072265625,
0.055206298828125,
0.0192108154296875,
-0.056488037109375,
-0.05322265625,
-0.004245758056640625,
-0.10601806640625,
-0.0120849609375,
0.05316162109375,
-0.01233673095703125,
-0.043182373046875,
0.01041412353515625,
-0.023468017578125,
0.035430908203125,
-0.031768798828125,
0.0543212890625,
0.0271148681640625,
-0.033721923828125,
0.00936126708984375,
-0.041778564453125,
0.0260772705078125,
0.01227569580078125,
-0.0699462890625,
-0.0182647705078125,
0.03485107421875,
0.042449951171875,
0.049896240234375,
0.0687255859375,
-0.0057830810546875,
0.025848388671875,
0.006748199462890625,
0.01515960693359375,
-0.002841949462890625,
-0.00815582275390625,
-0.022247314453125,
0.01309967041015625,
-0.01438140869140625,
-0.036712646484375
]
] |
BEE-spoke-data/TinyLlama-1.1bee | 2023-09-24T16:13:56.000Z | [
"transformers",
"safetensors",
"llama",
"text-generation",
"bees",
"beekeeping",
"honey",
"en",
"dataset:BEE-spoke-data/bees-internal",
"license:apache-2.0",
"endpoints_compatible",
"text-generation-inference",
"region:us"
] | text-generation | BEE-spoke-data | null | null | BEE-spoke-data/TinyLlama-1.1bee | 1 | 5,979 | transformers | 2023-09-19T22:30:43 | ---
license: apache-2.0
base_model: PY007/TinyLlama-1.1B-intermediate-step-240k-503b
tags:
- bees
- beekeeping
- honey
metrics:
- accuracy
inference:
parameters:
max_new_tokens: 64
do_sample: true
repetition_penalty: 1.1
no_repeat_ngram_size: 5
eta_cutoff: 0.0008
widget:
- text: In beekeeping, the term "queen excluder" refers to
example_title: Queen Excluder
- text: One way to encourage a honey bee colony to produce more honey is by
example_title: Increasing Honey Production
- text: The lifecycle of a worker bee consists of several stages, starting with
example_title: Lifecycle of a Worker Bee
- text: Varroa destructor is a type of mite that
example_title: Varroa Destructor
- text: In the world of beekeeping, the acronym PPE stands for
example_title: Beekeeping PPE
- text: The term "robbing" in beekeeping refers to the act of
example_title: Robbing in Beekeeping
- text: |-
Question: What's the primary function of drone bees in a hive?
Answer:
example_title: Role of Drone Bees
- text: To harvest honey from a hive, beekeepers often use a device known as a
example_title: Honey Harvesting Device
- text: >-
Problem: You have a hive that produces 60 pounds of honey per year. You
decide to split the hive into two. Assuming each hive now produces at a 70%
rate compared to before, how much honey will you get from both hives next
year?
To calculate
example_title: Beekeeping Math Problem
- text: In beekeeping, "swarming" is the process where
example_title: Swarming
pipeline_tag: text-generation
datasets:
- BEE-spoke-data/bees-internal
language:
- en
---
# TinyLlama-1.1bee 🐝
As we feverishly hit the refresh button on hf.co's homepage, on the hunt for the newest waifu chatbot to grace the AI stage, an epiphany struck us like a bee sting. What could we offer to the hive-mind of the community? The answer was as clear as honey—beekeeping, naturally. And thus, this un-bee-lievable model was born.
## Details
This model is a fine-tuned version of [PY007/TinyLlama-1.1B-intermediate-step-240k-503b](https://huggingface.co/PY007/TinyLlama-1.1B-intermediate-step-240k-503b) on the `BEE-spoke-data/bees-internal` dataset.
It achieves the following results on the evaluation set:
- Loss: 2.4285
- Accuracy: 0.4969
```
***** eval metrics *****
eval_accuracy = 0.4972
eval_loss = 2.4283
eval_runtime = 0:00:53.12
eval_samples = 239
eval_samples_per_second = 4.499
eval_steps_per_second = 1.129
perplexity = 11.3391
```
## 📜 Intended Uses & Limitations 📜
### Intended Uses:
1. **Educational Engagement**: Whether you're a novice beekeeper, an enthusiast, or someone just looking to understand the buzz around bees, this model aims to serve as an informative and entertaining resource.
2. **General Queries**: Have questions about hive management, bee species, or honey extraction? Feel free to consult the model for general insights.
3. **Academic & Research Inspiration**: If you're diving into the world of apiculture studies or environmental science, our model could offer some preliminary insights and ideas.
### Limitations:
1. **Not a Beekeeping Expert**: As much as we admire bees and their hard work, this model is not a certified apiculturist. Please consult professional beekeeping resources or experts for serious decisions related to hive management, bee health, and honey production.
2. **Licensing**: Apache-2.0, following TinyLlama
3. **Infallibility**: Our model can err, just like any other piece of technology (or bee). Always double-check the information before applying it to your own hive or research.
4. **Ethical Constraints**: This model may not be used for any illegal or unethical activities, including but not limited to: bioterrorism & standard terrorism, harassment, or spreading disinformation.
## Training and evaluation data
While the full dataset is not yet complete and therefore not yet released for "safety reasons", you can check out a preliminary sample at: [bees-v0](https://huggingface.co/datasets/BEE-spoke-data/bees-v0)
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 0.0001
- train_batch_size: 4
- eval_batch_size: 4
- seed: 80085
- gradient_accumulation_steps: 8
- total_train_batch_size: 32
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- lr_scheduler_warmup_ratio: 0.03
- num_epochs: 2.0 | 4,695 | [
[
-0.029876708984375,
-0.074951171875,
0.0380859375,
0.002471923828125,
-0.017333984375,
-0.015777587890625,
-0.027435302734375,
-0.05755615234375,
0.0205078125,
0.0217132568359375,
-0.0711669921875,
-0.022430419921875,
-0.0306854248046875,
0.007343292236328125,
-0.03192138671875,
0.08209228515625,
0.036285400390625,
0.000499725341796875,
-0.03375244140625,
0.0024509429931640625,
-0.039581298828125,
-0.0291900634765625,
-0.041015625,
-0.0181732177734375,
0.0455322265625,
0.042633056640625,
0.054046630859375,
0.061279296875,
0.04541015625,
0.015716552734375,
-0.01849365234375,
-0.004199981689453125,
-0.04278564453125,
-0.022735595703125,
0.015411376953125,
-0.048431396484375,
-0.0210113525390625,
0.025909423828125,
0.0252227783203125,
0.05560302734375,
-0.017547607421875,
0.06280517578125,
-0.0021305084228515625,
0.040069580078125,
-0.033050537109375,
0.029083251953125,
-0.034576416015625,
0.002063751220703125,
-0.023223876953125,
0.015625,
-0.030242919921875,
-0.04229736328125,
0.0196533203125,
-0.04229736328125,
-0.00589752197265625,
0.016998291015625,
0.0721435546875,
0.0469970703125,
-0.0169677734375,
-0.0198516845703125,
-0.06170654296875,
0.053741455078125,
-0.06378173828125,
0.01351165771484375,
0.0294036865234375,
0.046356201171875,
-0.001636505126953125,
-0.04742431640625,
-0.04168701171875,
-0.02569580078125,
0.0099639892578125,
0.01348876953125,
-0.0279998779296875,
-0.0251312255859375,
0.01216888427734375,
0.012359619140625,
-0.029266357421875,
0.004940032958984375,
-0.0153656005859375,
-0.0067138671875,
0.0227508544921875,
0.026092529296875,
0.00765228271484375,
-0.0204925537109375,
-0.030914306640625,
-0.0209197998046875,
-0.03582763671875,
0.039337158203125,
0.017730712890625,
0.04931640625,
-0.023712158203125,
0.032958984375,
-0.006557464599609375,
0.044525146484375,
0.017974853515625,
-0.0181732177734375,
0.0033664703369140625,
-0.01421356201171875,
-0.0234832763671875,
-0.0210113525390625,
0.03594970703125,
-0.0002646446228027344,
-0.0077362060546875,
-0.017425537109375,
0.0126953125,
0.0273590087890625,
0.02587890625,
-0.051055908203125,
-0.0290985107421875,
0.026123046875,
-0.0426025390625,
-0.0504150390625,
-0.0005164146423339844,
-0.0550537109375,
-0.016876220703125,
-0.017822265625,
0.047332763671875,
-0.027374267578125,
-0.0192718505859375,
-0.017333984375,
-0.0092620849609375,
0.00844573974609375,
0.01004791259765625,
-0.0516357421875,
0.015899658203125,
0.042938232421875,
0.0419921875,
0.005771636962890625,
-0.019561767578125,
0.0160064697265625,
-0.0105438232421875,
-0.0164031982421875,
0.059234619140625,
-0.029937744140625,
-0.028656005859375,
-0.023468017578125,
-0.0126953125,
-0.0065765380859375,
-0.039947509765625,
0.0159454345703125,
-0.03839111328125,
0.01258087158203125,
-0.0174560546875,
-0.0482177734375,
-0.028778076171875,
0.00704193115234375,
-0.0369873046875,
0.054290771484375,
0.01276397705078125,
-0.01558685302734375,
0.0494384765625,
-0.056060791015625,
-0.01024627685546875,
-0.006610870361328125,
0.0137786865234375,
-0.033843994140625,
-0.0131378173828125,
-0.01275634765625,
0.007373809814453125,
-0.0207366943359375,
-0.0011854171752929688,
-0.040802001953125,
-0.0196990966796875,
0.042205810546875,
-0.0240631103515625,
0.08984375,
0.003025054931640625,
-0.0219879150390625,
0.011566162109375,
-0.06365966796875,
0.004116058349609375,
0.0151214599609375,
-0.037353515625,
-0.0032215118408203125,
-0.018218994140625,
0.0037708282470703125,
0.0243072509765625,
0.0251312255859375,
-0.03759765625,
0.01160430908203125,
-0.040130615234375,
0.0252685546875,
0.0618896484375,
0.008880615234375,
0.037078857421875,
-0.045928955078125,
0.0294342041015625,
-0.0271148681640625,
0.056060791015625,
0.01311492919921875,
-0.05950927734375,
-0.0804443359375,
-0.018096923828125,
0.03399658203125,
0.0238189697265625,
-0.007503509521484375,
0.04229736328125,
0.0034122467041015625,
-0.07366943359375,
-0.056243896484375,
0.007549285888671875,
0.045501708984375,
0.06024169921875,
0.019500732421875,
0.01264190673828125,
-0.045623779296875,
-0.057891845703125,
0.0116119384765625,
-0.017181396484375,
0.00218963623046875,
0.0265350341796875,
0.048919677734375,
-0.0301971435546875,
0.09246826171875,
-0.031463623046875,
-0.0299835205078125,
0.0028591156005859375,
0.00688934326171875,
0.03521728515625,
0.033599853515625,
0.045501708984375,
-0.05181884765625,
-0.02874755859375,
-0.0269775390625,
-0.064208984375,
0.0008058547973632812,
0.006626129150390625,
-0.038665771484375,
0.0258026123046875,
-0.00957489013671875,
-0.0687255859375,
0.03985595703125,
0.03851318359375,
-0.0230712890625,
0.044036865234375,
0.005046844482421875,
-0.01214599609375,
-0.056427001953125,
0.0135040283203125,
0.002716064453125,
-0.01222991943359375,
-0.036651611328125,
0.037322998046875,
0.001476287841796875,
-0.01739501953125,
-0.04071044921875,
0.03228759765625,
-0.0168914794921875,
0.005950927734375,
-0.0316162109375,
-0.0167999267578125,
-0.016265869140625,
0.065673828125,
-0.0228729248046875,
0.046966552734375,
0.0282440185546875,
-0.060882568359375,
0.02410888671875,
0.046142578125,
-0.01166534423828125,
0.023712158203125,
-0.035400390625,
0.01479339599609375,
0.01763916015625,
0.0211334228515625,
-0.04681396484375,
-0.026092529296875,
0.039031982421875,
-0.048980712890625,
0.0096282958984375,
-0.012664794921875,
-0.02728271484375,
-0.0634765625,
-0.0138397216796875,
-0.006259918212890625,
0.07373046875,
-0.0472412109375,
0.038330078125,
0.04443359375,
0.01129913330078125,
-0.001956939697265625,
-0.041168212890625,
-0.020263671875,
-0.0285797119140625,
-0.053863525390625,
-0.0166473388671875,
-0.01898193359375,
-0.0275115966796875,
0.0052642822265625,
-0.0277252197265625,
-0.00276947021484375,
0.00839996337890625,
0.044158935546875,
0.0321044921875,
0.0009474754333496094,
0.0034027099609375,
-0.026641845703125,
-0.0159149169921875,
0.018035888671875,
-0.013885498046875,
0.048919677734375,
-0.03472900390625,
-0.01629638671875,
-0.0648193359375,
0.0054931640625,
0.026885986328125,
0.016998291015625,
0.0526123046875,
0.03985595703125,
-0.05206298828125,
0.0101318359375,
-0.037384033203125,
-0.021087646484375,
-0.033935546875,
-0.0014677047729492188,
-0.008087158203125,
-0.0294647216796875,
0.035614013671875,
0.0105438232421875,
0.0069732666015625,
0.057220458984375,
0.043670654296875,
-0.047821044921875,
0.06951904296875,
0.0679931640625,
0.00939178466796875,
0.019073486328125,
-0.04205322265625,
-0.00560760498046875,
-0.05670166015625,
-0.04412841796875,
-0.043701171875,
-0.021270751953125,
-0.0264434814453125,
-0.02362060546875,
0.037872314453125,
-0.004192352294921875,
-0.039886474609375,
0.034515380859375,
-0.0253143310546875,
0.0165252685546875,
0.059356689453125,
0.03814697265625,
0.00768280029296875,
-0.02301025390625,
0.0213470458984375,
-0.004833221435546875,
-0.05670166015625,
-0.044525146484375,
0.08453369140625,
0.027313232421875,
0.058135986328125,
0.0214691162109375,
0.0455322265625,
0.037322998046875,
0.027099609375,
-0.061279296875,
0.040008544921875,
0.00555419921875,
-0.06427001953125,
-0.039398193359375,
-0.0255279541015625,
-0.0687255859375,
0.0012989044189453125,
-0.034423828125,
-0.05474853515625,
0.02392578125,
0.0248565673828125,
-0.054229736328125,
0.00308990478515625,
-0.05963134765625,
0.070556640625,
-0.01424407958984375,
-0.00962066650390625,
-0.00960540771484375,
-0.0555419921875,
0.0239105224609375,
-0.026275634765625,
0.0203094482421875,
-0.026153564453125,
-0.00458526611328125,
0.07623291015625,
-0.06610107421875,
0.08990478515625,
-0.020782470703125,
0.0234527587890625,
0.03668212890625,
0.010284423828125,
0.0290985107421875,
0.004093170166015625,
-0.004421234130859375,
0.03509521484375,
-0.0042724609375,
-0.034515380859375,
-0.0180816650390625,
0.038665771484375,
-0.0791015625,
-0.0535888671875,
-0.037841796875,
-0.033447265625,
-0.0283660888671875,
0.0187530517578125,
0.02606201171875,
0.00531768798828125,
-0.0140228271484375,
0.018402099609375,
0.037384033203125,
0.0011043548583984375,
0.0260467529296875,
0.052032470703125,
-0.02337646484375,
-0.0078277587890625,
0.06805419921875,
-0.0034122467041015625,
0.012664794921875,
0.0098724365234375,
0.044189453125,
-0.027862548828125,
-0.0143890380859375,
-0.041778564453125,
0.0021514892578125,
-0.04254150390625,
-0.03350830078125,
-0.0531005859375,
-0.00482940673828125,
-0.03033447265625,
0.0060577392578125,
-0.0284881591796875,
-0.026275634765625,
-0.04815673828125,
0.003204345703125,
0.051605224609375,
0.050079345703125,
-0.0112152099609375,
0.0439453125,
-0.045135498046875,
0.01383209228515625,
0.031494140625,
0.003021240234375,
0.010650634765625,
-0.05499267578125,
-0.0291900634765625,
0.03765869140625,
-0.015472412109375,
-0.0679931640625,
0.028289794921875,
0.01392364501953125,
0.038848876953125,
0.0374755859375,
-0.005558013916015625,
0.067626953125,
-0.0172576904296875,
0.0565185546875,
0.014190673828125,
-0.055084228515625,
0.0545654296875,
-0.039642333984375,
0.015228271484375,
0.047943115234375,
0.016265869140625,
-0.01410675048828125,
-0.031463623046875,
-0.0654296875,
-0.0635986328125,
0.0687255859375,
0.0224761962890625,
-0.002063751220703125,
0.018280029296875,
0.01500701904296875,
0.022308349609375,
0.0223236083984375,
-0.052764892578125,
-0.0166473388671875,
-0.0075225830078125,
-0.0023212432861328125,
0.011688232421875,
-0.00383758544921875,
-0.012115478515625,
-0.036773681640625,
0.07513427734375,
0.015472412109375,
0.0301055908203125,
-0.0196075439453125,
-0.01715087890625,
-0.037811279296875,
-0.004085540771484375,
0.0066986083984375,
0.045684814453125,
-0.045745849609375,
-0.0249176025390625,
0.01166534423828125,
-0.0262298583984375,
0.004238128662109375,
-0.0001614093780517578,
-0.02593994140625,
-0.0217742919921875,
0.0217437744140625,
0.046600341796875,
0.0300750732421875,
-0.045623779296875,
0.0289154052734375,
0.00848388671875,
0.01377105712890625,
-0.04351806640625,
0.0187225341796875,
0.016632080078125,
0.02532958984375,
0.02874755859375,
0.046630859375,
0.01251983642578125,
-0.052520751953125,
0.0071868896484375,
0.0191192626953125,
-0.01096343994140625,
-0.015472412109375,
0.0679931640625,
0.0223388671875,
-0.03125,
0.045013427734375,
-0.0180511474609375,
-0.0163421630859375,
0.07916259765625,
0.04595947265625,
0.039794921875,
0.00537109375,
0.0189361572265625,
0.0199432373046875,
0.0201873779296875,
0.001605987548828125,
0.006931304931640625,
0.00911712646484375,
-0.0280914306640625,
0.00611114501953125,
-0.0677490234375,
-0.05291748046875,
0.0004184246063232422,
-0.062744140625,
0.012359619140625,
-0.0506591796875,
-0.035491943359375,
0.02362060546875,
0.013275146484375,
-0.07464599609375,
0.0174560546875,
0.020172119140625,
0.0882568359375,
-0.0635986328125,
0.050262451171875,
0.043670654296875,
-0.0645751953125,
-0.072998046875,
0.0076141357421875,
0.02069091796875,
-0.07867431640625,
0.0117034912109375,
0.007843017578125,
0.0013494491577148438,
0.0028209686279296875,
-0.048583984375,
-0.086669921875,
0.08221435546875,
0.00672149658203125,
-0.04364013671875,
-0.005825042724609375,
-0.00580596923828125,
0.051055908203125,
-0.018585205078125,
0.004924774169921875,
0.055328369140625,
0.044677734375,
-0.0104827880859375,
-0.060638427734375,
-0.0031414031982421875,
-0.0206451416015625,
0.00984954833984375,
-0.0091552734375,
-0.07464599609375,
0.08056640625,
-0.0252227783203125,
-0.014312744140625,
0.03863525390625,
0.0472412109375,
0.012786865234375,
0.0246429443359375,
0.035369873046875,
0.041351318359375,
0.06591796875,
-0.0214080810546875,
0.08758544921875,
-0.0182037353515625,
0.0271453857421875,
0.07000732421875,
-0.0032958984375,
0.0660400390625,
0.03277587890625,
-0.03955078125,
0.032623291015625,
0.0814208984375,
-0.0207366943359375,
0.041534423828125,
-0.01149749755859375,
0.0008182525634765625,
-0.02294921875,
-0.0090179443359375,
-0.028411865234375,
0.038482666015625,
0.01558685302734375,
-0.00011831521987915039,
0.0029582977294921875,
0.027801513671875,
-0.004627227783203125,
0.002033233642578125,
-0.0224761962890625,
0.055908203125,
0.002788543701171875,
-0.020355224609375,
0.02874755859375,
-0.01024627685546875,
0.06719970703125,
-0.046905517578125,
0.01003265380859375,
-0.0099639892578125,
0.007404327392578125,
-0.0171661376953125,
-0.039581298828125,
0.018341064453125,
0.01476287841796875,
-0.02099609375,
-0.0011014938354492188,
0.0526123046875,
-0.0015125274658203125,
-0.04595947265625,
0.0199432373046875,
0.042510986328125,
0.0180206298828125,
-0.00347137451171875,
-0.05340576171875,
0.0078887939453125,
-0.004772186279296875,
-0.023834228515625,
0.0073394775390625,
-0.00010603666305541992,
0.0143585205078125,
0.05853271484375,
0.0716552734375,
0.0002942085266113281,
0.004962921142578125,
-0.00856781005859375,
0.053955078125,
-0.05474853515625,
-0.051025390625,
-0.0540771484375,
0.0230712890625,
0.0024166107177734375,
-0.0312347412109375,
0.068359375,
0.042388916015625,
0.05902099609375,
-0.0200958251953125,
0.054534912109375,
-0.017913818359375,
0.0506591796875,
-0.018798828125,
0.0643310546875,
-0.0455322265625,
0.0025424957275390625,
-0.01056671142578125,
-0.05120849609375,
0.005558013916015625,
0.047760009765625,
-0.0136871337890625,
0.0082244873046875,
0.029998779296875,
0.047943115234375,
0.011871337890625,
0.0191497802734375,
0.00879669189453125,
0.034576416015625,
0.0028076171875,
0.03741455078125,
0.06805419921875,
-0.02081298828125,
0.06951904296875,
-0.0567626953125,
-0.03558349609375,
-0.047149658203125,
-0.034149169921875,
-0.054840087890625,
-0.0233154296875,
-0.01467132568359375,
-0.033203125,
-0.0021800994873046875,
0.08782958984375,
0.06768798828125,
-0.05120849609375,
-0.0126495361328125,
0.0104522705078125,
-0.00902557373046875,
-0.0305023193359375,
-0.0169677734375,
0.01641845703125,
0.0010061264038085938,
-0.04534912109375,
0.0260162353515625,
0.0021877288818359375,
0.01885986328125,
-0.021759033203125,
-0.00452423095703125,
-0.012847900390625,
0.016204833984375,
0.0257415771484375,
0.043914794921875,
-0.060150146484375,
-0.03009033203125,
0.004940032958984375,
-0.0178375244140625,
-0.004016876220703125,
0.0098876953125,
-0.057403564453125,
0.0101776123046875,
-0.0029468536376953125,
0.022918701171875,
0.037841796875,
-0.017578125,
0.004375457763671875,
-0.041290283203125,
0.022186279296875,
0.02874755859375,
0.0245819091796875,
0.0047149658203125,
-0.046234130859375,
0.042022705078125,
0.0249176025390625,
-0.04168701171875,
-0.0635986328125,
0.0017108917236328125,
-0.098876953125,
-0.0195770263671875,
0.08270263671875,
0.0141754150390625,
-0.0533447265625,
0.0149383544921875,
-0.016326904296875,
0.039215087890625,
-0.041046142578125,
0.078125,
0.044921875,
-0.040863037109375,
0.01605224609375,
-0.0290069580078125,
0.0321044921875,
0.008270263671875,
-0.0667724609375,
-0.01508331298828125,
0.0411376953125,
0.04315185546875,
0.02471923828125,
0.055328369140625,
0.007320404052734375,
0.042449951171875,
0.028533935546875,
0.0086517333984375,
-0.022064208984375,
-0.032928466796875,
-0.020721435546875,
-0.006626129150390625,
0.00274658203125,
-0.0271453857421875
]
] |
Aeala/VicUnlocked-alpaca-30b | 2023-05-17T17:00:32.000Z | [
"transformers",
"pytorch",
"llama",
"text-generation",
"endpoints_compatible",
"has_space",
"text-generation-inference",
"region:us"
] | text-generation | Aeala | null | null | Aeala/VicUnlocked-alpaca-30b | 7 | 5,978 | transformers | 2023-05-17T02:18:42 | ## Model Info
Merge of my [VicUnlocked-alpaca-half-30b LoRA](https://huggingface.co/Aeala/VicUnlocked-alpaca-half-30b-LoRA)
**Important Note**: While this is trained on a cleaned ShareGPT dataset like Vicuna used, this was trained in the *Alpaca* format, so prompting should be something like:
```
### Instruction:
<prompt> (without the <>)
### Response:
```
## Benchmarks
wikitext2: 4.372413635253906
ptb-new: 24.69171714782715
c4-new: 6.469308853149414
Results generated with GPTQ evals (not quantized) thanks to [Neko-Institute-of-Science](https://huggingface.co/Neko-Institute-of-Science) | 598 | [
[
-0.048187255859375,
-0.05902099609375,
0.0255584716796875,
0.0216827392578125,
-0.03753662109375,
-0.027374267578125,
0.006683349609375,
-0.029998779296875,
0.0168914794921875,
0.0214996337890625,
-0.0552978515625,
-0.04876708984375,
-0.06060791015625,
-0.0113067626953125,
-0.0360107421875,
0.08856201171875,
-0.0132293701171875,
0.0010843276977539062,
0.02587890625,
-0.0269622802734375,
-0.0347900390625,
-0.0341796875,
-0.08380126953125,
-0.018707275390625,
0.048919677734375,
0.01076507568359375,
0.053314208984375,
0.0316162109375,
0.0167694091796875,
0.01517486572265625,
-0.01052093505859375,
0.0217742919921875,
-0.02362060546875,
-0.0173492431640625,
0.0069122314453125,
-0.005184173583984375,
-0.05352783203125,
0.01113128662109375,
0.03369140625,
0.03948974609375,
-0.0219268798828125,
0.02001953125,
0.01045989990234375,
0.040985107421875,
-0.04547119140625,
0.019622802734375,
-0.0245513916015625,
0.0222930908203125,
-0.01024627685546875,
-0.00608062744140625,
-0.00222015380859375,
-0.04852294921875,
-0.00582122802734375,
-0.0635986328125,
0.037750244140625,
-0.00789642333984375,
0.09246826171875,
0.0219879150390625,
-0.0426025390625,
-0.00714111328125,
-0.017303466796875,
0.0294647216796875,
-0.04254150390625,
0.0252838134765625,
0.055389404296875,
0.028167724609375,
-0.0259246826171875,
-0.047760009765625,
-0.05474853515625,
-0.0015993118286132812,
0.0068817138671875,
0.0156402587890625,
-0.0182952880859375,
-0.012176513671875,
0.0290985107421875,
0.032958984375,
-0.030059814453125,
0.012786865234375,
-0.046142578125,
-0.0216827392578125,
0.04156494140625,
0.00746917724609375,
-0.0033550262451171875,
0.007381439208984375,
-0.032470703125,
-0.051971435546875,
-0.03741455078125,
0.01277923583984375,
0.0369873046875,
0.0247344970703125,
-0.044189453125,
0.06988525390625,
-0.037384033203125,
0.050079345703125,
0.007038116455078125,
0.0183258056640625,
0.0560302734375,
-0.03173828125,
-0.030059814453125,
-0.01317596435546875,
0.09033203125,
0.045989990234375,
0.0024356842041015625,
0.004108428955078125,
-0.006671905517578125,
-0.017303466796875,
0.0014715194702148438,
-0.05999755859375,
-0.01898193359375,
0.01172637939453125,
-0.0230255126953125,
-0.0209197998046875,
0.0271759033203125,
-0.05426025390625,
-0.01065826416015625,
-0.006267547607421875,
0.0634765625,
-0.03912353515625,
-0.0012578964233398438,
0.024871826171875,
-0.02264404296875,
0.049835205078125,
0.02996826171875,
-0.059661865234375,
0.023834228515625,
0.033233642578125,
0.049041748046875,
0.01953125,
-0.0341796875,
-0.038604736328125,
-0.007335662841796875,
-0.0222930908203125,
0.04852294921875,
0.00597381591796875,
-0.031219482421875,
-0.0113677978515625,
0.0131988525390625,
-0.00572967529296875,
-0.03717041015625,
0.06634521484375,
-0.0219573974609375,
0.0330810546875,
-0.053131103515625,
-0.00713348388671875,
-0.0207672119140625,
0.027496337890625,
-0.06610107421875,
0.10162353515625,
0.040771484375,
-0.060455322265625,
0.044677734375,
-0.050933837890625,
0.00250244140625,
-0.001316070556640625,
-0.0095977783203125,
-0.051910400390625,
-0.0091552734375,
0.0274658203125,
0.02825927734375,
-0.02606201171875,
0.0157012939453125,
-0.0265350341796875,
-0.0296173095703125,
0.02532958984375,
-0.046783447265625,
0.043609619140625,
0.011688232421875,
-0.020050048828125,
0.0018787384033203125,
-0.052459716796875,
0.011810302734375,
0.0158233642578125,
-0.01273345947265625,
0.005313873291015625,
-0.055267333984375,
-0.0021114349365234375,
-0.0010290145874023438,
0.039825439453125,
-0.037933349609375,
0.0472412109375,
0.0234832763671875,
0.014190673828125,
0.051116943359375,
0.00372314453125,
-0.00479888916015625,
-0.035430908203125,
0.02728271484375,
0.007678985595703125,
0.037567138671875,
0.028656005859375,
-0.0484619140625,
-0.06439208984375,
-0.03814697265625,
0.01959228515625,
0.020782470703125,
-0.044036865234375,
0.0357666015625,
0.01099395751953125,
-0.04925537109375,
-0.03680419921875,
-0.0146942138671875,
0.0225372314453125,
0.04229736328125,
0.043365478515625,
-0.034912109375,
-0.045166015625,
-0.06536865234375,
0.01314544677734375,
-0.018157958984375,
-0.01273345947265625,
0.003688812255859375,
0.0303802490234375,
-0.02886962890625,
0.06561279296875,
-0.04595947265625,
0.0025959014892578125,
0.01366424560546875,
0.0205078125,
0.032958984375,
0.03802490234375,
0.043853759765625,
-0.050384521484375,
-0.0258026123046875,
-0.01116943359375,
-0.05810546875,
-0.022979736328125,
0.0298614501953125,
-0.033294677734375,
-0.009185791015625,
0.0164947509765625,
-0.06463623046875,
0.058837890625,
0.035552978515625,
-0.04840087890625,
0.05474853515625,
-0.00690460205078125,
0.032958984375,
-0.07196044921875,
0.012969970703125,
-0.00868988037109375,
-0.00855255126953125,
-0.01479339599609375,
-0.00716400146484375,
-0.0023174285888671875,
0.0040283203125,
-0.03692626953125,
0.041046142578125,
-0.0248565673828125,
-0.007419586181640625,
-0.01049041748046875,
-0.0221710205078125,
0.0144500732421875,
0.030303955078125,
-0.01125335693359375,
0.063232421875,
0.038330078125,
-0.048675537109375,
0.036712646484375,
0.043365478515625,
-0.01239776611328125,
0.01081085205078125,
-0.059722900390625,
0.0054779052734375,
0.01335906982421875,
0.02294921875,
-0.035400390625,
-0.02142333984375,
0.047515869140625,
-0.017730712890625,
0.0029850006103515625,
-0.009613037109375,
-0.03631591796875,
-0.02777099609375,
-0.0394287109375,
0.0474853515625,
0.0275115966796875,
-0.056060791015625,
0.0328369140625,
-0.00856781005859375,
-0.004413604736328125,
-0.055999755859375,
-0.0369873046875,
-0.03192138671875,
-0.0200653076171875,
-0.024261474609375,
0.020751953125,
-0.0210418701171875,
0.0117950439453125,
0.0032939910888671875,
-0.0036602020263671875,
-0.0185394287109375,
-0.01052093505859375,
0.03070068359375,
0.042633056640625,
-0.0113067626953125,
-0.00801849365234375,
0.015899658203125,
-0.0027008056640625,
-0.006801605224609375,
0.0157318115234375,
0.071044921875,
-0.027130126953125,
-0.0249481201171875,
-0.0298309326171875,
0.0152130126953125,
0.0208892822265625,
-0.0016984939575195312,
0.08245849609375,
0.06488037109375,
-0.0212860107421875,
0.011138916015625,
-0.032440185546875,
0.0036678314208984375,
-0.032501220703125,
0.015777587890625,
-0.037872314453125,
-0.05450439453125,
0.0523681640625,
0.02783203125,
-0.01052093505859375,
0.056182861328125,
0.034027099609375,
0.005283355712890625,
0.045562744140625,
0.036712646484375,
0.0057830810546875,
0.042266845703125,
-0.041290283203125,
0.00362396240234375,
-0.066650390625,
-0.048126220703125,
-0.03802490234375,
0.005214691162109375,
-0.0531005859375,
-0.040985107421875,
0.02008056640625,
0.036865234375,
-0.042327880859375,
0.05487060546875,
-0.03759765625,
0.0219573974609375,
0.046600341796875,
0.032562255859375,
0.01262664794921875,
-0.010284423828125,
0.004528045654296875,
0.0306549072265625,
-0.05072021484375,
-0.0219268798828125,
0.08099365234375,
0.0199737548828125,
0.051513671875,
0.0032901763916015625,
0.062286376953125,
0.00878143310546875,
0.021209716796875,
-0.0487060546875,
0.045806884765625,
-0.00835418701171875,
-0.0173492431640625,
-0.023773193359375,
-0.039947509765625,
-0.08526611328125,
0.030029296875,
-0.02325439453125,
-0.0489501953125,
0.01953125,
0.0121307373046875,
-0.032989501953125,
0.00353240966796875,
-0.055572509765625,
0.06829833984375,
-0.002777099609375,
0.0014848709106445312,
0.0165863037109375,
-0.01537322998046875,
0.029144287109375,
-0.0162200927734375,
-0.007381439208984375,
-0.018463134765625,
-0.0007634162902832031,
0.06781005859375,
-0.0615234375,
0.05126953125,
-0.01107025146484375,
-0.014068603515625,
0.0245513916015625,
-0.0272216796875,
0.0262298583984375,
0.006252288818359375,
-0.004192352294921875,
0.0197906494140625,
0.013885498046875,
-0.04669189453125,
-0.0208587646484375,
0.033966064453125,
-0.076416015625,
-0.0160675048828125,
-0.04351806640625,
-0.03790283203125,
-0.02044677734375,
-0.00553131103515625,
0.06121826171875,
0.03515625,
-0.020477294921875,
0.0031452178955078125,
0.033782958984375,
-0.005596160888671875,
0.035186767578125,
0.03607177734375,
-0.0249176025390625,
-0.039703369140625,
0.0283203125,
0.0020465850830078125,
0.005420684814453125,
0.005706787109375,
0.01114654541015625,
-0.030853271484375,
-0.0270538330078125,
-0.019073486328125,
0.030487060546875,
-0.041748046875,
-0.02484130859375,
-0.043121337890625,
-0.0347900390625,
-0.034942626953125,
0.0173492431640625,
-0.036041259765625,
-0.03863525390625,
-0.007755279541015625,
-0.02337646484375,
0.04522705078125,
0.044677734375,
-0.01354217529296875,
0.050079345703125,
-0.06829833984375,
0.0307159423828125,
0.0280914306640625,
0.01708984375,
-0.0150909423828125,
-0.0513916015625,
-0.01995849609375,
0.012359619140625,
-0.031402587890625,
-0.06439208984375,
0.033538818359375,
0.00269317626953125,
0.04644775390625,
0.0219879150390625,
0.01151275634765625,
0.06707763671875,
-0.0051422119140625,
0.0528564453125,
0.022735595703125,
-0.06304931640625,
0.03515625,
-0.02471923828125,
0.02105712890625,
0.049224853515625,
0.043548583984375,
-0.01861572265625,
-0.013946533203125,
-0.06689453125,
-0.076904296875,
0.0294647216796875,
0.0257110595703125,
-0.0124969482421875,
0.00324249267578125,
0.016357421875,
0.040008544921875,
0.0268402099609375,
-0.0625,
-0.01922607421875,
-0.0268402099609375,
-0.0163421630859375,
0.004787445068359375,
-0.0117340087890625,
-0.0399169921875,
-0.0081939697265625,
0.061004638671875,
-0.02362060546875,
0.02789306640625,
0.006420135498046875,
0.02423095703125,
0.0032749176025390625,
0.005767822265625,
0.0176544189453125,
0.0294189453125,
-0.03692626953125,
-0.0194091796875,
0.00687408447265625,
-0.0469970703125,
-0.0074462890625,
0.0323486328125,
-0.01245880126953125,
-0.007038116455078125,
0.034149169921875,
0.0794677734375,
-0.01340484619140625,
-0.0172882080078125,
0.03369140625,
-0.043731689453125,
-0.02459716796875,
-0.0207366943359375,
0.0110015869140625,
-0.0157623291015625,
0.0234222412109375,
0.02813720703125,
-0.01216888427734375,
-0.0014133453369140625,
-0.03802490234375,
-0.0008139610290527344,
0.029937744140625,
-0.0035839080810546875,
-0.015655517578125,
0.02728271484375,
0.0007495880126953125,
0.0014123916625976562,
0.054351806640625,
-0.01861572265625,
-0.0372314453125,
0.048431396484375,
0.0222930908203125,
0.0565185546875,
0.004138946533203125,
0.01434326171875,
0.043243408203125,
0.005962371826171875,
-0.007564544677734375,
0.0298004150390625,
-0.0121612548828125,
-0.06695556640625,
-0.01354217529296875,
-0.044921875,
-0.041290283203125,
0.0362548828125,
-0.09161376953125,
0.0178070068359375,
-0.021270751953125,
-0.03173828125,
-0.0015888214111328125,
0.0116729736328125,
-0.06597900390625,
0.037139892578125,
-0.00707244873046875,
0.0738525390625,
-0.09295654296875,
0.0697021484375,
0.0265960693359375,
-0.043243408203125,
-0.09716796875,
-0.008148193359375,
-0.0223846435546875,
-0.055755615234375,
0.02386474609375,
0.0186614990234375,
0.00421142578125,
-0.0364990234375,
-0.05755615234375,
-0.06011962890625,
0.10162353515625,
0.03509521484375,
-0.040130615234375,
0.01922607421875,
-0.0126495361328125,
0.0225067138671875,
-0.01302337646484375,
0.034820556640625,
0.032745361328125,
0.0211334228515625,
0.01233673095703125,
-0.08306884765625,
0.00644683837890625,
-0.0305938720703125,
-0.02337646484375,
0.021270751953125,
-0.06988525390625,
0.08660888671875,
0.0035114288330078125,
0.004123687744140625,
0.0275421142578125,
0.05731201171875,
0.0531005859375,
0.0404052734375,
0.040740966796875,
0.055694580078125,
0.0616455078125,
-0.0205078125,
0.06549072265625,
-0.00019991397857666016,
0.03729248046875,
0.09991455078125,
-0.010040283203125,
0.054107666015625,
0.0396728515625,
-0.021148681640625,
0.0251922607421875,
0.06939697265625,
-0.0087127685546875,
0.042266845703125,
-0.0218048095703125,
-0.0286712646484375,
0.00707244873046875,
0.0019140243530273438,
-0.076171875,
0.0261077880859375,
-0.0010805130004882812,
-0.0132293701171875,
-0.004154205322265625,
-0.0090789794921875,
0.00595855712890625,
-0.0263519287109375,
-0.01953125,
0.03857421875,
0.006793975830078125,
-0.0526123046875,
0.042022705078125,
-0.006290435791015625,
0.051361083984375,
-0.064208984375,
-0.00290679931640625,
-0.0259246826171875,
-0.0027332305908203125,
-0.01519012451171875,
-0.06298828125,
0.0162200927734375,
0.003299713134765625,
-0.0175933837890625,
0.01259613037109375,
0.0215301513671875,
-0.029541015625,
-0.0168304443359375,
0.0307769775390625,
0.023956298828125,
0.02484130859375,
0.0018186569213867188,
-0.04876708984375,
0.0216064453125,
0.0024929046630859375,
-0.0214385986328125,
0.03594970703125,
0.032562255859375,
-0.0177154541015625,
0.059722900390625,
0.05438232421875,
-0.0069580078125,
0.01551055908203125,
0.0154266357421875,
0.08758544921875,
-0.05242919921875,
-0.03192138671875,
-0.036102294921875,
0.009918212890625,
-0.0079345703125,
-0.049591064453125,
0.04241943359375,
0.058013916015625,
0.07464599609375,
-0.008087158203125,
0.04205322265625,
-0.019073486328125,
0.0161285400390625,
-0.05987548828125,
0.03753662109375,
-0.0259246826171875,
0.00010627508163452148,
0.003345489501953125,
-0.0732421875,
0.029388427734375,
0.04486083984375,
-0.0100250244140625,
0.024932861328125,
0.05377197265625,
0.059722900390625,
0.005413055419921875,
0.0033512115478515625,
0.01430511474609375,
0.006542205810546875,
0.01334381103515625,
0.046142578125,
0.05377197265625,
-0.06732177734375,
0.023651123046875,
-0.039459228515625,
-0.0158538818359375,
-0.00026106834411621094,
-0.06695556640625,
-0.03839111328125,
-0.0276947021484375,
-0.036651611328125,
-0.040771484375,
0.003360748291015625,
0.06048583984375,
0.05316162109375,
-0.048126220703125,
-0.00972747802734375,
0.0023365020751953125,
-0.031707763671875,
-0.01910400390625,
-0.013671875,
0.041168212890625,
0.029144287109375,
-0.05859375,
0.0199737548828125,
-0.0254974365234375,
0.041534423828125,
0.0161895751953125,
-0.015838623046875,
0.0024566650390625,
0.00030803680419921875,
0.015899658203125,
0.042022705078125,
-0.0302581787109375,
-0.015838623046875,
-0.0345458984375,
-0.0201416015625,
0.0035686492919921875,
0.022216796875,
-0.05340576171875,
-0.0006194114685058594,
0.02117919921875,
0.007785797119140625,
0.07012939453125,
0.021636962890625,
0.038299560546875,
-0.04217529296875,
0.02801513671875,
-0.0018053054809570312,
0.04345703125,
0.033721923828125,
-0.0276641845703125,
0.06085205078125,
0.0016155242919921875,
-0.0362548828125,
-0.053070068359375,
-0.0018987655639648438,
-0.1181640625,
0.0164031982421875,
0.0614013671875,
-0.0216827392578125,
-0.032073974609375,
0.0298309326171875,
-0.031341552734375,
0.05364990234375,
-0.0364990234375,
0.04254150390625,
0.033447265625,
-0.00760650634765625,
-0.0153045654296875,
-0.0265960693359375,
0.0216522216796875,
0.022735595703125,
-0.0721435546875,
-0.02838134765625,
0.013702392578125,
0.0311126708984375,
-0.00823974609375,
0.05352783203125,
-0.00934600830078125,
0.029022216796875,
0.0114898681640625,
0.02294921875,
-0.0306549072265625,
-0.0027675628662109375,
-0.0227203369140625,
0.00826263427734375,
0.00881195068359375,
-0.050872802734375
]
] |
ehartford/WizardLM-33B-V1.0-Uncensored | 2023-06-24T11:17:20.000Z | [
"transformers",
"pytorch",
"llama",
"text-generation",
"en",
"dataset:ehartford/WizardLM_evol_instruct_V2_196k_unfiltered_merged_split",
"license:other",
"endpoints_compatible",
"has_space",
"text-generation-inference",
"region:us"
] | text-generation | ehartford | null | null | ehartford/WizardLM-33B-V1.0-Uncensored | 46 | 5,978 | transformers | 2023-06-24T11:08:38 | ---
license: other
datasets:
- ehartford/WizardLM_evol_instruct_V2_196k_unfiltered_merged_split
language:
- en
---
This is a retraining of https://huggingface.co/WizardLM/WizardLM-30B-V1.0 with a filtered dataset, intended to reduce refusals, avoidance, and bias.
Note that LLaMA itself has inherent ethical beliefs, so there's no such thing as a "truly uncensored" model. But this model will be more compliant than WizardLM/WizardLM-7B-V1.0.
Shout out to the open source AI/ML community, and everyone who helped me out.
Note: An uncensored model has no guardrails. You are responsible for anything you do with the model, just as you are responsible for anything you do with any dangerous object such as a knife, gun, lighter, or car. Publishing anything this model generates is the same as publishing it yourself. You are responsible for the content you publish, and you cannot blame the model any more than you can blame the knife, gun, lighter, or car for what you do with it.
Like WizardLM/WizardLM-30B-V1.0, this model is trained with Vicuna-1.1 style prompts.
```
You are a helpful AI assistant.
USER: <prompt>
ASSISTANT:
```
Thank you [chirper.ai](https://chirper.ai) for sponsoring some of my compute! | 1,218 | [
[
-0.01007843017578125,
-0.03887939453125,
0.0205841064453125,
0.01334381103515625,
-0.041778564453125,
-0.01119232177734375,
0.0148468017578125,
-0.03717041015625,
0.004131317138671875,
0.0789794921875,
-0.04962158203125,
-0.042724609375,
-0.039642333984375,
0.01340484619140625,
-0.04241943359375,
0.098388671875,
-0.0008378028869628906,
0.01131439208984375,
-0.03131103515625,
-0.00556182861328125,
-0.029632568359375,
-0.0450439453125,
-0.054443359375,
-0.03363037109375,
0.057647705078125,
-0.00276947021484375,
0.04901123046875,
0.04437255859375,
0.018035888671875,
0.019012451171875,
-0.005649566650390625,
0.01123809814453125,
-0.05328369140625,
-0.00196075439453125,
-0.0302886962890625,
-0.0207061767578125,
-0.040191650390625,
0.0271453857421875,
0.003955841064453125,
0.0193634033203125,
-0.035858154296875,
0.0440673828125,
0.00478363037109375,
0.040496826171875,
-0.051910400390625,
-0.009918212890625,
-0.052154541015625,
0.0089874267578125,
-0.00670623779296875,
-0.0024509429931640625,
-0.03192138671875,
-0.02044677734375,
-0.01483917236328125,
-0.08673095703125,
0.02020263671875,
0.0218963623046875,
0.0662841796875,
0.053863525390625,
-0.033416748046875,
-0.002223968505859375,
-0.044189453125,
0.048736572265625,
-0.0477294921875,
0.0168609619140625,
0.06317138671875,
0.04345703125,
-0.0228424072265625,
-0.052642822265625,
-0.047088623046875,
-0.01812744140625,
0.01666259765625,
-0.0043182373046875,
-0.01898193359375,
-0.004131317138671875,
-0.0002435445785522461,
0.01407623291015625,
-0.0220947265625,
0.0224609375,
-0.0465087890625,
-0.0240478515625,
0.06890869140625,
0.01203155517578125,
0.01035308837890625,
0.01238250732421875,
-0.02789306640625,
-0.002796173095703125,
-0.05633544921875,
-0.0022125244140625,
0.044097900390625,
0.0083770751953125,
-0.02276611328125,
0.07122802734375,
-0.0282135009765625,
0.03173828125,
0.004741668701171875,
-0.01152801513671875,
0.02142333984375,
-0.0010662078857421875,
-0.038482666015625,
0.0017766952514648438,
0.046600341796875,
0.052093505859375,
0.032684326171875,
-0.00901031494140625,
-0.00458526611328125,
-0.0104827880859375,
0.0418701171875,
-0.046630859375,
-0.0221099853515625,
0.0234222412109375,
-0.05560302734375,
-0.04083251953125,
-0.0161590576171875,
-0.029998779296875,
-0.06475830078125,
-0.0278472900390625,
0.0253753662109375,
-0.00342559814453125,
-0.029693603515625,
0.01342010498046875,
-0.009979248046875,
0.03125,
0.047271728515625,
-0.047271728515625,
0.004024505615234375,
0.041168212890625,
0.04266357421875,
0.009490966796875,
-0.0184326171875,
-0.0193023681640625,
0.01233673095703125,
-0.032318115234375,
0.040069580078125,
-0.0273590087890625,
-0.038665771484375,
-0.00801849365234375,
0.01078033447265625,
0.0233154296875,
-0.0455322265625,
0.0289306640625,
-0.0572509765625,
-0.00217437744140625,
-0.00594329833984375,
-0.02471923828125,
-0.0296783447265625,
0.0206756591796875,
-0.052154541015625,
0.05474853515625,
0.0007410049438476562,
-0.051177978515625,
0.0380859375,
-0.044189453125,
0.0019989013671875,
-0.0200958251953125,
-0.01092529296875,
-0.033966064453125,
-0.012847900390625,
0.00911712646484375,
0.0090179443359375,
-0.021575927734375,
0.0380859375,
-0.03765869140625,
-0.0190887451171875,
0.02459716796875,
-0.05718994140625,
0.09661865234375,
0.0120849609375,
-0.0161590576171875,
0.01983642578125,
-0.0772705078125,
-0.022369384765625,
0.025421142578125,
-0.02984619140625,
-0.0118408203125,
-0.00525665283203125,
0.0062103271484375,
-0.0014562606811523438,
0.02093505859375,
-0.0478515625,
0.0145263671875,
0.004032135009765625,
-0.0146026611328125,
0.08599853515625,
0.0006508827209472656,
0.018035888671875,
-0.0271759033203125,
0.042572021484375,
-0.005401611328125,
0.03955078125,
0.028411865234375,
-0.047576904296875,
-0.0540771484375,
-0.028778076171875,
0.0078582763671875,
0.040985107421875,
-0.039520263671875,
0.03424072265625,
0.0163726806640625,
-0.05706787109375,
-0.058563232421875,
0.0231475830078125,
0.025115966796875,
0.045623779296875,
0.0286865234375,
-0.01471710205078125,
-0.033935546875,
-0.07232666015625,
-0.0058441162109375,
-0.0169830322265625,
-0.024505615234375,
0.004108428955078125,
0.026275634765625,
-0.0172576904296875,
0.06854248046875,
-0.017578125,
-0.025543212890625,
0.00743865966796875,
-0.013427734375,
0.009857177734375,
0.055908203125,
0.0401611328125,
-0.054718017578125,
-0.023651123046875,
-0.000009298324584960938,
-0.09161376953125,
-0.01128387451171875,
0.0001341104507446289,
-0.0307159423828125,
0.0019311904907226562,
0.016387939453125,
-0.034637451171875,
0.0703125,
0.024993896484375,
-0.033905029296875,
0.043060302734375,
-0.0126800537109375,
-0.0080718994140625,
-0.06646728515625,
0.001583099365234375,
-0.00849151611328125,
-0.0186004638671875,
-0.03350830078125,
-0.0001882314682006836,
-0.0252532958984375,
-0.0021190643310546875,
-0.053619384765625,
0.0491943359375,
-0.01323699951171875,
0.00891876220703125,
-0.040557861328125,
-0.017059326171875,
0.01029205322265625,
0.030914306640625,
0.0127105712890625,
0.046295166015625,
0.05975341796875,
-0.060394287109375,
0.031768798828125,
0.0401611328125,
-0.02362060546875,
0.04058837890625,
-0.058319091796875,
0.025421142578125,
-0.0298004150390625,
0.02520751953125,
-0.0299072265625,
-0.00315093994140625,
0.0634765625,
-0.038238525390625,
0.01104736328125,
-0.01373291015625,
-0.0223541259765625,
-0.004779815673828125,
-0.0166778564453125,
0.023681640625,
0.0391845703125,
-0.057342529296875,
0.04278564453125,
0.0341796875,
0.0249176025390625,
-0.062744140625,
-0.053619384765625,
-0.0173797607421875,
-0.034332275390625,
-0.019622802734375,
-0.004730224609375,
-0.01432037353515625,
-0.03570556640625,
0.001007080078125,
-0.005523681640625,
-0.0222320556640625,
0.0164337158203125,
0.040069580078125,
0.0338134765625,
0.005031585693359375,
-0.016693115234375,
0.006244659423828125,
-0.0016679763793945312,
0.0120697021484375,
0.0002256631851196289,
0.015625,
0.0231475830078125,
-0.04290771484375,
-0.047576904296875,
0.03497314453125,
0.0099334716796875,
-0.00530242919921875,
0.05755615234375,
0.041107177734375,
-0.03350830078125,
0.00676727294921875,
-0.033843994140625,
-0.0087432861328125,
-0.04180908203125,
0.0035991668701171875,
-0.00495147705078125,
-0.061248779296875,
0.02838134765625,
0.0263214111328125,
0.036041259765625,
0.034759521484375,
0.052642822265625,
-0.006443023681640625,
0.054962158203125,
0.06341552734375,
-0.008697509765625,
0.0228271484375,
-0.0024051666259765625,
0.0017004013061523438,
-0.05889892578125,
-0.055450439453125,
-0.030120849609375,
-0.004131317138671875,
-0.041748046875,
-0.0237884521484375,
0.015899658203125,
0.00606536865234375,
-0.05535888671875,
0.0187225341796875,
-0.056610107421875,
0.041168212890625,
0.044464111328125,
0.0277557373046875,
0.03289794921875,
0.007328033447265625,
0.021820068359375,
0.01345062255859375,
-0.03692626953125,
-0.0626220703125,
0.1021728515625,
0.03131103515625,
0.0955810546875,
0.00579833984375,
0.053619384765625,
0.046722412109375,
0.044952392578125,
-0.036651611328125,
0.04119873046875,
0.0102996826171875,
-0.07373046875,
-0.024688720703125,
-0.01324462890625,
-0.08184814453125,
0.02276611328125,
-0.00806427001953125,
-0.07293701171875,
0.028106689453125,
0.0207366943359375,
-0.01456451416015625,
0.04010009765625,
-0.032745361328125,
0.042205810546875,
-0.0145111083984375,
-0.02435302734375,
-0.015838623046875,
-0.032135009765625,
0.0273284912109375,
-0.0117340087890625,
0.002681732177734375,
-0.036468505859375,
-0.005329132080078125,
0.07196044921875,
-0.042816162109375,
0.10760498046875,
-0.00994110107421875,
-0.0155181884765625,
0.0440673828125,
0.00038242340087890625,
0.0225067138671875,
0.01065826416015625,
-0.0184783935546875,
0.033447265625,
-0.0031414031982421875,
-0.040374755859375,
-0.00994110107421875,
0.03204345703125,
-0.08978271484375,
-0.0726318359375,
-0.035003662109375,
-0.026641845703125,
0.005352020263671875,
0.0143585205078125,
0.026275634765625,
0.00042724609375,
-0.01250457763671875,
-0.00350189208984375,
0.040924072265625,
-0.0193939208984375,
0.01181793212890625,
0.044158935546875,
-0.034210205078125,
-0.0169830322265625,
0.047454833984375,
-0.0024776458740234375,
0.0035247802734375,
0.0019779205322265625,
0.01522064208984375,
-0.0305023193359375,
-0.0352783203125,
-0.045013427734375,
0.01302337646484375,
-0.069580078125,
-0.02435302734375,
-0.038421630859375,
-0.03070068359375,
-0.038665771484375,
-0.017974853515625,
-0.02801513671875,
-0.0219268798828125,
-0.04998779296875,
-0.0203704833984375,
0.06298828125,
0.07183837890625,
-0.018890380859375,
0.0235748291015625,
-0.03594970703125,
0.02081298828125,
0.021209716796875,
0.007427215576171875,
-0.0030364990234375,
-0.06097412109375,
-0.01087188720703125,
-0.0012960433959960938,
-0.03765869140625,
-0.04638671875,
0.0213165283203125,
-0.006687164306640625,
0.06610107421875,
0.032318115234375,
0.032562255859375,
0.029205322265625,
-0.03802490234375,
0.05596923828125,
0.0209808349609375,
-0.050262451171875,
0.0271759033203125,
-0.0216522216796875,
-0.014923095703125,
0.04644775390625,
0.036468505859375,
-0.0144195556640625,
-0.036041259765625,
-0.0447998046875,
-0.053619384765625,
0.0308685302734375,
0.0165557861328125,
0.032135009765625,
0.0182037353515625,
0.0299530029296875,
0.03143310546875,
0.038665771484375,
-0.075927734375,
-0.033721923828125,
-0.0540771484375,
0.0186767578125,
0.0157623291015625,
-0.0137176513671875,
-0.034576416015625,
-0.0217132568359375,
0.06353759765625,
-0.00130462646484375,
0.014007568359375,
0.0164947509765625,
0.006153106689453125,
-0.0053863525390625,
-0.002841949462890625,
0.0194549560546875,
0.05560302734375,
-0.011199951171875,
-0.00685882568359375,
0.004070281982421875,
-0.0377197265625,
0.021392822265625,
-0.00516510009765625,
-0.0182952880859375,
-0.01519012451171875,
0.0176239013671875,
0.062744140625,
-0.029022216796875,
-0.038909912109375,
0.025482177734375,
0.0011768341064453125,
-0.01139068603515625,
-0.03497314453125,
0.02020263671875,
-0.00557708740234375,
0.0293426513671875,
0.010986328125,
-0.006023406982421875,
0.025238037109375,
-0.019744873046875,
0.005985260009765625,
0.0236358642578125,
0.0021820068359375,
-0.003162384033203125,
0.07940673828125,
0.01378631591796875,
-0.033416748046875,
0.05572509765625,
-0.01035308837890625,
-0.0103302001953125,
0.056396484375,
0.048553466796875,
0.049835205078125,
-0.01177978515625,
0.035736083984375,
0.030731201171875,
0.036590576171875,
0.00955963134765625,
0.005527496337890625,
-0.0021820068359375,
-0.055145263671875,
-0.0139923095703125,
-0.04461669921875,
-0.054931640625,
0.0148162841796875,
-0.0640869140625,
0.01329803466796875,
-0.05572509765625,
-0.003204345703125,
-0.01320648193359375,
-0.00792694091796875,
-0.046417236328125,
0.027984619140625,
0.0033626556396484375,
0.0816650390625,
-0.054656982421875,
0.0738525390625,
0.0277252197265625,
-0.04315185546875,
-0.05517578125,
-0.007389068603515625,
0.015777587890625,
-0.07781982421875,
0.013946533203125,
0.00965118408203125,
0.001461029052734375,
-0.01371002197265625,
-0.0701904296875,
-0.059600830078125,
0.0919189453125,
0.039093017578125,
-0.033721923828125,
-0.0038585662841796875,
0.00470733642578125,
0.031402587890625,
-0.024627685546875,
0.00189971923828125,
0.0241546630859375,
0.02960205078125,
0.00275421142578125,
-0.06689453125,
0.00510406494140625,
-0.020660400390625,
-0.004047393798828125,
-0.039764404296875,
-0.0555419921875,
0.059814453125,
-0.012359619140625,
0.005855560302734375,
0.0258636474609375,
0.059600830078125,
0.047454833984375,
0.044097900390625,
0.037353515625,
0.0094146728515625,
0.0794677734375,
0.03289794921875,
0.09112548828125,
0.006725311279296875,
0.01271820068359375,
0.087890625,
-0.024993896484375,
0.048614501953125,
0.036285400390625,
-0.0105133056640625,
0.0274810791015625,
0.08038330078125,
-0.019500732421875,
0.047119140625,
0.0267791748046875,
-0.0193328857421875,
-0.0253753662109375,
-0.03173828125,
-0.039093017578125,
0.028045654296875,
-0.004291534423828125,
-0.0226287841796875,
-0.0158233642578125,
0.01145172119140625,
0.0107269287109375,
-0.0008487701416015625,
-0.032928466796875,
0.042999267578125,
0.0150604248046875,
-0.011749267578125,
0.054656982421875,
-0.010467529296875,
0.055023193359375,
-0.042449951171875,
0.0018062591552734375,
-0.014129638671875,
-0.013641357421875,
-0.018463134765625,
-0.0494384765625,
0.0227203369140625,
0.0250244140625,
-0.0126190185546875,
0.005462646484375,
0.05780029296875,
-0.01238250732421875,
-0.0494384765625,
0.0199737548828125,
0.028411865234375,
0.04583740234375,
0.0005474090576171875,
-0.0604248046875,
0.00841522216796875,
0.0024280548095703125,
-0.03515625,
0.0301971435546875,
0.00489044189453125,
-0.017120361328125,
0.075439453125,
0.058013916015625,
-0.01136016845703125,
0.006587982177734375,
0.00994110107421875,
0.08099365234375,
-0.0273590087890625,
-0.0260162353515625,
-0.042724609375,
0.03533935546875,
-0.00682830810546875,
-0.0265350341796875,
0.05535888671875,
0.031951904296875,
0.0665283203125,
-0.0106201171875,
0.054412841796875,
0.004283905029296875,
0.00885009765625,
-0.0467529296875,
0.08331298828125,
-0.056793212890625,
0.0160980224609375,
0.01366424560546875,
-0.045379638671875,
-0.00470733642578125,
0.05523681640625,
-0.002620697021484375,
0.00014269351959228516,
0.01345062255859375,
0.0693359375,
0.00948333740234375,
-0.01013946533203125,
0.0418701171875,
-0.013092041015625,
0.01995849609375,
0.003932952880859375,
0.05389404296875,
-0.01163482666015625,
0.04693603515625,
-0.037200927734375,
-0.0167388916015625,
-0.00295257568359375,
-0.0672607421875,
-0.08477783203125,
-0.025421142578125,
-0.03173828125,
-0.04693603515625,
0.0007300376892089844,
0.0714111328125,
0.0640869140625,
-0.042572021484375,
-0.0158843994140625,
0.00612640380859375,
0.0207366943359375,
-0.0135345458984375,
-0.0087738037109375,
0.011962890625,
0.024993896484375,
-0.049468994140625,
0.035919189453125,
-0.020416259765625,
0.021209716796875,
-0.040557861328125,
-0.01409912109375,
-0.0032329559326171875,
0.0243682861328125,
0.0280609130859375,
0.034393310546875,
-0.0546875,
-0.0303192138671875,
-0.0203704833984375,
-0.00740814208984375,
0.00949859619140625,
0.02764892578125,
-0.0533447265625,
0.0249176025390625,
-0.006565093994140625,
0.04302978515625,
0.031463623046875,
-0.01023101806640625,
0.028900146484375,
-0.03289794921875,
0.0261993408203125,
0.00615692138671875,
0.025543212890625,
0.032470703125,
-0.06591796875,
0.05572509765625,
0.004322052001953125,
-0.046844482421875,
-0.055755615234375,
0.0137176513671875,
-0.06939697265625,
-0.01161956787109375,
0.0816650390625,
-0.01483917236328125,
-0.03948974609375,
-0.01922607421875,
-0.03741455078125,
0.039520263671875,
-0.034210205078125,
0.0616455078125,
0.021087646484375,
-0.00534820556640625,
-0.0113525390625,
-0.046875,
0.024261474609375,
0.00783538818359375,
-0.059600830078125,
0.00592803955078125,
0.038787841796875,
0.03692626953125,
-0.0011262893676757812,
0.07110595703125,
0.0013799667358398438,
0.01983642578125,
0.0199737548828125,
0.017913818359375,
-0.0310821533203125,
-0.0238494873046875,
-0.02099609375,
-0.00789642333984375,
0.01314544677734375,
-0.040557861328125
]
] |
CHIH-HUNG/llama-2-13b-FINETUNE2_3w-gate_up_down_proj | 2023-09-06T04:55:54.000Z | [
"transformers",
"pytorch",
"llama",
"text-generation",
"dataset:huangyt/FINETUNE2",
"license:llama2",
"endpoints_compatible",
"text-generation-inference",
"region:us"
] | text-generation | CHIH-HUNG | null | null | CHIH-HUNG/llama-2-13b-FINETUNE2_3w-gate_up_down_proj | 0 | 5,978 | transformers | 2023-09-01T20:45:20 | ---
license: llama2
datasets:
- huangyt/FINETUNE2
---
# Model Card for Model ID
<!-- Provide a quick summary of what the model is/does. -->
在llama-2-13b上使用huangyt/FINETUNE2資料集進行訓練，總資料筆數約3w
# Fine-Tuning Information
- **GPU:** RTX4090 (single core / 24564MiB)
- **model:** meta-llama/Llama-2-13b-hf
- **dataset:** huangyt/FINETUNE2 (共約3w筆訓練集)
- **peft_type:** LoRA
- **lora_rank:** 8
- **lora_target:** gate_proj, up_proj, down_proj
- **per_device_train_batch_size:** 8
- **gradient_accumulation_steps:** 8
- **learning_rate :** 5e-5
- **epoch:** 1
- **precision:** bf16
- **quantization:** load_in_4bit
# Fine-Tuning Detail
- **train_loss:** 0.614
- **train_runtime:** 3:42:14 (use deepspeed)
# Evaluation
- 評估結果來自**HuggingFaceH4/open_llm_leaderboard**
- 與Llama-2-13b比較4種Benchmark，包含**ARC**、**HellaSwag**、**MMLU**、**TruthfulQA**
| Model |Average| ARC |HellaSwag| MMLU | TruthfulQA |
|-----------------------------------------------------|-------|-------|---------|-------|------------|
|meta-llama/Llama-2-13b-hf | 56.9 | 58.11 | 80.97 | 54.34 | 34.17 |
|meta-llama/Llama-2-13b-chat-hf | 59.93 | 59.04 | 81.94 | 54.64 | 44.12 |
|CHIH-HUNG/llama-2-13b-FINETUNE2_3w | 58.34 | 58.62 | 82.32 | 54.25 | 38.17 |
|CHIH-HUNG/llama-2-13b-FINETUNE2_3w-q_k_v_o_proj | 58.21 | 58.53 | 82.47 | 53.9 | 37.92 |
|CHIH-HUNG/llama-2-13b-FINETUNE2_3w-gate_up_down_proj | 58.65 | 57.42 | 82.42 | 55.57 | 39.19 |
# How to convert dataset to json
- 在**load_dataset**中輸入資料集名稱，並且在**take**中輸入要取前幾筆資料
- 觀察該資料集的欄位名稱，填入**example**欄位中(例如system_prompt、question、response)
- 最後指定json檔儲存位置 (**json_filename**)
```py
import json
from datasets import load_dataset
# 讀取數據集，take可以取得該數據集前n筆資料
dataset = load_dataset("huangyt/FINETUNE2", split="train", streaming=True)
# 提取所需欄位並建立新的字典列表
extracted_data = []
for example in dataset:
extracted_example = {
"instruction": example["instruction"],
"input": example["input"],
"output": example["output"]
}
extracted_data.append(extracted_example)
# 指定 JSON 文件名稱
json_filename = "huangyt_FINETUNE2.json"
# 寫入 JSON 文件
with open(json_filename, "w") as json_file:
json.dump(extracted_data, json_file, indent=4)
print(f"數據已提取並保存為 {json_filename}")
``` | 2,363 | [
[
-0.04339599609375,
-0.05145263671875,
0.01227569580078125,
0.01244354248046875,
-0.047821044921875,
0.003986358642578125,
-0.012786865234375,
-0.021697998046875,
0.01434326171875,
0.033233642578125,
-0.045989990234375,
-0.038177490234375,
-0.044189453125,
0.00897216796875,
-0.020355224609375,
0.0841064453125,
-0.007633209228515625,
-0.010040283203125,
0.021026611328125,
0.007434844970703125,
-0.040557861328125,
-0.0243988037109375,
-0.0550537109375,
-0.03179931640625,
0.0215911865234375,
0.018951416015625,
0.04962158203125,
0.07049560546875,
0.053253173828125,
0.019195556640625,
-0.01175689697265625,
0.0184326171875,
-0.048065185546875,
-0.0186767578125,
0.0181884765625,
-0.0452880859375,
-0.047821044921875,
-0.00525665283203125,
0.0496826171875,
0.0216217041015625,
0.004638671875,
0.045562744140625,
0.01690673828125,
0.043182373046875,
-0.023468017578125,
0.0195770263671875,
-0.0242462158203125,
0.0086212158203125,
-0.0267486572265625,
-0.025299072265625,
-0.0028839111328125,
-0.0243988037109375,
-0.0115966796875,
-0.06646728515625,
0.00449371337890625,
0.01099395751953125,
0.10174560546875,
0.031890869140625,
-0.0235748291015625,
0.00595855712890625,
-0.03973388671875,
0.06304931640625,
-0.07501220703125,
0.0005307197570800781,
0.0249481201171875,
0.0333251953125,
-0.007663726806640625,
-0.052459716796875,
-0.053619384765625,
0.00848388671875,
-0.01158905029296875,
0.0142059326171875,
-0.00934600830078125,
-0.01873779296875,
0.02825927734375,
0.037261962890625,
-0.034454345703125,
0.005970001220703125,
-0.037322998046875,
0.005802154541015625,
0.0634765625,
0.031463623046875,
0.0054473876953125,
-0.0245361328125,
-0.0212860107421875,
-0.0215301513671875,
-0.040130615234375,
0.0183258056640625,
0.03271484375,
0.0305023193359375,
-0.039093017578125,
0.036529541015625,
-0.03790283203125,
0.032806396484375,
0.010467529296875,
-0.0304412841796875,
0.044921875,
-0.015777587890625,
-0.041259765625,
0.0020999908447265625,
0.0775146484375,
0.045379638671875,
-0.00408935546875,
0.018402099609375,
-0.00783538818359375,
-0.0134429931640625,
-0.00730133056640625,
-0.07037353515625,
-0.025909423828125,
0.042388916015625,
-0.053955078125,
-0.032684326171875,
0.007717132568359375,
-0.06512451171875,
-0.0091094970703125,
-0.0094451904296875,
0.0220184326171875,
-0.0236358642578125,
-0.044281005859375,
-0.0023479461669921875,
-0.01324462890625,
0.025543212890625,
0.0245361328125,
-0.0596923828125,
0.01300048828125,
0.04962158203125,
0.052520751953125,
0.009490966796875,
-0.0233154296875,
-0.00933837890625,
0.0139617919921875,
-0.024658203125,
0.04962158203125,
-0.005096435546875,
-0.025970458984375,
-0.015716552734375,
0.02008056640625,
-0.0008530616760253906,
-0.03948974609375,
0.058319091796875,
-0.0309600830078125,
-0.00576019287109375,
-0.038238525390625,
-0.0206756591796875,
-0.03680419921875,
0.032745361328125,
-0.05224609375,
0.08026123046875,
0.00606536865234375,
-0.06475830078125,
0.020904541015625,
-0.0506591796875,
-0.017120361328125,
0.003795623779296875,
-0.00014579296112060547,
-0.03509521484375,
-0.0206146240234375,
0.0169525146484375,
0.042816162109375,
-0.03656005859375,
0.01678466796875,
-0.0175323486328125,
-0.0443115234375,
0.0239410400390625,
-0.0305023193359375,
0.0723876953125,
0.031829833984375,
-0.0185394287109375,
0.0007977485656738281,
-0.07275390625,
0.005611419677734375,
0.047027587890625,
-0.039276123046875,
-0.0040435791015625,
-0.007320404052734375,
0.00653076171875,
-0.0004794597625732422,
0.030059814453125,
-0.01495361328125,
0.02471923828125,
-0.015533447265625,
0.032012939453125,
0.0670166015625,
-0.002155303955078125,
0.0083770751953125,
-0.037322998046875,
0.0223236083984375,
0.0078125,
0.0176239013671875,
-0.0020999908447265625,
-0.03521728515625,
-0.0745849609375,
-0.0171051025390625,
0.0128173828125,
0.042999267578125,
-0.033172607421875,
0.05145263671875,
-0.022552490234375,
-0.052703857421875,
-0.055267333984375,
0.007442474365234375,
0.02056884765625,
0.039215087890625,
0.040740966796875,
0.0083160400390625,
-0.054656982421875,
-0.06573486328125,
0.0025482177734375,
-0.00421142578125,
0.004146575927734375,
0.0286865234375,
0.04705810546875,
-0.0239105224609375,
0.0374755859375,
-0.040008544921875,
-0.0228729248046875,
-0.0261077880859375,
0.0025653839111328125,
0.0662841796875,
0.04376220703125,
0.0504150390625,
-0.035186767578125,
-0.0322265625,
0.00418853759765625,
-0.08428955078125,
0.01078033447265625,
-0.007244110107421875,
-0.0210418701171875,
-0.009429931640625,
0.0025482177734375,
-0.04656982421875,
0.02838134765625,
0.033843994140625,
-0.016693115234375,
0.041595458984375,
0.0086669921875,
0.0234375,
-0.08123779296875,
0.01078033447265625,
-0.016326904296875,
0.005413055419921875,
-0.03240966796875,
0.0166168212890625,
-0.01186370849609375,
0.0224609375,
-0.029205322265625,
0.023284912109375,
-0.0233154296875,
0.00966644287109375,
-0.017364501953125,
-0.0018167495727539062,
0.0009565353393554688,
0.04632568359375,
-0.01174163818359375,
0.04510498046875,
0.039398193359375,
-0.05450439453125,
0.0443115234375,
0.034881591796875,
-0.028472900390625,
0.015533447265625,
-0.0411376953125,
0.00374603271484375,
0.006805419921875,
0.02178955078125,
-0.07415771484375,
-0.0287322998046875,
0.0450439453125,
-0.03253173828125,
0.0186004638671875,
-0.02935791015625,
-0.0305633544921875,
-0.04754638671875,
-0.030853271484375,
0.0214691162109375,
0.0269622802734375,
-0.044464111328125,
0.01384735107421875,
0.011871337890625,
0.015625,
-0.04864501953125,
-0.067138671875,
-0.00574493408203125,
-0.0178375244140625,
-0.035491943359375,
0.016143798828125,
-0.01107025146484375,
-0.007061004638671875,
0.00408935546875,
0.0007262229919433594,
-0.0009899139404296875,
0.00881195068359375,
0.0130767822265625,
0.036285400390625,
-0.0254669189453125,
-0.027374267578125,
0.0096893310546875,
-0.00885009765625,
0.0055389404296875,
0.01081085205078125,
0.060638427734375,
-0.01534271240234375,
-0.014190673828125,
-0.06121826171875,
0.005992889404296875,
0.025115966796875,
0.00626373291015625,
0.040618896484375,
0.062225341796875,
-0.0197601318359375,
0.0032863616943359375,
-0.01953125,
-0.003185272216796875,
-0.037872314453125,
0.027191162109375,
-0.043182373046875,
-0.052215576171875,
0.049072265625,
-0.002437591552734375,
0.0163421630859375,
0.0648193359375,
0.0290069580078125,
-0.0186004638671875,
0.0777587890625,
0.0121612548828125,
-0.01873779296875,
0.017486572265625,
-0.072265625,
0.003246307373046875,
-0.07562255859375,
-0.0246734619140625,
-0.034820556640625,
-0.04669189453125,
-0.049407958984375,
-0.0138397216796875,
0.014129638671875,
0.02264404296875,
-0.049835205078125,
0.031097412109375,
-0.062408447265625,
0.02215576171875,
0.046356201171875,
0.0161285400390625,
0.013763427734375,
-0.006191253662109375,
0.0138397216796875,
0.0032501220703125,
-0.03656005859375,
-0.032257080078125,
0.09881591796875,
0.026275634765625,
0.050811767578125,
0.005237579345703125,
0.0504150390625,
0.00981903076171875,
0.00698089599609375,
-0.048614501953125,
0.047271728515625,
-0.003543853759765625,
-0.052215576171875,
-0.01157379150390625,
-0.02099609375,
-0.04833984375,
0.0238189697265625,
-0.0174713134765625,
-0.05609130859375,
0.00508880615234375,
0.00347900390625,
-0.036590576171875,
0.041900634765625,
-0.03173828125,
0.052764892578125,
-0.0289764404296875,
-0.0251312255859375,
0.000843048095703125,
-0.042724609375,
0.054779052734375,
0.007762908935546875,
0.01076507568359375,
-0.0254364013671875,
0.01021575927734375,
0.08197021484375,
-0.042877197265625,
0.045654296875,
-0.0236358642578125,
0.0002067089080810547,
0.041046142578125,
0.00313568115234375,
0.053497314453125,
0.02496337890625,
0.0017652511596679688,
0.04376220703125,
0.0028057098388671875,
-0.0145111083984375,
-0.023712158203125,
0.056610107421875,
-0.08892822265625,
-0.043853759765625,
-0.042449951171875,
-0.025146484375,
0.01479339599609375,
0.0267333984375,
0.039093017578125,
-0.00514984130859375,
0.0155487060546875,
0.01953125,
0.03546142578125,
-0.003368377685546875,
0.04205322265625,
0.02227783203125,
-0.0141754150390625,
-0.055450439453125,
0.06182861328125,
0.003040313720703125,
-0.0015668869018554688,
0.0275726318359375,
0.00965118408203125,
-0.0187835693359375,
-0.046295166015625,
-0.04425048828125,
0.0193939208984375,
-0.038177490234375,
-0.04669189453125,
-0.03619384765625,
-0.039093017578125,
-0.037933349609375,
0.0026302337646484375,
-0.0413818359375,
-0.0200042724609375,
-0.06036376953125,
-0.01216888427734375,
0.04974365234375,
0.028839111328125,
-0.00592803955078125,
0.0562744140625,
-0.0592041015625,
0.03125,
0.0156097412109375,
0.01435089111328125,
0.00910186767578125,
-0.0628662109375,
-0.0230712890625,
0.009429931640625,
-0.03570556640625,
-0.04510498046875,
0.040557861328125,
0.00017404556274414062,
0.037078857421875,
0.06011962890625,
-0.0023097991943359375,
0.0855712890625,
-0.0157470703125,
0.06732177734375,
0.0162811279296875,
-0.050262451171875,
0.042572021484375,
-0.031768798828125,
-0.0107421875,
0.0379638671875,
0.0232391357421875,
-0.028167724609375,
-0.0000998377799987793,
-0.036376953125,
-0.05828857421875,
0.07904052734375,
0.01300048828125,
-0.004322052001953125,
0.02099609375,
0.01702880859375,
0.0080108642578125,
0.01812744140625,
-0.06549072265625,
-0.045562744140625,
-0.037139892578125,
-0.0036869049072265625,
0.00606536865234375,
-0.01531219482421875,
-0.027191162109375,
-0.03729248046875,
0.058013916015625,
-0.0031375885009765625,
0.03875732421875,
0.009735107421875,
0.0141754150390625,
-0.0181121826171875,
0.006961822509765625,
0.0285186767578125,
0.0275421142578125,
-0.04376220703125,
-0.007724761962890625,
0.009918212890625,
-0.041595458984375,
0.001270294189453125,
0.01264190673828125,
-0.018157958984375,
-0.01084136962890625,
0.0338134765625,
0.0673828125,
0.0016384124755859375,
-0.028533935546875,
0.022125244140625,
0.0033283233642578125,
-0.0216827392578125,
-0.033111572265625,
0.022369384765625,
-0.003116607666015625,
0.039276123046875,
0.040985107421875,
0.0004930496215820312,
0.00921630859375,
-0.0236053466796875,
-0.00836944580078125,
0.0196533203125,
0.01104736328125,
-0.0199737548828125,
0.0714111328125,
0.0028438568115234375,
-0.0121612548828125,
0.043365478515625,
-0.0147552490234375,
-0.0305328369140625,
0.057525634765625,
0.04071044921875,
0.05731201171875,
-0.010528564453125,
-0.0034923553466796875,
0.059814453125,
0.02978515625,
-0.01477813720703125,
0.041900634765625,
-0.0018033981323242188,
-0.049957275390625,
-0.01531219482421875,
-0.054046630859375,
-0.00829315185546875,
0.045745849609375,
-0.052734375,
0.0241241455078125,
-0.055419921875,
-0.021636962890625,
-0.0026397705078125,
0.0255889892578125,
-0.05145263671875,
0.0215911865234375,
0.01290130615234375,
0.064453125,
-0.054595947265625,
0.06927490234375,
0.0251007080078125,
-0.039825439453125,
-0.0728759765625,
-0.0204925537109375,
-0.00982666015625,
-0.07275390625,
0.0404052734375,
0.01324462890625,
0.0217437744140625,
-0.0009832382202148438,
-0.06689453125,
-0.07659912109375,
0.10791015625,
0.01512908935546875,
-0.04791259765625,
0.006832122802734375,
0.0164337158203125,
0.0243377685546875,
-0.013580322265625,
0.032012939453125,
0.05523681640625,
0.04766845703125,
0.004489898681640625,
-0.057373046875,
0.02215576171875,
-0.033843994140625,
-0.00988006591796875,
-0.00131988525390625,
-0.08892822265625,
0.10009765625,
-0.01500701904296875,
0.00211334228515625,
0.01111602783203125,
0.05096435546875,
0.0406494140625,
0.0306396484375,
0.025482177734375,
0.05450439453125,
0.05133056640625,
-0.0251617431640625,
0.054779052734375,
-0.007625579833984375,
0.042938232421875,
0.0662841796875,
-0.005695343017578125,
0.057708740234375,
0.0269775390625,
-0.03704833984375,
0.03814697265625,
0.06951904296875,
-0.03656005859375,
0.052825927734375,
-0.01003265380859375,
-0.004913330078125,
-0.01146697998046875,
0.0034198760986328125,
-0.054901123046875,
0.0279541015625,
0.0290374755859375,
-0.02801513671875,
0.005748748779296875,
-0.020355224609375,
0.0182037353515625,
-0.0283966064453125,
-0.0235748291015625,
0.042694091796875,
-0.01183319091796875,
-0.02606201171875,
0.07794189453125,
-0.003139495849609375,
0.056549072265625,
-0.043975830078125,
-0.0113525390625,
-0.0171051025390625,
0.0110626220703125,
-0.036590576171875,
-0.0606689453125,
-0.0017347335815429688,
0.0022220611572265625,
-0.01235198974609375,
0.016357421875,
0.03582763671875,
-0.01128387451171875,
-0.037872314453125,
0.026580810546875,
0.006626129150390625,
0.0258026123046875,
0.00653076171875,
-0.068603515625,
0.0265960693359375,
0.0189056396484375,
-0.040618896484375,
0.0184326171875,
0.0209197998046875,
0.0224761962890625,
0.05145263671875,
0.07073974609375,
0.007144927978515625,
0.01413726806640625,
-0.009674072265625,
0.07562255859375,
-0.0618896484375,
-0.0287017822265625,
-0.05670166015625,
0.0369873046875,
-0.0195770263671875,
-0.037628173828125,
0.05291748046875,
0.059722900390625,
0.06610107421875,
-0.0025234222412109375,
0.072021484375,
-0.0213775634765625,
0.0394287109375,
-0.03363037109375,
0.057464599609375,
-0.05609130859375,
0.01287841796875,
-0.020904541015625,
-0.0419921875,
-0.00630950927734375,
0.057891845703125,
-0.0022945404052734375,
-0.003505706787109375,
0.04180908203125,
0.043304443359375,
-0.0039043426513671875,
0.01068115234375,
0.0010900497436523438,
0.0288543701171875,
0.026153564453125,
0.06683349609375,
0.047210693359375,
-0.0802001953125,
0.0543212890625,
-0.052642822265625,
-0.00782012939453125,
-0.031890869140625,
-0.044952392578125,
-0.06488037109375,
-0.0197601318359375,
-0.018890380859375,
-0.0297698974609375,
-0.019439697265625,
0.061309814453125,
0.040283203125,
-0.061065673828125,
-0.0267486572265625,
0.0022792816162109375,
0.0090484619140625,
-0.0333251953125,
-0.022735595703125,
0.049072265625,
0.00662994384765625,
-0.06097412109375,
0.028839111328125,
-0.0099334716796875,
0.01007080078125,
-0.0022640228271484375,
-0.0215606689453125,
-0.016876220703125,
-0.0227508544921875,
0.0290679931640625,
0.023162841796875,
-0.05194091796875,
-0.01287841796875,
-0.01450347900390625,
-0.00021576881408691406,
0.0193023681640625,
0.0181884765625,
-0.037689208984375,
0.01099395751953125,
0.036712646484375,
0.024566650390625,
0.046478271484375,
-0.004718780517578125,
-0.005908966064453125,
-0.0306243896484375,
0.022705078125,
-0.0004150867462158203,
0.0264892578125,
0.00710296630859375,
-0.039306640625,
0.05352783203125,
0.035369873046875,
-0.045928955078125,
-0.07763671875,
-0.0296173095703125,
-0.09619140625,
-0.013458251953125,
0.08587646484375,
-0.0008358955383300781,
-0.0400390625,
0.021087646484375,
-0.0230865478515625,
0.045562744140625,
-0.043182373046875,
0.0484619140625,
0.0284423828125,
-0.00978851318359375,
-0.0029888153076171875,
-0.0565185546875,
0.03033447265625,
-0.00592803955078125,
-0.0513916015625,
-0.00342559814453125,
0.0097198486328125,
0.0247344970703125,
0.017242431640625,
0.033111572265625,
0.0043792724609375,
0.0114898681640625,
0.01270294189453125,
0.0077667236328125,
-0.0228271484375,
-0.01171875,
-0.005970001220703125,
-0.00592041015625,
-0.0200958251953125,
-0.0435791015625
]
] |
kandinsky-community/kandinsky-2-2-controlnet-depth | 2023-10-09T11:32:45.000Z | [
"diffusers",
"text-to-image",
"kandinsky",
"license:apache-2.0",
"has_space",
"diffusers:KandinskyV22ControlnetPipeline",
"region:us"
] | text-to-image | kandinsky-community | null | null | kandinsky-community/kandinsky-2-2-controlnet-depth | 15 | 5,977 | diffusers | 2023-06-28T22:51:45 | ---
license: apache-2.0
prior:
- kandinsky-community/kandinsky-2-2-prior
tags:
- text-to-image
- kandinsky
inference: false
---
# Kandinsky 2.2
Kandinsky inherits best practices from Dall-E 2 and Latent diffusion while introducing some new ideas.
It uses the CLIP model as a text and image encoder, and diffusion image prior (mapping) between latent spaces of CLIP modalities. This approach increases the visual performance of the model and unveils new horizons in blending images and text-guided image manipulation.
The Kandinsky model is created by [Arseniy Shakhmatov](https://github.com/cene555), [Anton Razzhigaev](https://github.com/razzant), [Aleksandr Nikolich](https://github.com/AlexWortega), [Igor Pavlov](https://github.com/boomb0om), [Andrey Kuznetsov](https://github.com/kuznetsoffandrey) and [Denis Dimitrov](https://github.com/denndimitrov)
## Usage
Kandinsky 2.2 is available in diffusers!
```python
pip install diffusers transformers accelerate
```
### Text-to-Image Generation with ControlNet Conditioning
```python
import torch
import numpy as np
from transformers import pipeline
from diffusers.utils import load_image
from diffusers import KandinskyV22PriorPipeline, KandinskyV22ControlnetPipeline
# let's take an image and extract its depth map.
def make_hint(image, depth_estimator):
image = depth_estimator(image)["depth"]
image = np.array(image)
image = image[:, :, None]
image = np.concatenate([image, image, image], axis=2)
detected_map = torch.from_numpy(image).float() / 255.0
hint = detected_map.permute(2, 0, 1)
return hint
img = load_image(
"https://huggingface.co/datasets/hf-internal-testing/diffusers-images/resolve/main/kandinskyv22/cat.png"
).resize((768, 768))
# We can use the `depth-estimation` pipeline from transformers to process the image and retrieve its depth map.
depth_estimator = pipeline("depth-estimation")
hint = make_hint(img, depth_estimator).unsqueeze(0).half().to("cuda")
# Now, we load the prior pipeline and the text-to-image controlnet pipeline
pipe_prior = KandinskyV22PriorPipeline.from_pretrained(
"kandinsky-community/kandinsky-2-2-prior", torch_dtype=torch.float16
)
pipe_prior = pipe_prior.to("cuda")
pipe = KandinskyV22ControlnetPipeline.from_pretrained(
"kandinsky-community/kandinsky-2-2-controlnet-depth", torch_dtype=torch.float16
)
pipe = pipe.to("cuda")
# We pass the prompt and negative prompt through the prior to generate image embeddings
prompt = "A robot, 4k photo"
negative_prior_prompt = "lowres, text, error, cropped, worst quality, low quality, jpeg artifacts, ugly, duplicate, morbid, mutilated, out of frame, extra fingers, mutated hands, poorly drawn hands, poorly drawn face, mutation, deformed, blurry, dehydrated, bad anatomy, bad proportions, extra limbs, cloned face, disfigured, gross proportions, malformed limbs, missing arms, missing legs, extra arms, extra legs, fused fingers, too many fingers, long neck, username, watermark, signature"
generator = torch.Generator(device="cuda").manual_seed(43)
image_emb, zero_image_emb = pipe_prior(
prompt=prompt, negative_prompt=negative_prior_prompt, generator=generator
).to_tuple()
# Now we can pass the image embeddings and the depth image we extracted to the controlnet pipeline. With Kandinsky 2.2, only prior pipelines accept `prompt` input. You do not need to pass the prompt to the controlnet pipeline.
images = pipe(
image_embeds=image_emb,
negative_image_embeds=zero_image_emb,
hint=hint,
num_inference_steps=50,
generator=generator,
height=768,
width=768,
).images
images[0].save("robot_cat.png")
```
### Image-to-Image Generation with ControlNet Conditioning
```python
import torch
import numpy as np
from diffusers import KandinskyV22PriorEmb2EmbPipeline, KandinskyV22ControlnetImg2ImgPipeline
from diffusers.utils import load_image
from transformers import pipeline
img = load_image(
"https://huggingface.co/datasets/hf-internal-testing/diffusers-images/resolve/main" "/kandinskyv22/cat.png"
).resize((768, 768))
def make_hint(image, depth_estimator):
image = depth_estimator(image)["depth"]
image = np.array(image)
image = image[:, :, None]
image = np.concatenate([image, image, image], axis=2)
detected_map = torch.from_numpy(image).float() / 255.0
hint = detected_map.permute(2, 0, 1)
return hint
depth_estimator = pipeline("depth-estimation")
hint = make_hint(img, depth_estimator).unsqueeze(0).half().to("cuda")
pipe_prior = KandinskyV22PriorEmb2EmbPipeline.from_pretrained(
"kandinsky-community/kandinsky-2-2-prior", torch_dtype=torch.float16
)
pipe_prior = pipe_prior.to("cuda")
pipe = KandinskyV22ControlnetImg2ImgPipeline.from_pretrained(
"kandinsky-community/kandinsky-2-2-controlnet-depth", torch_dtype=torch.float16
)
pipe = pipe.to("cuda")
prompt = "A robot, 4k photo"
negative_prior_prompt = "lowres, text, error, cropped, worst quality, low quality, jpeg artifacts, ugly, duplicate, morbid, mutilated, out of frame, extra fingers, mutated hands, poorly drawn hands, poorly drawn face, mutation, deformed, blurry, dehydrated, bad anatomy, bad proportions, extra limbs, cloned face, disfigured, gross proportions, malformed limbs, missing arms, missing legs, extra arms, extra legs, fused fingers, too many fingers, long neck, username, watermark, signature"
generator = torch.Generator(device="cuda").manual_seed(43)
# run prior pipeline
img_emb = pipe_prior(prompt=prompt, image=img, strength=0.85, generator=generator)
negative_emb = pipe_prior(prompt=negative_prior_prompt, image=img, strength=1, generator=generator)
# run controlnet img2img pipeline
images = pipe(
image=img,
strength=0.5,
image_embeds=img_emb.image_embeds,
negative_image_embeds=negative_emb.image_embeds,
hint=hint,
num_inference_steps=50,
generator=generator,
height=768,
width=768,
).images
images[0].save("robot_cat.png")
```
Here is the output. Compared with the output from our text-to-image controlnet example, it kept a lot more cat facial details from the original image and worked into the robot style we asked for.
## Model Architecture
### Overview
Kandinsky 2.2 is a text-conditional diffusion model based on unCLIP and latent diffusion, composed of a transformer-based image prior model, a unet diffusion model, and a decoder.
The model architectures are illustrated in the figure below - the chart on the left describes the process to train the image prior model, the figure in the center is the text-to-image generation process, and the figure on the right is image interpolation.
<p float="left">
<img src="https://raw.githubusercontent.com/ai-forever/Kandinsky-2/main/content/kandinsky21.png"/>
</p>
Specifically, the image prior model was trained on CLIP text and image embeddings generated with a pre-trained [CLIP-ViT-G model](https://huggingface.co/laion/CLIP-ViT-g-14-laion2B-s12B-b42K). The trained image prior model is then used to generate CLIP image embeddings for input text prompts. Both the input text prompts and its CLIP image embeddings are used in the diffusion process. A [MoVQGAN](https://openreview.net/forum?id=Qb-AoSw4Jnm) model acts as the final block of the model, which decodes the latent representation into an actual image.
### Details
The image prior training of the model was performed on the [LAION Improved Aesthetics dataset](https://huggingface.co/datasets/bhargavsdesai/laion_improved_aesthetics_6.5plus_with_images), and then fine-tuning was performed on the [LAION HighRes data](https://huggingface.co/datasets/laion/laion-high-resolution).
The main Text2Image diffusion model was trained on [LAION HighRes dataset](https://huggingface.co/datasets/laion/laion-high-resolution) and then fine-tuned with a dataset of 2M very high-quality high-resolution images with descriptions (COYO, anime, landmarks_russia, and a number of others) was used separately collected from open sources.
The main change in Kandinsky 2.2 is the replacement of CLIP-ViT-G. Its image encoder significantly increases the model's capability to generate more aesthetic pictures and better understand text, thus enhancing its overall performance.
Due to the switch CLIP model, the image prior model was retrained, and the Text2Image diffusion model was fine-tuned for 2000 iterations. Kandinsky 2.2 was trained on data of various resolutions, from 512 x 512 to 1536 x 1536, and also as different aspect ratios. As a result, Kandinsky 2.2 can generate 1024 x 1024 outputs with any aspect ratio.
### Evaluation
We quantitatively measure the performance of Kandinsky 2.1 on the COCO_30k dataset, in zero-shot mode. The table below presents FID.
FID metric values for generative models on COCO_30k
| | FID (30k)|
|:------|----:|
| eDiff-I (2022) | 6.95 |
| Image (2022) | 7.27 |
| Kandinsky 2.1 (2023) | 8.21|
| Stable Diffusion 2.1 (2022) | 8.59 |
| GigaGAN, 512x512 (2023) | 9.09 |
| DALL-E 2 (2022) | 10.39 |
| GLIDE (2022) | 12.24 |
| Kandinsky 1.0 (2022) | 15.40 |
| DALL-E (2021) | 17.89 |
| Kandinsky 2.0 (2022) | 20.00 |
| GLIGEN (2022) | 21.04 |
For more information, please refer to the upcoming technical report.
## BibTex
If you find this repository useful in your research, please cite:
```
@misc{kandinsky 2.2,
title = {kandinsky 2.2},
author = {Arseniy Shakhmatov, Anton Razzhigaev, Aleksandr Nikolich, Vladimir Arkhipkin, Igor Pavlov, Andrey Kuznetsov, Denis Dimitrov},
year = {2023},
howpublished = {},
}
``` | 9,894 | [
[
-0.0248260498046875,
-0.047393798828125,
0.03790283203125,
-0.0020351409912109375,
-0.0294036865234375,
-0.006107330322265625,
0.0015287399291992188,
-0.00882720947265625,
0.0081329345703125,
0.03643798828125,
-0.03131103515625,
-0.042449951171875,
-0.050506591796875,
-0.0174560546875,
-0.0089263916015625,
0.04791259765625,
-0.0191650390625,
0.001007080078125,
0.0017938613891601562,
0.011688232421875,
-0.0274810791015625,
0.0033054351806640625,
-0.05657958984375,
-0.0244140625,
0.0280303955078125,
0.0265960693359375,
0.05340576171875,
0.04022216796875,
0.037994384765625,
0.021331787109375,
0.007617950439453125,
0.004924774169921875,
-0.035888671875,
0.0009512901306152344,
0.01080322265625,
-0.0279998779296875,
-0.005687713623046875,
-0.01012420654296875,
0.05780029296875,
0.004810333251953125,
0.0203857421875,
-0.0200653076171875,
0.0005640983581542969,
0.053680419921875,
-0.04833984375,
-0.010040283203125,
-0.015533447265625,
0.0279693603515625,
0.005466461181640625,
-0.0157470703125,
-0.0255126953125,
-0.0033130645751953125,
0.0167236328125,
-0.062255859375,
0.00962066650390625,
-0.00896453857421875,
0.1011962890625,
0.02032470703125,
-0.023040771484375,
-0.011932373046875,
-0.038787841796875,
0.06634521484375,
-0.058807373046875,
0.00894927978515625,
0.01898193359375,
0.0232086181640625,
-0.0031147003173828125,
-0.07989501953125,
-0.0426025390625,
-0.0077667236328125,
-0.022369384765625,
0.041412353515625,
-0.01267242431640625,
0.01055908203125,
0.0247955322265625,
0.01114654541015625,
-0.040618896484375,
0.00101470947265625,
-0.044219970703125,
-0.0223846435546875,
0.056854248046875,
0.007396697998046875,
0.028533935546875,
-0.034271240234375,
-0.03778076171875,
-0.0460205078125,
-0.0217132568359375,
0.01458740234375,
0.025634765625,
-0.027099609375,
-0.03460693359375,
0.035858154296875,
-0.00710296630859375,
0.0426025390625,
0.0140380859375,
-0.0153045654296875,
0.0251007080078125,
-0.019073486328125,
-0.0328369140625,
-0.0173187255859375,
0.08984375,
0.03790283203125,
0.0090179443359375,
0.02288818359375,
-0.00441741943359375,
-0.00665283203125,
0.0015745162963867188,
-0.07757568359375,
-0.04541015625,
0.027618408203125,
-0.037567138671875,
-0.0290985107421875,
-0.002117156982421875,
-0.062164306640625,
-0.0159149169921875,
0.002437591552734375,
0.056488037109375,
-0.04046630859375,
-0.02520751953125,
0.00632476806640625,
-0.02313232421875,
0.036346435546875,
0.0275115966796875,
-0.036407470703125,
0.006626129150390625,
0.0163116455078125,
0.0927734375,
-0.0084991455078125,
-0.01114654541015625,
-0.0198211669921875,
-0.0067596435546875,
-0.036651611328125,
0.041107177734375,
-0.0206451416015625,
-0.0194091796875,
-0.0184783935546875,
0.0235443115234375,
0.004940032958984375,
-0.02972412109375,
0.01372528076171875,
-0.04443359375,
0.0261993408203125,
0.0021686553955078125,
-0.0249786376953125,
-0.0214996337890625,
0.0011510848999023438,
-0.039642333984375,
0.0592041015625,
0.0175933837890625,
-0.0672607421875,
0.0151519775390625,
-0.056243896484375,
-0.0174407958984375,
-0.0162811279296875,
-0.00921630859375,
-0.053680419921875,
-0.028564453125,
0.0166168212890625,
0.04510498046875,
-0.007091522216796875,
0.024322509765625,
-0.005451202392578125,
-0.01441192626953125,
0.01021575927734375,
-0.021820068359375,
0.088623046875,
0.023773193359375,
-0.0399169921875,
0.0129547119140625,
-0.02899169921875,
-0.0036067962646484375,
0.00787353515625,
-0.003322601318359375,
-0.00476837158203125,
-0.0249176025390625,
0.0025424957275390625,
0.0313720703125,
0.01065826416015625,
-0.040435791015625,
0.003139495849609375,
-0.035430908203125,
0.03460693359375,
0.057647705078125,
0.0143890380859375,
0.048797607421875,
-0.0210113525390625,
0.0506591796875,
0.02862548828125,
0.01142120361328125,
-0.029510498046875,
-0.04071044921875,
-0.07196044921875,
-0.029754638671875,
0.00885009765625,
0.05047607421875,
-0.07891845703125,
0.02301025390625,
-0.00384521484375,
-0.0506591796875,
-0.02081298828125,
-0.01007843017578125,
0.026397705078125,
0.04345703125,
0.03070068359375,
-0.033660888671875,
-0.039459228515625,
-0.08245849609375,
0.0169219970703125,
0.018463134765625,
-0.0204315185546875,
0.022979736328125,
0.0565185546875,
-0.009307861328125,
0.055328369140625,
-0.046722412109375,
-0.0139617919921875,
0.01232147216796875,
0.0068359375,
0.028900146484375,
0.069091796875,
0.024688720703125,
-0.050933837890625,
-0.050933837890625,
0.009490966796875,
-0.06939697265625,
0.0215301513671875,
-0.0286865234375,
-0.04034423828125,
0.01953125,
0.034454345703125,
-0.052978515625,
0.05010986328125,
0.03753662109375,
-0.0251922607421875,
0.0421142578125,
-0.01342010498046875,
0.0157318115234375,
-0.09942626953125,
0.021209716796875,
0.0026645660400390625,
-0.0270538330078125,
-0.040740966796875,
-0.00021910667419433594,
0.0002849102020263672,
-0.01065826416015625,
-0.043243408203125,
0.046905517578125,
-0.0457763671875,
0.0024585723876953125,
-0.0084381103515625,
-0.004627227783203125,
0.01293182373046875,
0.05633544921875,
0.018280029296875,
0.03558349609375,
0.07794189453125,
-0.04437255859375,
0.045501708984375,
0.0091094970703125,
-0.032196044921875,
0.033233642578125,
-0.07562255859375,
0.0246124267578125,
-0.00811767578125,
0.010986328125,
-0.0833740234375,
-0.0241241455078125,
0.053680419921875,
-0.053009033203125,
0.036773681640625,
-0.01070404052734375,
-0.0232086181640625,
-0.01505279541015625,
-0.0254974365234375,
0.022705078125,
0.0682373046875,
-0.0221099853515625,
0.043243408203125,
0.00974273681640625,
0.013275146484375,
-0.05682373046875,
-0.07257080078125,
-0.0011339187622070312,
-0.0100555419921875,
-0.06317138671875,
0.0277099609375,
-0.0010824203491210938,
-0.002437591552734375,
0.007541656494140625,
0.0157318115234375,
-0.0145416259765625,
-0.01076507568359375,
0.01885986328125,
0.0143585205078125,
-0.0171966552734375,
-0.0130462646484375,
0.02264404296875,
-0.01061248779296875,
0.0047607421875,
-0.0233612060546875,
0.049041748046875,
0.003322601318359375,
-0.01332855224609375,
-0.06890869140625,
0.0118865966796875,
0.0231781005859375,
0.01251220703125,
0.04595947265625,
0.0740966796875,
-0.02398681640625,
0.007633209228515625,
-0.01345062255859375,
-0.0277099609375,
-0.03900146484375,
0.03668212890625,
-0.0172882080078125,
-0.05145263671875,
0.03985595703125,
0.011474609375,
-0.000014424324035644531,
0.046966552734375,
0.0284576416015625,
-0.026519775390625,
0.0703125,
0.02783203125,
0.0249176025390625,
0.040924072265625,
-0.08465576171875,
-0.01267242431640625,
-0.06982421875,
-0.026031494140625,
-0.005977630615234375,
-0.048736572265625,
-0.0276641845703125,
-0.05078125,
0.03363037109375,
0.0251007080078125,
-0.032012939453125,
0.022979736328125,
-0.043243408203125,
0.02130126953125,
0.041839599609375,
0.029693603515625,
-0.01070404052734375,
0.0245361328125,
-0.021697998046875,
-0.012847900390625,
-0.050384521484375,
-0.0269927978515625,
0.06903076171875,
0.0311431884765625,
0.049530029296875,
-0.01074981689453125,
0.04583740234375,
0.00739288330078125,
0.0032749176025390625,
-0.05517578125,
0.034332275390625,
0.0001512765884399414,
-0.037445068359375,
-0.028076171875,
-0.02001953125,
-0.081787109375,
0.00787353515625,
-0.005252838134765625,
-0.040863037109375,
0.03106689453125,
0.0242156982421875,
-0.01690673828125,
0.031768798828125,
-0.0614013671875,
0.068359375,
0.0019245147705078125,
-0.046417236328125,
-0.0103607177734375,
-0.06732177734375,
0.0288543701171875,
0.01519012451171875,
-0.0153961181640625,
0.01445770263671875,
-0.0150604248046875,
0.07232666015625,
-0.034393310546875,
0.056976318359375,
-0.034912109375,
0.0006895065307617188,
0.0380859375,
0.0097808837890625,
0.027435302734375,
0.0159912109375,
0.0119781494140625,
0.0139312744140625,
-0.005077362060546875,
-0.036651611328125,
-0.052154541015625,
0.045684814453125,
-0.0726318359375,
-0.031402587890625,
-0.03082275390625,
-0.0302581787109375,
0.0282135009765625,
0.022979736328125,
0.057647705078125,
0.040679931640625,
0.019989013671875,
0.024505615234375,
0.045745849609375,
-0.035858154296875,
0.0419921875,
-0.005916595458984375,
-0.0145416259765625,
-0.04559326171875,
0.053924560546875,
0.004917144775390625,
0.043182373046875,
0.03515625,
0.020111083984375,
-0.01285552978515625,
-0.0214996337890625,
-0.0270843505859375,
0.0158538818359375,
-0.052032470703125,
-0.04046630859375,
-0.057373046875,
-0.038787841796875,
-0.02557373046875,
-0.0301055908203125,
-0.024749755859375,
-0.01532745361328125,
-0.059539794921875,
0.0241546630859375,
0.040374755859375,
0.03265380859375,
-0.02459716796875,
0.03741455078125,
-0.03033447265625,
0.0280609130859375,
0.035491943359375,
0.0118865966796875,
0.0019969940185546875,
-0.06280517578125,
-0.01837158203125,
0.0201416015625,
-0.039093017578125,
-0.0533447265625,
0.041290283203125,
0.00806427001953125,
0.0184326171875,
0.0248870849609375,
-0.01544952392578125,
0.040771484375,
-0.010650634765625,
0.06195068359375,
0.029998779296875,
-0.0643310546875,
0.051361083984375,
-0.0240631103515625,
0.038787841796875,
0.004222869873046875,
0.040924072265625,
-0.042388916015625,
-0.0162811279296875,
-0.05291748046875,
-0.05572509765625,
0.06707763671875,
0.030975341796875,
-0.0015439987182617188,
0.027252197265625,
0.0224609375,
-0.007297515869140625,
-0.0023021697998046875,
-0.0579833984375,
-0.0245513916015625,
-0.041412353515625,
-0.011077880859375,
-0.007843017578125,
-0.006961822509765625,
0.0003211498260498047,
-0.0306549072265625,
0.05023193359375,
-0.0025424957275390625,
0.052581787109375,
0.0546875,
-0.0192413330078125,
-0.00347900390625,
-0.021881103515625,
0.0501708984375,
0.0396728515625,
-0.011688232421875,
0.004528045654296875,
0.0006995201110839844,
-0.05291748046875,
0.0189056396484375,
-0.00931549072265625,
-0.02587890625,
0.009490966796875,
0.03692626953125,
0.07647705078125,
-0.016937255859375,
-0.0216064453125,
0.04522705078125,
0.002429962158203125,
-0.039093017578125,
-0.032745361328125,
0.005046844482421875,
0.0108642578125,
0.0208740234375,
0.02777099609375,
0.037200927734375,
0.007061004638671875,
-0.004070281982421875,
0.0170745849609375,
0.026214599609375,
-0.0300140380859375,
-0.023956298828125,
0.03753662109375,
-0.00799560546875,
-0.0198211669921875,
0.03656005859375,
-0.01284027099609375,
-0.0296173095703125,
0.07666015625,
0.0294036865234375,
0.06866455078125,
-0.01142120361328125,
0.033843994140625,
0.06512451171875,
0.02581787109375,
0.006999969482421875,
0.0144500732421875,
0.00690460205078125,
-0.0595703125,
-0.028717041015625,
-0.02178955078125,
0.0033054351806640625,
0.0153656005859375,
-0.021392822265625,
0.034454345703125,
-0.0584716796875,
0.0031032562255859375,
-0.017303466796875,
0.005039215087890625,
-0.06524658203125,
0.018341064453125,
-0.0088348388671875,
0.04791259765625,
-0.05584716796875,
0.055877685546875,
0.03515625,
-0.026458740234375,
-0.0645751953125,
-0.01500701904296875,
-0.0189208984375,
-0.06427001953125,
0.049591064453125,
0.0157012939453125,
-0.0162200927734375,
0.0233917236328125,
-0.07550048828125,
-0.0657958984375,
0.10809326171875,
0.02618408203125,
-0.030609130859375,
0.0157012939453125,
-0.01904296875,
0.03570556640625,
-0.0323486328125,
0.036590576171875,
0.0191497802734375,
0.035614013671875,
0.01342010498046875,
-0.039642333984375,
0.0099334716796875,
-0.0250244140625,
0.0216217041015625,
0.003910064697265625,
-0.0478515625,
0.07318115234375,
-0.024261474609375,
-0.03594970703125,
0.0103302001953125,
0.05120849609375,
0.012786865234375,
0.006748199462890625,
0.039642333984375,
0.04888916015625,
0.0240325927734375,
0.004825592041015625,
0.071044921875,
0.0038909912109375,
0.03271484375,
0.0531005859375,
0.0242156982421875,
0.04150390625,
0.0301666259765625,
-0.00653076171875,
0.054473876953125,
0.054595947265625,
-0.01226806640625,
0.0570068359375,
0.0113525390625,
-0.0006093978881835938,
0.0017833709716796875,
0.0017461776733398438,
-0.03887939453125,
0.027191162109375,
0.0211181640625,
-0.044586181640625,
0.004787445068359375,
0.0182952880859375,
0.004825592041015625,
-0.01259613037109375,
-0.0007920265197753906,
0.03912353515625,
0.001392364501953125,
-0.037353515625,
0.06732177734375,
-0.0028076171875,
0.07489013671875,
-0.039215087890625,
-0.01507568359375,
-0.00849151611328125,
-0.004848480224609375,
-0.0192413330078125,
-0.0794677734375,
0.0153350830078125,
-0.02130126953125,
0.004108428955078125,
-0.00745391845703125,
0.054840087890625,
-0.035980224609375,
-0.0352783203125,
0.0182647705078125,
0.0010747909545898438,
0.03790283203125,
0.00911712646484375,
-0.0645751953125,
0.00970458984375,
0.0158843994140625,
-0.033203125,
-0.005985260009765625,
0.0150604248046875,
0.031829833984375,
0.03900146484375,
0.035400390625,
-0.00728607177734375,
0.006103515625,
-0.025115966796875,
0.060089111328125,
-0.040924072265625,
-0.021392822265625,
-0.062347412109375,
0.069580078125,
-0.020721435546875,
-0.038818359375,
0.041046142578125,
0.03472900390625,
0.06378173828125,
-0.022796630859375,
0.05059814453125,
-0.025115966796875,
0.0056304931640625,
-0.066162109375,
0.06365966796875,
-0.055938720703125,
-0.015350341796875,
-0.035552978515625,
-0.058349609375,
-0.01422119140625,
0.06439208984375,
-0.017852783203125,
0.0236663818359375,
0.062255859375,
0.0927734375,
-0.0165252685546875,
-0.045196533203125,
0.020965576171875,
0.0189056396484375,
0.0154266357421875,
0.061309814453125,
0.055511474609375,
-0.0694580078125,
0.036773681640625,
-0.053192138671875,
-0.00806427001953125,
-0.00199127197265625,
-0.061767578125,
-0.059417724609375,
-0.06451416015625,
-0.035797119140625,
-0.040771484375,
-0.01629638671875,
0.03912353515625,
0.0809326171875,
-0.050201416015625,
-0.01345062255859375,
-0.0247344970703125,
0.01287078857421875,
0.0024967193603515625,
-0.02374267578125,
0.035003662109375,
0.007114410400390625,
-0.05328369140625,
-0.00803375244140625,
0.0165863037109375,
0.02081298828125,
-0.027069091796875,
-0.0293121337890625,
-0.04248046875,
-0.02581787109375,
0.026123046875,
0.02197265625,
-0.050384521484375,
-0.019012451171875,
0.0053863525390625,
-0.00911712646484375,
0.0250091552734375,
0.034210205078125,
-0.051727294921875,
0.052581787109375,
0.057281494140625,
0.021942138671875,
0.059906005859375,
-0.025390625,
0.03070068359375,
-0.04376220703125,
0.036651611328125,
0.02081298828125,
0.00792694091796875,
0.033233642578125,
-0.040863037109375,
0.021148681640625,
0.0190887451171875,
-0.048126220703125,
-0.041015625,
0.01568603515625,
-0.07470703125,
-0.016082763671875,
0.0782470703125,
-0.019744873046875,
-0.0283050537109375,
0.005474090576171875,
-0.0318603515625,
0.036468505859375,
-0.027587890625,
0.0257720947265625,
0.036834716796875,
0.0025577545166015625,
-0.056396484375,
-0.026397705078125,
0.043243408203125,
0.02374267578125,
-0.051239013671875,
-0.051849365234375,
0.030242919921875,
0.025054931640625,
0.032073974609375,
0.051666259765625,
-0.00196075439453125,
0.0271759033203125,
0.01416778564453125,
0.01313018798828125,
-0.002727508544921875,
-0.0228424072265625,
-0.03509521484375,
-0.017852783203125,
-0.018157958984375,
-0.032958984375
]
] |
TheBloke/VicUnlocked-alpaca-65B-QLoRA-fp16 | 2023-06-05T00:10:34.000Z | [
"transformers",
"pytorch",
"llama",
"text-generation",
"dataset:gozfarb/ShareGPT_Vicuna_unfiltered",
"license:other",
"has_space",
"text-generation-inference",
"region:us"
] | text-generation | TheBloke | null | null | TheBloke/VicUnlocked-alpaca-65B-QLoRA-fp16 | 9 | 5,976 | transformers | 2023-05-30T00:10:50 | ---
inference: false
license: other
datasets:
- gozfarb/ShareGPT_Vicuna_unfiltered
---
<!-- header start -->
<div style="width: 100%;">
<img src="https://i.imgur.com/EBdldam.jpg" alt="TheBlokeAI" style="width: 100%; min-width: 400px; display: block; margin: auto;">
</div>
<div style="display: flex; justify-content: space-between; width: 100%;">
<div style="display: flex; flex-direction: column; align-items: flex-start;">
<p><a href="https://discord.gg/Jq4vkcDakD">Chat & support: my new Discord server</a></p>
</div>
<div style="display: flex; flex-direction: column; align-items: flex-end;">
<p><a href="https://www.patreon.com/TheBlokeAI">Want to contribute? TheBloke's Patreon page</a></p>
</div>
</div>
<!-- header end -->
# Aeala's VicUnlocked Alpaca 65B QLoRA GGML
These files are fp16 mergedmodel files for [Aeala's VicUnlocked Alpaca 65B QLoRA](https://huggingface.co/Aeala/VicUnlocked-alpaca-65b-QLoRA).
## Other repositories available
* [4-bit GPTQ models for GPU inference](https://huggingface.co/Aeala/VicUnlocked-alpaca-65b-4bit)
* [4-bit, 5-bit, and 8-bit GGML models for CPU+GPU inference](https://huggingface.co/TheBloke/VicUnlocked-alpaca-65B-QLoRA-GGML)
* [Original unquantised fp16 model in HF format](https://huggingface.co/TheBloke/VicUnlocked-alpaca-65B-QLoRA-fp16)
<!-- footer start -->
## Discord
For further support, and discussions on these models and AI in general, join us at:
[TheBloke AI's Discord server](https://discord.gg/Jq4vkcDakD)
## Thanks, and how to contribute.
Thanks to the [chirper.ai](https://chirper.ai) team!
I've had a lot of people ask if they can contribute. I enjoy providing models and helping people, and would love to be able to spend even more time doing it, as well as expanding into new projects like fine tuning/training.
If you're able and willing to contribute it will be most gratefully received and will help me to keep providing more models, and to start work on new AI projects.
Donaters will get priority support on any and all AI/LLM/model questions and requests, access to a private Discord room, plus other benefits.
* Patreon: https://patreon.com/TheBlokeAI
* Ko-Fi: https://ko-fi.com/TheBlokeAI
**Patreon special mentions**: Aemon Algiz, Dmitriy Samsonov, Nathan LeClaire, Trenton Dambrowitz, Mano Prime, David Flickinger, vamX, Nikolai Manek, senxiiz, Khalefa Al-Ahmad, Illia Dulskyi, Jonathan Leane, Talal Aujan, V. Lukas, Joseph William Delisle, Pyrater, Oscar Rangel, Lone Striker, Luke Pendergrass, Eugene Pentland, Sebastain Graf, Johann-Peter Hartman.
Thank you to all my generous patrons and donaters!
<!-- footer end -->
Patreon special mention: Jonathan Leane; Talal Aujan. Thank you both, and to all my other patrons and donaters.
# Original model card: Aeala's VicUnlocked Alpaca 65B QLoRA
## LoRA Info:
Please note that this is a highly experimental LoRA model. It may do some good stuff, it might do some undesirable stuff. Training is paused for now. Feel free to try it!~
**Important Note**: While this is trained on a cleaned ShareGPT dataset like Vicuna used, this was trained in the *Alpaca* format, so prompting should be something like:
```
### Instruction:
<prompt> (without the <>)
### Response:
```
Current upload: checkpoint of step 1200 in training.
## Benchmarks
**wikitext2:** Coming soon...
**ptb-new:** Coming soon...
**c4-new:** Coming soon...
| 3,412 | [
[
-0.0462646484375,
-0.060760498046875,
0.003330230712890625,
0.004825592041015625,
-0.0305328369140625,
-0.020477294921875,
0.0131683349609375,
-0.059906005859375,
0.05157470703125,
0.01398468017578125,
-0.058990478515625,
-0.0222015380859375,
-0.03741455078125,
-0.0022487640380859375,
-0.03753662109375,
0.066650390625,
0.0179595947265625,
-0.01168060302734375,
0.02264404296875,
-0.00852203369140625,
-0.05078125,
-0.0188140869140625,
-0.0714111328125,
-0.02752685546875,
0.043121337890625,
0.004878997802734375,
0.057647705078125,
0.051849365234375,
0.0226287841796875,
0.03167724609375,
-0.00959014892578125,
0.01242828369140625,
-0.0364990234375,
-0.01056671142578125,
0.0127716064453125,
-0.0220947265625,
-0.052886962890625,
-0.0048675537109375,
0.0257110595703125,
0.0248260498046875,
-0.0201873779296875,
0.0110626220703125,
0.0082550048828125,
0.03839111328125,
-0.037353515625,
0.0165252685546875,
-0.0362548828125,
-0.004909515380859375,
-0.0172119140625,
0.00995635986328125,
-0.00021123886108398438,
-0.035797119140625,
-0.0199127197265625,
-0.086669921875,
0.006488800048828125,
0.00547027587890625,
0.0814208984375,
0.015838623046875,
-0.01374053955078125,
-0.01154327392578125,
-0.04931640625,
0.043701171875,
-0.05303955078125,
0.028289794921875,
0.032989501953125,
0.0248260498046875,
-0.0100555419921875,
-0.05902099609375,
-0.052734375,
0.007228851318359375,
0.006549835205078125,
0.01947021484375,
-0.04119873046875,
-0.020111083984375,
-0.00487518310546875,
0.03607177734375,
-0.022979736328125,
0.003803253173828125,
-0.0489501953125,
-0.0120086669921875,
0.04193115234375,
-0.00858306884765625,
0.0281219482421875,
0.0208740234375,
-0.0238494873046875,
-0.0377197265625,
-0.04071044921875,
0.014373779296875,
0.0299072265625,
0.0285186767578125,
-0.07281494140625,
0.05267333984375,
-0.005481719970703125,
0.038299560546875,
0.0190887451171875,
0.003177642822265625,
0.016510009765625,
-0.045806884765625,
-0.0291595458984375,
-0.023406982421875,
0.07611083984375,
0.0306243896484375,
-0.01303863525390625,
0.0170135498046875,
0.0020599365234375,
0.0029277801513671875,
0.00292205810546875,
-0.06500244140625,
-0.01296234130859375,
0.0285186767578125,
-0.045562744140625,
-0.026763916015625,
0.0033397674560546875,
-0.0697021484375,
-0.0322265625,
-0.0028209686279296875,
0.033843994140625,
-0.035858154296875,
-0.032867431640625,
0.0198822021484375,
-0.0242462158203125,
0.044891357421875,
0.043212890625,
-0.060089111328125,
0.0180511474609375,
0.04168701171875,
0.0523681640625,
0.04364013671875,
-0.01514434814453125,
-0.03466796875,
0.00594329833984375,
-0.019378662109375,
0.046356201171875,
-0.014007568359375,
-0.040130615234375,
-0.00807952880859375,
0.015228271484375,
0.01279449462890625,
-0.0222320556640625,
0.039825439453125,
-0.01398468017578125,
0.00603485107421875,
-0.03948974609375,
-0.0221710205078125,
-0.00669097900390625,
0.00937652587890625,
-0.06072998046875,
0.046051025390625,
0.018646240234375,
-0.054443359375,
0.019500732421875,
-0.05426025390625,
-0.00380706787109375,
0.004718780517578125,
-0.0065460205078125,
-0.03204345703125,
-0.01352691650390625,
0.008697509765625,
0.01284027099609375,
-0.037322998046875,
-0.013885498046875,
-0.047332763671875,
-0.02069091796875,
0.030853271484375,
-0.04254150390625,
0.08087158203125,
0.0137786865234375,
-0.0286407470703125,
-0.0162811279296875,
-0.056488037109375,
-0.01239013671875,
0.035919189453125,
-0.022491455078125,
0.0109100341796875,
-0.01708984375,
0.0105133056640625,
-0.004177093505859375,
0.0275726318359375,
-0.04095458984375,
0.0261688232421875,
-0.0111541748046875,
0.0333251953125,
0.06256103515625,
-0.008331298828125,
0.01873779296875,
-0.053955078125,
0.0372314453125,
-0.01277923583984375,
0.048980712890625,
0.0104827880859375,
-0.052642822265625,
-0.06817626953125,
-0.0362548828125,
0.01447296142578125,
0.0276031494140625,
-0.042724609375,
0.046295166015625,
0.0117645263671875,
-0.05328369140625,
-0.055999755859375,
0.0005145072937011719,
0.0176849365234375,
0.023834228515625,
0.0321044921875,
-0.044189453125,
-0.039276123046875,
-0.057525634765625,
0.01593017578125,
-0.0377197265625,
-0.0012693405151367188,
0.0274810791015625,
0.04022216796875,
-0.03240966796875,
0.047027587890625,
-0.0241546630859375,
-0.0345458984375,
-0.007488250732421875,
-0.00789642333984375,
0.019683837890625,
0.057830810546875,
0.0628662109375,
-0.058441162109375,
-0.01898193359375,
0.01285552978515625,
-0.045623779296875,
-0.004878997802734375,
0.0038280487060546875,
-0.03155517578125,
0.0012655258178710938,
0.0120697021484375,
-0.07318115234375,
0.050933837890625,
0.04205322265625,
-0.04949951171875,
0.031982421875,
-0.0196685791015625,
0.01103973388671875,
-0.07916259765625,
0.0157623291015625,
-0.001983642578125,
-0.0159759521484375,
-0.032440185546875,
0.0016326904296875,
-0.0307159423828125,
-0.00696563720703125,
-0.034210205078125,
0.056243896484375,
-0.042633056640625,
0.0037994384765625,
-0.007457733154296875,
-0.00948333740234375,
0.0208587646484375,
0.028106689453125,
-0.0219879150390625,
0.039886474609375,
0.04345703125,
-0.03240966796875,
0.040924072265625,
0.04669189453125,
-0.0170135498046875,
0.0211029052734375,
-0.08453369140625,
0.003238677978515625,
-0.002704620361328125,
0.03167724609375,
-0.058197021484375,
-0.0197601318359375,
0.0557861328125,
-0.047515869140625,
0.040008544921875,
-0.0202178955078125,
-0.0174560546875,
-0.0263519287109375,
-0.03485107421875,
0.037017822265625,
0.05072021484375,
-0.04638671875,
0.043365478515625,
0.034271240234375,
0.003063201904296875,
-0.057891845703125,
-0.053192138671875,
-0.021820068359375,
-0.021514892578125,
-0.0227813720703125,
0.0239105224609375,
-0.023040771484375,
-0.02374267578125,
-0.00027370452880859375,
0.0117034912109375,
-0.01166534423828125,
0.004669189453125,
0.038177490234375,
0.0289154052734375,
-0.01140594482421875,
-0.024322509765625,
-0.00823974609375,
0.00782012939453125,
-0.0148468017578125,
-0.006450653076171875,
0.07122802734375,
-0.043487548828125,
-0.0303802490234375,
-0.069091796875,
0.03387451171875,
0.042633056640625,
-0.0226593017578125,
0.07049560546875,
0.0389404296875,
-0.0306549072265625,
-0.0002741813659667969,
-0.0379638671875,
-0.01015472412109375,
-0.0413818359375,
0.007656097412109375,
-0.00382232666015625,
-0.0499267578125,
0.0550537109375,
0.03717041015625,
0.0200042724609375,
0.04541015625,
0.037384033203125,
-0.0182342529296875,
0.05224609375,
0.057891845703125,
-0.0247802734375,
0.050201416015625,
-0.063232421875,
0.007152557373046875,
-0.042144775390625,
-0.036224365234375,
-0.049774169921875,
-0.0291290283203125,
-0.053314208984375,
-0.045440673828125,
0.0170440673828125,
0.01409149169921875,
-0.037750244140625,
0.0297393798828125,
-0.03277587890625,
0.0219573974609375,
0.01947021484375,
0.02410888671875,
0.0020275115966796875,
-0.005191802978515625,
0.0208587646484375,
0.025848388671875,
-0.05938720703125,
-0.01384735107421875,
0.05712890625,
0.040435791015625,
0.047698974609375,
0.02008056640625,
0.057891845703125,
0.0242462158203125,
0.0268707275390625,
-0.044921875,
0.0340576171875,
-0.00946044921875,
-0.054779052734375,
-0.02630615234375,
-0.01331329345703125,
-0.082763671875,
0.006866455078125,
-0.02679443359375,
-0.04608154296875,
0.052154541015625,
0.01457977294921875,
-0.033477783203125,
0.03076171875,
-0.0255126953125,
0.061492919921875,
-0.0070343017578125,
-0.02978515625,
-0.003692626953125,
-0.04461669921875,
0.0191497802734375,
0.00730133056640625,
0.029296875,
-0.0133209228515625,
0.0021305084228515625,
0.04644775390625,
-0.07861328125,
0.0927734375,
-0.00815582275390625,
-0.01529693603515625,
0.054779052734375,
-0.000949859619140625,
0.02752685546875,
0.010040283203125,
-0.00937652587890625,
0.01861572265625,
0.008514404296875,
-0.0352783203125,
-0.010498046875,
0.038299560546875,
-0.08673095703125,
-0.0280303955078125,
-0.02227783203125,
-0.0309600830078125,
0.0263214111328125,
0.0182342529296875,
0.04278564453125,
0.04608154296875,
-0.0198822021484375,
0.029693603515625,
0.01983642578125,
-0.0169219970703125,
0.040863037109375,
0.0120697021484375,
-0.0062408447265625,
-0.035400390625,
0.060302734375,
0.0004718303680419922,
0.004047393798828125,
0.0164794921875,
0.01261138916015625,
-0.027557373046875,
-0.0191192626953125,
-0.0305023193359375,
0.0460205078125,
-0.0458984375,
-0.0308685302734375,
-0.032958984375,
-0.01465606689453125,
-0.038055419921875,
-0.01453399658203125,
-0.04901123046875,
-0.041229248046875,
-0.04412841796875,
0.011322021484375,
0.046173095703125,
0.046051025390625,
-0.027801513671875,
0.0361328125,
-0.044952392578125,
0.0207672119140625,
0.0214080810546875,
0.01210784912109375,
-0.0018768310546875,
-0.050384521484375,
-0.00485992431640625,
0.02386474609375,
-0.03033447265625,
-0.059814453125,
0.053466796875,
0.018646240234375,
0.0377197265625,
0.03753662109375,
-0.00014519691467285156,
0.0640869140625,
-0.022796630859375,
0.052337646484375,
0.0350341796875,
-0.0579833984375,
0.038665771484375,
-0.039337158203125,
0.005115509033203125,
0.050079345703125,
0.03564453125,
-0.01058197021484375,
-0.0299530029296875,
-0.0517578125,
-0.053985595703125,
0.0257110595703125,
0.0183868408203125,
0.00901031494140625,
0.00028967857360839844,
0.039642333984375,
0.00832366943359375,
0.0037975311279296875,
-0.0772705078125,
-0.0291290283203125,
-0.032562255859375,
0.0012426376342773438,
0.01428985595703125,
0.0069122314453125,
-0.019256591796875,
-0.0361328125,
0.0828857421875,
-0.0186767578125,
0.052703857421875,
0.0171966552734375,
0.0328369140625,
-0.0131988525390625,
0.005168914794921875,
0.02239990234375,
0.0504150390625,
-0.0118865966796875,
-0.029876708984375,
-0.00435638427734375,
-0.02325439453125,
0.001499176025390625,
0.0201873779296875,
-0.0251312255859375,
-0.007274627685546875,
0.02227783203125,
0.06585693359375,
-0.00995635986328125,
-0.034423828125,
0.036407470703125,
-0.0234527587890625,
-0.01641845703125,
-0.027923583984375,
0.0162353515625,
0.01255035400390625,
0.04754638671875,
0.0130615234375,
-0.004547119140625,
0.01099395751953125,
-0.06353759765625,
-0.00574493408203125,
0.052886962890625,
-0.0142059326171875,
-0.0279388427734375,
0.06903076171875,
0.005268096923828125,
-0.03466796875,
0.041534423828125,
0.0033359527587890625,
-0.0275421142578125,
0.07647705078125,
0.049163818359375,
0.057830810546875,
-0.0172271728515625,
0.0182037353515625,
0.03668212890625,
0.0177001953125,
-0.002140045166015625,
0.0218505859375,
-0.00537109375,
-0.0594482421875,
0.0042877197265625,
-0.053131103515625,
-0.035980224609375,
0.0278167724609375,
-0.059478759765625,
0.0372314453125,
-0.0574951171875,
-0.027069091796875,
0.004093170166015625,
0.0032939910888671875,
-0.0517578125,
0.0223846435546875,
0.01302337646484375,
0.07049560546875,
-0.05908203125,
0.06243896484375,
0.028167724609375,
-0.044036865234375,
-0.06524658203125,
-0.0211639404296875,
-0.002208709716796875,
-0.0631103515625,
0.0063629150390625,
0.01203155517578125,
-0.01122283935546875,
-0.0120391845703125,
-0.06475830078125,
-0.05682373046875,
0.106689453125,
0.0190582275390625,
-0.04541015625,
-0.00118255615234375,
0.0008955001831054688,
0.0313720703125,
-0.03033447265625,
0.0269927978515625,
0.03070068359375,
0.0212554931640625,
0.005359649658203125,
-0.061676025390625,
-0.0006241798400878906,
-0.045318603515625,
-0.00775146484375,
0.00029468536376953125,
-0.10369873046875,
0.06549072265625,
-0.0111236572265625,
-0.004302978515625,
0.040283203125,
0.05841064453125,
0.03790283203125,
0.0106964111328125,
0.048004150390625,
0.03179931640625,
0.05487060546875,
-0.00007349252700805664,
0.08575439453125,
-0.020294189453125,
0.0176849365234375,
0.0555419921875,
0.002231597900390625,
0.05841064453125,
0.025299072265625,
-0.0207672119140625,
0.02655029296875,
0.046173095703125,
-0.0158538818359375,
0.0234527587890625,
-0.0135955810546875,
-0.0277862548828125,
-0.0026454925537109375,
-0.032958984375,
-0.0648193359375,
0.033447265625,
0.01462554931640625,
-0.0104827880859375,
0.0139312744140625,
-0.0126190185546875,
-0.00817108154296875,
-0.0244140625,
-0.0206298828125,
0.037109375,
0.024658203125,
-0.0170440673828125,
0.06591796875,
0.0020503997802734375,
0.059661865234375,
-0.063720703125,
-0.001453399658203125,
-0.044158935546875,
0.0196685791015625,
-0.003993988037109375,
-0.0322265625,
0.0124359130859375,
-0.001922607421875,
-0.005489349365234375,
-0.005062103271484375,
0.04864501953125,
-0.0146331787109375,
-0.044158935546875,
0.04998779296875,
0.0181427001953125,
0.02093505859375,
0.0308380126953125,
-0.0633544921875,
0.040863037109375,
0.0012998580932617188,
-0.0206451416015625,
0.0282440185546875,
0.0248260498046875,
0.00940704345703125,
0.060302734375,
0.044891357421875,
0.00650787353515625,
0.01409912109375,
-0.004261016845703125,
0.08758544921875,
-0.0289764404296875,
-0.03369140625,
-0.0562744140625,
0.038238525390625,
0.00899505615234375,
-0.025177001953125,
0.03961181640625,
0.046112060546875,
0.057220458984375,
-0.006778717041015625,
0.053009033203125,
-0.0237884521484375,
0.0161895751953125,
-0.025421142578125,
0.07183837890625,
-0.06964111328125,
0.00954437255859375,
-0.0164794921875,
-0.056976318359375,
0.00281524658203125,
0.065673828125,
0.03131103515625,
0.0141754150390625,
0.004787445068359375,
0.06536865234375,
-0.00394439697265625,
0.005855560302734375,
0.016448974609375,
0.033172607421875,
0.033233642578125,
0.05352783203125,
0.06329345703125,
-0.07177734375,
0.040374755859375,
-0.04254150390625,
-0.013885498046875,
-0.00782012939453125,
-0.0594482421875,
-0.045562744140625,
-0.0333251953125,
-0.03240966796875,
-0.036529541015625,
-0.00009369850158691406,
0.06536865234375,
0.052703857421875,
-0.04998779296875,
-0.035308837890625,
0.0026416778564453125,
-0.007503509521484375,
-0.00807952880859375,
-0.0135345458984375,
0.006801605224609375,
0.037353515625,
-0.056488037109375,
0.039642333984375,
-0.0082855224609375,
0.053009033203125,
-0.00222015380859375,
-0.015350341796875,
-0.0228118896484375,
0.01335906982421875,
0.0243988037109375,
0.06475830078125,
-0.038787841796875,
-0.01186370849609375,
-0.0151824951171875,
0.0097198486328125,
0.0160980224609375,
0.0268402099609375,
-0.0419921875,
-0.0004024505615234375,
0.0296630859375,
0.030303955078125,
0.048431396484375,
0.007022857666015625,
0.0259857177734375,
-0.0211944580078125,
0.04254150390625,
0.0039825439453125,
0.04864501953125,
0.020843505859375,
-0.0293731689453125,
0.040374755859375,
0.0182647705078125,
-0.046173095703125,
-0.062286376953125,
-0.01324462890625,
-0.0965576171875,
-0.01898193359375,
0.06866455078125,
0.0026187896728515625,
-0.03790283203125,
0.01335906982421875,
-0.0180511474609375,
0.044158935546875,
-0.0274505615234375,
0.0384521484375,
0.0239105224609375,
-0.0015964508056640625,
-0.016937255859375,
-0.04315185546875,
0.0250244140625,
0.0182037353515625,
-0.070556640625,
-0.0006947517395019531,
0.055206298828125,
0.0224456787109375,
0.0230255126953125,
0.07647705078125,
-0.0146026611328125,
0.03131103515625,
0.0128326416015625,
0.01279449462890625,
-0.00785064697265625,
-0.032073974609375,
-0.03094482421875,
0.004665374755859375,
-0.002105712890625,
-0.0287017822265625
]
] |
KoboldAI/fairseq-dense-2.7B | 2023-06-22T21:25:40.000Z | [
"transformers",
"pytorch",
"safetensors",
"xglm",
"text-generation",
"en",
"arxiv:2112.10684",
"endpoints_compatible",
"has_space",
"region:us"
] | text-generation | KoboldAI | null | null | KoboldAI/fairseq-dense-2.7B | 2 | 5,975 | transformers | 2022-03-02T23:29:04 | ---
language: en
---
This is a Hugging Face transformers-compatible conversion of the original dense 2.7B-parameter model from the paper "[Efficient Large Scale Language Modeling with Mixtures of Experts](https://arxiv.org/abs/2112.10684)" from Artetxe et al. Please refer to the original model card, which can be found at https://github.com/facebookresearch/fairseq/blob/main/examples/moe_lm/model_card.md.
| 408 | [
[
-0.048370361328125,
-0.05938720703125,
0.01439666748046875,
0.034271240234375,
-0.022308349609375,
-0.04632568359375,
-0.0210723876953125,
-0.027618408203125,
0.03240966796875,
0.06341552734375,
-0.053466796875,
-0.0083465576171875,
-0.037750244140625,
-0.03033447265625,
-0.03363037109375,
0.065673828125,
-0.004344940185546875,
0.008056640625,
-0.0167083740234375,
0.006214141845703125,
-0.0028076171875,
-0.0184478759765625,
-0.0667724609375,
-0.0306396484375,
0.0406494140625,
0.0159454345703125,
0.07135009765625,
0.03692626953125,
0.0295257568359375,
0.0193634033203125,
-0.0033664703369140625,
-0.0256195068359375,
-0.04473876953125,
-0.006847381591796875,
-0.01332855224609375,
-0.0285186767578125,
-0.06658935546875,
0.03271484375,
0.060516357421875,
0.053070068359375,
-0.049285888671875,
0.00949859619140625,
-0.004299163818359375,
0.040924072265625,
-0.004840850830078125,
0.007476806640625,
-0.049652099609375,
-0.004528045654296875,
-0.00952911376953125,
0.0059051513671875,
-0.055084228515625,
-0.004917144775390625,
-0.0092926025390625,
-0.0210113525390625,
0.0144500732421875,
-0.0014095306396484375,
0.0792236328125,
0.030975341796875,
-0.032440185546875,
0.01128387451171875,
-0.054290771484375,
0.040802001953125,
-0.0382080078125,
0.05474853515625,
0.00827789306640625,
0.054931640625,
-0.0245513916015625,
-0.054351806640625,
-0.051788330078125,
0.00887298583984375,
0.0200347900390625,
0.01568603515625,
-0.01369476318359375,
0.00470733642578125,
0.0169677734375,
0.038604736328125,
-0.02386474609375,
0.0012617111206054688,
-0.050323486328125,
-0.0106048583984375,
0.07049560546875,
-0.00006955862045288086,
0.01395416259765625,
-0.01030731201171875,
-0.05157470703125,
-0.021453857421875,
-0.0300445556640625,
-0.004974365234375,
0.019439697265625,
0.0175323486328125,
-0.040679931640625,
0.033294677734375,
-0.0102386474609375,
0.046478271484375,
0.01215362548828125,
0.00151824951171875,
0.01910400390625,
0.0162353515625,
-0.007965087890625,
-0.017852783203125,
0.045684814453125,
0.05517578125,
0.034393310546875,
-0.00229644775390625,
-0.01369476318359375,
-0.0225067138671875,
0.035369873046875,
-0.089111328125,
-0.038787841796875,
-0.0092926025390625,
-0.048553466796875,
-0.01444244384765625,
0.01276397705078125,
-0.052337646484375,
-0.006458282470703125,
-0.0240478515625,
0.01444244384765625,
-0.019622802734375,
-0.052764892578125,
0.0014200210571289062,
0.01280975341796875,
0.0430908203125,
0.0208740234375,
-0.04150390625,
0.01323699951171875,
0.0298919677734375,
0.044281005859375,
0.0022430419921875,
0.003772735595703125,
-0.0261077880859375,
0.007144927978515625,
-0.006908416748046875,
0.0545654296875,
-0.0266571044921875,
-0.0340576171875,
0.0016126632690429688,
0.00543975830078125,
0.0006532669067382812,
-0.047210693359375,
0.0648193359375,
-0.048370361328125,
0.00897216796875,
0.0070343017578125,
-0.021514892578125,
-0.0278778076171875,
0.012725830078125,
-0.07110595703125,
0.0980224609375,
0.05377197265625,
-0.03021240234375,
0.010589599609375,
-0.0263671875,
-0.00269317626953125,
0.0193328857421875,
0.0063018798828125,
-0.023193359375,
0.029205322265625,
-0.0030879974365234375,
0.032379150390625,
-0.0293426513671875,
0.0309906005859375,
-0.0606689453125,
-0.01220703125,
0.01959228515625,
-0.016632080078125,
0.079833984375,
0.0235595703125,
0.007381439208984375,
0.00411224365234375,
-0.05560302734375,
0.003749847412109375,
0.020782470703125,
-0.0140838623046875,
-0.015472412109375,
-0.015594482421875,
0.018218994140625,
0.0472412109375,
0.0275726318359375,
-0.0284423828125,
0.034027099609375,
-0.010711669921875,
0.00881195068359375,
0.0256195068359375,
-0.0190582275390625,
0.0240478515625,
-0.0165252685546875,
0.036224365234375,
0.0126495361328125,
0.01528167724609375,
0.0056915283203125,
-0.052734375,
-0.053955078125,
-0.042449951171875,
0.0212554931640625,
0.01641845703125,
-0.048370361328125,
0.046539306640625,
-0.0167999267578125,
-0.080322265625,
-0.04620361328125,
0.006317138671875,
-0.006069183349609375,
0.0273590087890625,
0.0185699462890625,
-0.022308349609375,
-0.03070068359375,
-0.0880126953125,
-0.00521087646484375,
-0.0262451171875,
-0.00905609130859375,
0.024383544921875,
0.00997161865234375,
-0.05126953125,
0.07086181640625,
-0.0309295654296875,
-0.01338958740234375,
-0.0175018310546875,
-0.012603759765625,
0.020782470703125,
0.061767578125,
0.0645751953125,
-0.036895751953125,
-0.04302978515625,
-0.01282501220703125,
-0.04736328125,
-0.03314208984375,
0.01523590087890625,
-0.03955078125,
0.0135040283203125,
0.0579833984375,
-0.05999755859375,
0.0226593017578125,
0.0684814453125,
-0.027862548828125,
0.0281524658203125,
0.0073394775390625,
-0.018890380859375,
-0.09832763671875,
0.01219940185546875,
0.0086822509765625,
-0.0279083251953125,
-0.0455322265625,
0.042144775390625,
0.0234832763671875,
-0.00518798828125,
-0.032958984375,
0.0704345703125,
-0.0343017578125,
0.0164337158203125,
-0.0145111083984375,
0.0121917724609375,
-0.01348876953125,
0.0243682861328125,
-0.01020050048828125,
0.027618408203125,
0.0614013671875,
-0.0295257568359375,
0.05133056640625,
0.035186767578125,
0.0028781890869140625,
0.06744384765625,
-0.05670166015625,
0.019805908203125,
-0.01549530029296875,
0.0262298583984375,
-0.0640869140625,
-0.03192138671875,
0.0220794677734375,
-0.0151214599609375,
0.032196044921875,
-0.00464630126953125,
-0.04095458984375,
-0.0288848876953125,
-0.006160736083984375,
0.05255126953125,
0.0693359375,
-0.046966552734375,
0.08380126953125,
0.036895751953125,
-0.03802490234375,
0.01325225830078125,
-0.046600341796875,
0.0038814544677734375,
-0.01241302490234375,
-0.06317138671875,
0.0218658447265625,
-0.014923095703125,
-0.0120697021484375,
0.0013580322265625,
0.019439697265625,
-0.01250457763671875,
-0.0141754150390625,
0.014556884765625,
0.008209228515625,
-0.037017822265625,
-0.0283660888671875,
0.034393310546875,
-0.0084228515625,
0.0286865234375,
0.018829345703125,
0.04443359375,
0.005847930908203125,
0.0013628005981445312,
-0.038482666015625,
0.040008544921875,
0.0457763671875,
-0.00406646728515625,
0.061553955078125,
0.0638427734375,
-0.040191650390625,
-0.0243072509765625,
-0.054107666015625,
-0.0247802734375,
-0.03741455078125,
0.0295867919921875,
-0.040252685546875,
-0.041412353515625,
0.060791015625,
-0.0010976791381835938,
-0.0254974365234375,
0.054229736328125,
0.04241943359375,
0.01462554931640625,
0.0831298828125,
0.040069580078125,
0.0117034912109375,
0.0335693359375,
-0.0001983642578125,
0.013458251953125,
-0.0643310546875,
-0.03338623046875,
-0.0199127197265625,
-0.034576416015625,
-0.03411865234375,
-0.042938232421875,
0.0150909423828125,
0.0340576171875,
-0.0037708282470703125,
0.0321044921875,
-0.01206207275390625,
0.0091400146484375,
0.041778564453125,
0.021636962890625,
0.0022258758544921875,
0.00969696044921875,
0.01708984375,
-0.009002685546875,
-0.0401611328125,
-0.020050048828125,
0.047119140625,
0.0523681640625,
0.0577392578125,
0.0154571533203125,
0.035186767578125,
-0.020782470703125,
0.046600341796875,
-0.051513671875,
0.055877685546875,
-0.0074310302734375,
-0.07464599609375,
0.0126495361328125,
-0.038238525390625,
-0.046234130859375,
0.01739501953125,
-0.0225067138671875,
-0.051300048828125,
0.0005249977111816406,
0.003925323486328125,
-0.00612640380859375,
0.0256500244140625,
-0.060211181640625,
0.08099365234375,
0.017333984375,
-0.004871368408203125,
-0.0175323486328125,
-0.0284423828125,
0.04296875,
0.0031147003173828125,
-0.010955810546875,
-0.015869140625,
0.0037021636962890625,
0.050994873046875,
-0.01302337646484375,
0.057098388671875,
-0.029998779296875,
-0.03204345703125,
0.0355224609375,
0.0173492431640625,
0.03289794921875,
0.006649017333984375,
-0.0209503173828125,
0.03533935546875,
-0.0071258544921875,
-0.04638671875,
-0.048248291015625,
0.058929443359375,
-0.058349609375,
-0.03302001953125,
-0.000370025634765625,
-0.05657958984375,
-0.009796142578125,
0.01146697998046875,
-0.0006923675537109375,
0.031768798828125,
-0.012725830078125,
0.042236328125,
0.030548095703125,
0.009765625,
0.01050567626953125,
0.036651611328125,
-0.03509521484375,
-0.0212249755859375,
0.040618896484375,
-0.01523590087890625,
0.0210723876953125,
-0.0006227493286132812,
0.006591796875,
-0.0294036865234375,
-0.02569580078125,
-0.044281005859375,
0.036163330078125,
-0.04132080078125,
-0.018218994140625,
-0.045196533203125,
-0.045166015625,
-0.029388427734375,
-0.0100555419921875,
-0.040740966796875,
-0.0060272216796875,
-0.025146484375,
0.0027332305908203125,
0.0278472900390625,
0.044830322265625,
0.0092010498046875,
0.056854248046875,
-0.065185546875,
0.01250457763671875,
0.0256195068359375,
0.056549072265625,
-0.00510406494140625,
-0.07073974609375,
-0.014129638671875,
0.002452850341796875,
-0.01389312744140625,
-0.07757568359375,
0.0188751220703125,
0.0121917724609375,
0.045196533203125,
0.037841796875,
0.0017185211181640625,
0.034942626953125,
-0.0278167724609375,
0.0430908203125,
0.00464630126953125,
-0.050750732421875,
-0.005435943603515625,
-0.035675048828125,
0.001789093017578125,
0.0408935546875,
0.0205535888671875,
-0.0311279296875,
-0.020660400390625,
-0.0650634765625,
-0.062225341796875,
0.06256103515625,
-0.005664825439453125,
0.032440185546875,
0.00946044921875,
0.04266357421875,
0.019287109375,
0.006595611572265625,
-0.0472412109375,
-0.0245208740234375,
-0.0038604736328125,
-0.04718017578125,
0.006885528564453125,
-0.051666259765625,
0.0140228271484375,
-0.02252197265625,
0.0552978515625,
-0.01241302490234375,
0.0266876220703125,
-0.00992584228515625,
-0.0031585693359375,
-0.0148468017578125,
-0.0242767333984375,
0.04473876953125,
-0.005279541015625,
-0.0183868408203125,
0.0090179443359375,
0.00518798828125,
-0.032684326171875,
-0.031585693359375,
0.027923583984375,
-0.0118560791015625,
-0.0055694580078125,
0.002239227294921875,
0.06884765625,
0.0290374755859375,
-0.03216552734375,
0.04302978515625,
0.013092041015625,
-0.00933837890625,
-0.043609619140625,
0.0067291259765625,
0.0185394287109375,
0.0241851806640625,
0.021209716796875,
0.0146026611328125,
0.00870513916015625,
-0.0311126708984375,
0.035064697265625,
0.0343017578125,
-0.0445556640625,
-0.048828125,
0.057586669921875,
0.0399169921875,
-0.04156494140625,
0.044586181640625,
-0.031524658203125,
-0.020660400390625,
0.01277923583984375,
0.043609619140625,
0.060577392578125,
-0.050445556640625,
0.0089263916015625,
0.0477294921875,
0.0210418701171875,
0.003154754638671875,
0.0167236328125,
0.007778167724609375,
-0.040985107421875,
-0.03216552734375,
-0.057037353515625,
-0.0169219970703125,
0.00608062744140625,
-0.06988525390625,
0.043426513671875,
-0.02734375,
0.007015228271484375,
-0.0252227783203125,
-0.03204345703125,
-0.03997802734375,
0.0223541259765625,
0.0208892822265625,
0.0887451171875,
-0.0675048828125,
0.07080078125,
0.045257568359375,
-0.00891876220703125,
-0.07220458984375,
-0.00018346309661865234,
-0.01824951171875,
-0.0784912109375,
0.016998291015625,
0.0131378173828125,
0.018402099609375,
-0.0005698204040527344,
-0.046600341796875,
-0.057708740234375,
0.04425048828125,
0.04461669921875,
-0.040374755859375,
-0.01120758056640625,
-0.0037860870361328125,
0.035919189453125,
-0.029815673828125,
0.038421630859375,
0.038421630859375,
0.0302886962890625,
0.019012451171875,
-0.0682373046875,
0.0223388671875,
-0.044342041015625,
0.0223388671875,
0.013214111328125,
-0.0653076171875,
0.07080078125,
0.0073699951171875,
0.007610321044921875,
0.0189361572265625,
0.08966064453125,
0.0297393798828125,
0.00658416748046875,
0.0489501953125,
0.045501708984375,
0.0129547119140625,
-0.00989532470703125,
0.0614013671875,
-0.0276336669921875,
0.042724609375,
0.045867919921875,
-0.0229339599609375,
0.0758056640625,
0.046966552734375,
-0.010650634765625,
0.055694580078125,
0.023681640625,
-0.006847381591796875,
0.01227569580078125,
-0.004730224609375,
0.0003304481506347656,
-0.0333251953125,
0.0024051666259765625,
-0.03961181640625,
0.039703369140625,
0.0275421142578125,
-0.0169677734375,
-0.0213470458984375,
-0.019927978515625,
0.0082855224609375,
0.008514404296875,
-0.0263214111328125,
0.043853759765625,
0.0102081298828125,
-0.0430908203125,
0.035369873046875,
0.0030765533447265625,
0.0426025390625,
-0.0278472900390625,
-0.0016164779663085938,
0.004241943359375,
0.01488494873046875,
-0.00521087646484375,
-0.0640869140625,
0.03985595703125,
-0.018798828125,
-0.01323699951171875,
-0.0142364501953125,
0.0341796875,
-0.06298828125,
-0.058074951171875,
0.035369873046875,
0.01503753662109375,
0.03570556640625,
-0.0243988037109375,
-0.06463623046875,
0.017303466796875,
-0.005695343017578125,
-0.038604736328125,
0.00539398193359375,
0.0301361083984375,
0.0015239715576171875,
0.033111572265625,
0.015960693359375,
-0.0017518997192382812,
0.01515960693359375,
0.0005326271057128906,
0.051300048828125,
-0.050567626953125,
-0.047943115234375,
-0.030517578125,
0.065673828125,
-0.015716552734375,
-0.037841796875,
0.0360107421875,
0.040740966796875,
0.054931640625,
-0.02960205078125,
0.0203704833984375,
-0.00612640380859375,
0.032867431640625,
-0.039306640625,
0.059539794921875,
-0.06365966796875,
-0.0300750732421875,
-0.0164337158203125,
-0.09490966796875,
-0.014251708984375,
0.0516357421875,
0.00997161865234375,
0.045806884765625,
0.039154052734375,
0.06146240234375,
-0.015594482421875,
-0.006839752197265625,
0.03143310546875,
0.03912353515625,
0.0156707763671875,
0.017822265625,
0.028472900390625,
-0.0501708984375,
0.031646728515625,
-0.01366424560546875,
-0.0191192626953125,
-0.046478271484375,
-0.067138671875,
-0.081787109375,
-0.0614013671875,
-0.04705810546875,
-0.032928466796875,
-0.04266357421875,
0.06317138671875,
0.07257080078125,
-0.048736572265625,
0.0019321441650390625,
0.016082763671875,
-0.027099609375,
0.01023101806640625,
-0.016998291015625,
-0.00836181640625,
0.0178375244140625,
-0.07489013671875,
0.0103302001953125,
-0.003253936767578125,
0.001781463623046875,
-0.034698486328125,
-0.01279449462890625,
0.0225982666015625,
0.0223236083984375,
0.0369873046875,
-0.005847930908203125,
-0.057952880859375,
-0.0251617431640625,
-0.01020050048828125,
-0.02227783203125,
-0.004810333251953125,
0.033111572265625,
-0.0252838134765625,
0.0016355514526367188,
0.030731201171875,
0.0377197265625,
0.040985107421875,
0.00392913818359375,
0.031280517578125,
-0.059661865234375,
0.040618896484375,
-0.00298309326171875,
0.045166015625,
0.039337158203125,
-0.009674072265625,
0.01116943359375,
0.02069091796875,
-0.0241546630859375,
-0.0704345703125,
0.032012939453125,
-0.13623046875,
0.019012451171875,
0.09930419921875,
0.0203704833984375,
-0.047821044921875,
0.024658203125,
-0.0452880859375,
0.01934814453125,
-0.034820556640625,
0.04083251953125,
0.03790283203125,
0.03790283203125,
-0.04718017578125,
-0.0265655517578125,
0.0017023086547851562,
0.023681640625,
-0.03045654296875,
-0.005794525146484375,
0.01163482666015625,
0.01476287841796875,
0.019134521484375,
0.025054931640625,
-0.0287628173828125,
0.004848480224609375,
0.005340576171875,
0.06243896484375,
-0.007579803466796875,
-0.0226287841796875,
-0.017333984375,
0.004486083984375,
0.0139007568359375,
0.0202178955078125
]
] |
vennify/t5-base-grammar-correction | 2022-01-14T16:35:23.000Z | [
"transformers",
"pytorch",
"t5",
"text2text-generation",
"grammar",
"en",
"dataset:jfleg",
"arxiv:1702.04066",
"license:cc-by-nc-sa-4.0",
"autotrain_compatible",
"endpoints_compatible",
"has_space",
"text-generation-inference",
"region:us"
] | text2text-generation | vennify | null | null | vennify/t5-base-grammar-correction | 84 | 5,973 | transformers | 2022-03-02T23:29:05 | ---
language: en
tags:
- grammar
- text2text-generation
license: cc-by-nc-sa-4.0
datasets:
- jfleg
---
# T5 Grammar Correction
This model generates a revised version of inputted text with the goal of containing fewer grammatical errors.
It was trained with [Happy Transformer](https://github.com/EricFillion/happy-transformer)
using a dataset called [JFLEG](https://arxiv.org/abs/1702.04066). Here's a [full article](https://www.vennify.ai/fine-tune-grammar-correction/) on how to train a similar model.
## Usage
`pip install happytransformer `
```python
from happytransformer import HappyTextToText, TTSettings
happy_tt = HappyTextToText("T5", "vennify/t5-base-grammar-correction")
args = TTSettings(num_beams=5, min_length=1)
# Add the prefix "grammar: " before each input
result = happy_tt.generate_text("grammar: This sentences has has bads grammar.", args=args)
print(result.text) # This sentence has bad grammar.
``` | 939 | [
[
0.00917816162109375,
-0.056365966796875,
0.0191497802734375,
0.018768310546875,
-0.01059722900390625,
-0.0218048095703125,
-0.0196075439453125,
-0.004638671875,
-0.0066375732421875,
0.0292205810546875,
-0.06402587890625,
-0.04437255859375,
-0.043731689453125,
0.04296875,
-0.027862548828125,
0.077880859375,
0.0111846923828125,
0.02154541015625,
-0.0109100341796875,
0.0282745361328125,
-0.03167724609375,
-0.0186767578125,
-0.06365966796875,
-0.037994384765625,
0.034149169921875,
0.041046142578125,
0.0391845703125,
0.0447998046875,
0.040679931640625,
0.0280609130859375,
-0.032958984375,
0.0091094970703125,
-0.046905517578125,
-0.01152801513671875,
-0.01739501953125,
-0.049224853515625,
-0.0301666259765625,
-0.031890869140625,
0.033538818359375,
0.024932861328125,
0.0013332366943359375,
0.01129150390625,
-0.005001068115234375,
0.0195770263671875,
-0.03948974609375,
0.0189208984375,
-0.0309906005859375,
0.0113677978515625,
-0.016937255859375,
0.0092315673828125,
-0.05072021484375,
-0.036651611328125,
-0.0208892822265625,
-0.04827880859375,
0.0183868408203125,
-0.024169921875,
0.1021728515625,
0.0438232421875,
-0.035247802734375,
0.00226593017578125,
-0.043701171875,
0.060546875,
-0.07232666015625,
0.0225372314453125,
0.038665771484375,
0.00850677490234375,
-0.02215576171875,
-0.07684326171875,
-0.040924072265625,
0.0007467269897460938,
-0.01318359375,
0.00907135009765625,
-0.0139617919921875,
0.01180267333984375,
0.0276336669921875,
0.0240936279296875,
-0.030670166015625,
-0.02337646484375,
-0.038238525390625,
-0.0183868408203125,
0.042877197265625,
0.0014142990112304688,
0.00653076171875,
-0.0270538330078125,
-0.00466156005859375,
-0.0151824951171875,
-0.0318603515625,
-0.000027000904083251953,
0.00568389892578125,
0.025726318359375,
-0.0081024169921875,
0.058441162109375,
-0.0198516845703125,
0.057220458984375,
0.0298004150390625,
-0.027008056640625,
0.052642822265625,
-0.0304107666015625,
-0.0288238525390625,
-0.0008187294006347656,
0.04827880859375,
0.04547119140625,
0.05224609375,
-0.003025054931640625,
-0.0278778076171875,
0.0167083740234375,
0.03271484375,
-0.046630859375,
-0.01593017578125,
0.0010938644409179688,
-0.0165557861328125,
-0.04974365234375,
-0.006366729736328125,
-0.0250244140625,
-0.0188446044921875,
-0.0185089111328125,
0.040740966796875,
-0.048553466796875,
0.01812744140625,
0.0256195068359375,
-0.0134735107421875,
0.0007295608520507812,
0.014862060546875,
-0.0714111328125,
0.01258087158203125,
0.036285400390625,
0.034271240234375,
0.042388916015625,
-0.0285797119140625,
-0.054229736328125,
-0.000720977783203125,
-0.01407623291015625,
0.07257080078125,
-0.01873779296875,
-0.027374267578125,
0.0034809112548828125,
0.0165252685546875,
-0.0110015869140625,
-0.04150390625,
0.07275390625,
-0.01299285888671875,
0.055023193359375,
0.00047779083251953125,
-0.04400634765625,
-0.0126953125,
0.038665771484375,
-0.0391845703125,
0.06964111328125,
0.0208282470703125,
-0.05511474609375,
0.0182647705078125,
-0.04620361328125,
-0.051300048828125,
-0.023284912109375,
0.0297088623046875,
-0.071533203125,
0.010284423828125,
0.02777099609375,
0.0177154541015625,
0.0208587646484375,
0.0169219970703125,
-0.0196685791015625,
-0.028472900390625,
0.00798797607421875,
-0.015777587890625,
0.07806396484375,
0.020751953125,
-0.00562286376953125,
0.033935546875,
-0.068115234375,
0.0010747909545898438,
-0.0035953521728515625,
-0.0406494140625,
-0.004764556884765625,
0.00074005126953125,
0.01415252685546875,
0.0096893310546875,
0.0200958251953125,
-0.053985595703125,
0.0195159912109375,
-0.04949951171875,
0.02630615234375,
0.043792724609375,
-0.005306243896484375,
0.016845703125,
-0.052581787109375,
0.004131317138671875,
0.00902557373046875,
0.0216217041015625,
-0.01116180419921875,
-0.0188446044921875,
-0.08380126953125,
-0.016937255859375,
0.035552978515625,
0.04473876953125,
-0.05535888671875,
0.0380859375,
-0.02166748046875,
-0.0360107421875,
-0.03936767578125,
-0.00217437744140625,
0.0149078369140625,
0.028411865234375,
0.036865234375,
0.007015228271484375,
-0.07855224609375,
-0.050567626953125,
-0.040008544921875,
-0.0073394775390625,
0.00959014892578125,
-0.0269622802734375,
0.03271484375,
-0.0264434814453125,
0.0653076171875,
-0.0242462158203125,
-0.0245361328125,
-0.01824951171875,
0.002651214599609375,
0.0261383056640625,
0.035797119140625,
0.040069580078125,
-0.034698486328125,
-0.038970947265625,
-0.0095672607421875,
-0.039337158203125,
-0.0188140869140625,
-0.00738525390625,
0.0063934326171875,
0.0178985595703125,
0.0229644775390625,
-0.0537109375,
0.02032470703125,
0.020416259765625,
-0.0538330078125,
0.03692626953125,
-0.0025424957275390625,
0.00324249267578125,
-0.103271484375,
-0.0097808837890625,
0.00864410400390625,
-0.060455322265625,
-0.042633056640625,
0.0110931396484375,
0.01143646240234375,
0.017303466796875,
-0.044921875,
0.037445068359375,
-0.00746917724609375,
0.0287017822265625,
-0.01345062255859375,
-0.0132904052734375,
-0.005306243896484375,
0.02178955078125,
-0.022308349609375,
0.0445556640625,
0.033111572265625,
-0.04022216796875,
0.0322265625,
0.03076171875,
0.01412200927734375,
0.0193328857421875,
-0.0360107421875,
-0.00583648681640625,
-0.01317596435546875,
-0.0037097930908203125,
-0.07073974609375,
-0.033355712890625,
0.029327392578125,
-0.034912109375,
0.01340484619140625,
0.0018281936645507812,
-0.043731689453125,
-0.029144287109375,
-0.03302001953125,
0.0008511543273925781,
0.044097900390625,
-0.042022705078125,
0.02398681640625,
-0.004241943359375,
-0.00739288330078125,
-0.0460205078125,
-0.05712890625,
0.0016813278198242188,
-0.00942230224609375,
-0.0518798828125,
0.0292205810546875,
-0.004241943359375,
-0.01318359375,
0.00537872314453125,
-0.0007219314575195312,
-0.01020050048828125,
0.01103973388671875,
-0.00148773193359375,
0.01203155517578125,
-0.015228271484375,
0.011749267578125,
-0.00366973876953125,
-0.0079345703125,
0.011199951171875,
-0.00528717041015625,
0.044342041015625,
-0.019287109375,
0.00164794921875,
-0.044525146484375,
-0.01568603515625,
0.032623291015625,
-0.008758544921875,
0.045562744140625,
0.042999267578125,
-0.045257568359375,
-0.01152801513671875,
-0.0193023681640625,
-0.0172576904296875,
-0.037353515625,
0.036956787109375,
-0.0196380615234375,
-0.05035400390625,
0.0224609375,
0.01490020751953125,
0.0034427642822265625,
0.04443359375,
0.034332275390625,
-0.01444244384765625,
0.06097412109375,
0.0447998046875,
0.00604248046875,
0.06298828125,
0.0013551712036132812,
0.017730712890625,
-0.01548004150390625,
-0.0260772705078125,
-0.035675048828125,
-0.00766754150390625,
-0.045013427734375,
-0.004146575927734375,
0.0006899833679199219,
0.007694244384765625,
-0.034515380859375,
0.022003173828125,
-0.022308349609375,
0.03350830078125,
0.049072265625,
0.016021728515625,
0.019287109375,
0.01995849609375,
0.0010519027709960938,
-0.01218414306640625,
-0.051300048828125,
-0.049591064453125,
0.06951904296875,
0.023345947265625,
0.06243896484375,
-0.003894805908203125,
0.079345703125,
0.0032196044921875,
0.021484375,
-0.06317138671875,
0.0310821533203125,
-0.051849365234375,
-0.028656005859375,
0.004673004150390625,
-0.024383544921875,
-0.053741455078125,
-0.019378662109375,
-0.0299835205078125,
-0.042144775390625,
-0.009979248046875,
0.029541015625,
-0.0267333984375,
0.006435394287109375,
-0.05999755859375,
0.08013916015625,
0.00489044189453125,
-0.041839599609375,
0.0010957717895507812,
-0.038421630859375,
0.0164794921875,
0.0242767333984375,
-0.0005669593811035156,
0.0234527587890625,
0.025177001953125,
0.035247802734375,
-0.036590576171875,
0.075439453125,
-0.003025054931640625,
-0.003993988037109375,
0.0159759521484375,
-0.010772705078125,
0.03326416015625,
-0.0038051605224609375,
-0.0175323486328125,
-0.0079803466796875,
-0.0131378173828125,
-0.022674560546875,
-0.0224151611328125,
0.033172607421875,
-0.06884765625,
-0.0236053466796875,
-0.0369873046875,
-0.03118896484375,
0.0096435546875,
0.042144775390625,
0.040130615234375,
0.022796630859375,
-0.01136016845703125,
0.0182647705078125,
0.034454345703125,
0.002750396728515625,
0.04022216796875,
0.0008392333984375,
-0.03656005859375,
-0.05218505859375,
0.04608154296875,
0.002941131591796875,
0.0008540153503417969,
0.033050537109375,
0.0156707763671875,
-0.0296630859375,
-0.01503753662109375,
-0.0106201171875,
0.02838134765625,
-0.051422119140625,
-0.0309600830078125,
-0.02911376953125,
-0.0350341796875,
-0.051055908203125,
0.00452423095703125,
-0.026275634765625,
-0.040130615234375,
-0.04388427734375,
-0.0116119384765625,
0.03411865234375,
0.0660400390625,
-0.01064300537109375,
0.061920166015625,
-0.053375244140625,
0.01910400390625,
-0.0024929046630859375,
0.02166748046875,
-0.052398681640625,
-0.045440673828125,
-0.015869140625,
0.02166748046875,
-0.04913330078125,
-0.05548095703125,
0.037628173828125,
0.0205230712890625,
0.0190887451171875,
0.0071258544921875,
-0.00443267822265625,
0.060699462890625,
-0.035430908203125,
0.0709228515625,
0.00884246826171875,
-0.0902099609375,
0.047088623046875,
-0.0237884521484375,
0.037384033203125,
0.0195465087890625,
-0.01044464111328125,
-0.05584716796875,
-0.03662109375,
-0.055908203125,
-0.056549072265625,
0.06585693359375,
0.0262908935546875,
0.008056640625,
-0.01079559326171875,
0.03662109375,
-0.0013027191162109375,
0.0019121170043945312,
-0.0928955078125,
-0.0147552490234375,
-0.056732177734375,
-0.058441162109375,
-0.007354736328125,
-0.0294952392578125,
-0.0032253265380859375,
-0.0128631591796875,
0.09271240234375,
0.0078277587890625,
0.0287017822265625,
0.017364501953125,
-0.01026153564453125,
-0.01222991943359375,
0.03863525390625,
0.061737060546875,
0.01690673828125,
-0.00627899169921875,
0.0177154541015625,
0.0251617431640625,
-0.02056884765625,
0.004299163818359375,
0.014434814453125,
-0.0231475830078125,
0.0229034423828125,
0.042633056640625,
0.0670166015625,
-0.005817413330078125,
-0.0222625732421875,
0.033172607421875,
-0.01230621337890625,
-0.0360107421875,
-0.0382080078125,
0.0298614501953125,
0.01027679443359375,
0.0172271728515625,
-0.002178192138671875,
0.0443115234375,
0.004547119140625,
-0.01666259765625,
0.0206756591796875,
-0.00612640380859375,
-0.0169830322265625,
-0.020751953125,
0.051666259765625,
0.013427734375,
-0.05206298828125,
0.052093505859375,
-0.024505615234375,
-0.040130615234375,
0.03369140625,
0.07183837890625,
0.05731201171875,
-0.013641357421875,
0.01520538330078125,
0.052001953125,
0.0289306640625,
-0.0244598388671875,
0.05224609375,
0.03070068359375,
-0.054718017578125,
-0.04827880859375,
-0.03326416015625,
-0.00787353515625,
0.0227813720703125,
-0.0256195068359375,
0.056427001953125,
-0.04083251953125,
-0.0264129638671875,
-0.01502227783203125,
-0.00567626953125,
-0.035369873046875,
0.041412353515625,
0.025054931640625,
0.065673828125,
-0.059234619140625,
0.04583740234375,
0.0877685546875,
-0.0284271240234375,
-0.06842041015625,
0.0155487060546875,
-0.00922393798828125,
-0.032806396484375,
0.033111572265625,
0.0172119140625,
-0.00370025634765625,
0.0162353515625,
-0.04901123046875,
-0.06146240234375,
0.056976318359375,
0.054534912109375,
-0.049957275390625,
-0.0035419464111328125,
0.01678466796875,
0.05712890625,
0.0015811920166015625,
0.021514892578125,
0.01904296875,
0.038665771484375,
-0.01297760009765625,
-0.06591796875,
-0.01287078857421875,
-0.04248046875,
0.0008363723754882812,
-0.00847625732421875,
-0.03656005859375,
0.07855224609375,
-0.0122528076171875,
-0.00604248046875,
0.0216827392578125,
0.0518798828125,
0.012908935546875,
0.01154327392578125,
0.030181884765625,
0.05181884765625,
0.0526123046875,
-0.02197265625,
0.06451416015625,
-0.013916015625,
0.07830810546875,
0.078369140625,
0.008056640625,
0.049407958984375,
0.04718017578125,
-0.00266265869140625,
0.046142578125,
0.05682373046875,
-0.007320404052734375,
0.054107666015625,
0.00025010108947753906,
-0.0164642333984375,
-0.0259857177734375,
0.00603485107421875,
-0.01444244384765625,
0.0274200439453125,
0.0057830810546875,
-0.0306854248046875,
-0.034759521484375,
0.0007581710815429688,
0.0263671875,
0.0131988525390625,
-0.0241851806640625,
0.0555419921875,
-0.01319122314453125,
-0.052276611328125,
0.061798095703125,
0.021759033203125,
0.05352783203125,
-0.0280914306640625,
0.001224517822265625,
-0.02288818359375,
0.05908203125,
-0.039276123046875,
-0.045745849609375,
0.037139892578125,
0.01409149169921875,
-0.0194854736328125,
-0.017730712890625,
0.06109619140625,
-0.0418701171875,
-0.05694580078125,
0.00543975830078125,
0.01030731201171875,
-0.007717132568359375,
0.0214385986328125,
-0.0333251953125,
-0.00377655029296875,
0.0272979736328125,
-0.0170135498046875,
-0.031585693359375,
0.0266265869140625,
0.0066375732421875,
0.040679931640625,
0.0213775634765625,
-0.00579071044921875,
0.0212249755859375,
0.02032470703125,
0.054595947265625,
-0.046112060546875,
-0.04095458984375,
-0.0771484375,
0.065185546875,
-0.01531219482421875,
-0.03863525390625,
0.045196533203125,
0.04742431640625,
0.08294677734375,
-0.04266357421875,
0.07257080078125,
-0.01837158203125,
0.031890869140625,
-0.042449951171875,
0.040069580078125,
-0.04443359375,
0.005649566650390625,
-0.01445770263671875,
-0.05572509765625,
-0.06060791015625,
0.09161376953125,
-0.0257568359375,
-0.003047943115234375,
0.05584716796875,
0.0791015625,
-0.0251617431640625,
0.0007967948913574219,
0.016754150390625,
0.04736328125,
0.0197296142578125,
0.02874755859375,
0.053009033203125,
-0.06060791015625,
0.0293426513671875,
-0.0369873046875,
0.0142364501953125,
-0.020843505859375,
-0.05364990234375,
-0.06781005859375,
-0.01058197021484375,
-0.0312347412109375,
-0.043212890625,
0.033355712890625,
0.09234619140625,
0.031158447265625,
-0.06842041015625,
-0.01534271240234375,
-0.0318603515625,
0.01531219482421875,
-0.017059326171875,
-0.0192413330078125,
-0.00025725364685058594,
-0.052001953125,
-0.069580078125,
0.017333984375,
-0.0282440185546875,
0.0026493072509765625,
0.0288238525390625,
-0.0006079673767089844,
-0.0191192626953125,
-0.0096588134765625,
0.039581298828125,
-0.013275146484375,
-0.040374755859375,
-0.054962158203125,
0.01222991943359375,
-0.0274658203125,
0.025909423828125,
0.0124664306640625,
-0.048858642578125,
0.0201263427734375,
0.03955078125,
0.02813720703125,
0.030548095703125,
-0.008697509765625,
0.042572021484375,
-0.076171875,
-0.01212310791015625,
-0.003261566162109375,
0.039306640625,
0.04632568359375,
-0.0276336669921875,
0.0360107421875,
0.0418701171875,
-0.038177490234375,
-0.04949951171875,
-0.0006237030029296875,
-0.0697021484375,
-0.0006427764892578125,
0.0926513671875,
-0.0102996826171875,
-0.041046142578125,
0.019317626953125,
-0.052398681640625,
0.067138671875,
-0.030975341796875,
0.062744140625,
0.056427001953125,
0.001873016357421875,
-0.0258331298828125,
0.0095977783203125,
0.035980224609375,
0.046844482421875,
-0.0633544921875,
-0.004611968994140625,
0.0391845703125,
0.056640625,
0.0181427001953125,
0.0214385986328125,
-0.00029349327087402344,
0.040069580078125,
0.00405120849609375,
0.0267181396484375,
-0.01141357421875,
-0.017822265625,
-0.03314208984375,
0.007251739501953125,
0.01064300537109375,
-0.0251922607421875
]
] |
Henk717/airochronos-33B | 2023-07-28T22:11:05.000Z | [
"transformers",
"pytorch",
"safetensors",
"llama",
"text-generation",
"license:other",
"endpoints_compatible",
"has_space",
"text-generation-inference",
"region:us"
] | text-generation | Henk717 | null | null | Henk717/airochronos-33B | 6 | 5,973 | transformers | 2023-07-10T09:46:47 | ---
license: other
---
After the initial experiment with chronoboros-33B it was evident that the merge was to unpredictable to be useful, testing the individual models it became clear that the bias should be weighted towards Chronos.
This is the new release of the merge with 75% chronos 33B, and 25% airoboros-1.4 33B.
Model has been tested with the Alpaca prompting format combined with KoboldAI Lite's instruct and chat modes, as well as regular story writing.
It has also been tested on basic reasoning tasks, but has not seen much testing for factual information. | 569 | [
[
-0.0640869140625,
-0.040252685546875,
0.031707763671875,
0.0192108154296875,
-0.04388427734375,
-0.0184478759765625,
-0.006244659423828125,
-0.0654296875,
0.0196075439453125,
0.033233642578125,
-0.0693359375,
-0.003200531005859375,
-0.041839599609375,
-0.005523681640625,
-0.0285491943359375,
0.0870361328125,
0.0047149658203125,
0.00302886962890625,
0.0245513916015625,
-0.019927978515625,
-0.047393798828125,
-0.023773193359375,
-0.055938720703125,
-0.042022705078125,
0.061065673828125,
0.034454345703125,
0.07574462890625,
0.0266265869140625,
0.025909423828125,
0.0149688720703125,
-0.0184173583984375,
-0.001316070556640625,
-0.00948333740234375,
-0.005840301513671875,
0.01357269287109375,
-0.0516357421875,
-0.033233642578125,
0.0203399658203125,
0.03765869140625,
0.041534423828125,
-0.0306243896484375,
-0.0130767822265625,
0.0007257461547851562,
0.03155517578125,
-0.04425048828125,
0.005641937255859375,
-0.0299072265625,
0.02032470703125,
-0.006984710693359375,
0.0036144256591796875,
-0.02886962890625,
-0.0499267578125,
0.0256195068359375,
-0.06683349609375,
0.01253509521484375,
0.015655517578125,
0.066650390625,
-0.0033321380615234375,
-0.05926513671875,
-0.047149658203125,
-0.037078857421875,
0.057342529296875,
-0.061065673828125,
0.04644775390625,
0.0231781005859375,
0.032562255859375,
-0.00917816162109375,
-0.0120697021484375,
-0.046417236328125,
-0.0186920166015625,
0.01320648193359375,
0.00426483154296875,
-0.0311279296875,
-0.00241851806640625,
0.0029926300048828125,
0.029876708984375,
-0.022216796875,
0.03155517578125,
-0.056915283203125,
-0.01120758056640625,
0.03985595703125,
0.00664520263671875,
0.005405426025390625,
-0.0185394287109375,
-0.037139892578125,
-0.0452880859375,
-0.0292816162109375,
0.004611968994140625,
0.0670166015625,
0.00011742115020751953,
-0.0213165283203125,
0.060150146484375,
-0.017913818359375,
0.0684814453125,
0.01548004150390625,
0.004901885986328125,
0.0156402587890625,
-0.0338134765625,
-0.03912353515625,
0.03228759765625,
0.055023193359375,
0.062255859375,
0.009002685546875,
0.01300811767578125,
0.0161590576171875,
0.0220947265625,
0.0182647705078125,
-0.053314208984375,
0.005336761474609375,
0.02557373046875,
-0.0295867919921875,
-0.042022705078125,
0.0011320114135742188,
-0.060272216796875,
-0.0258331298828125,
-0.005908966064453125,
0.00875091552734375,
-0.0711669921875,
-0.0024509429931640625,
0.0010013580322265625,
-0.03466796875,
0.0233001708984375,
0.056304931640625,
-0.06427001953125,
0.0128021240234375,
0.03546142578125,
0.02337646484375,
-0.0253753662109375,
-0.036102294921875,
-0.0064697265625,
-0.0275726318359375,
-0.0528564453125,
0.056915283203125,
-0.005039215087890625,
-0.039337158203125,
-0.005832672119140625,
-0.0035228729248046875,
0.0021152496337890625,
-0.024169921875,
0.0292510986328125,
-0.04400634765625,
0.04266357421875,
-0.03802490234375,
-0.04266357421875,
-0.0260009765625,
0.0004596710205078125,
-0.045501708984375,
0.057647705078125,
0.033203125,
-0.0259552001953125,
0.057769775390625,
-0.0164031982421875,
-0.00013816356658935547,
-0.0157928466796875,
-0.0029773712158203125,
-0.0244598388671875,
-0.00936126708984375,
0.00646209716796875,
-0.003971099853515625,
-0.035186767578125,
0.0257720947265625,
-0.039337158203125,
-0.02197265625,
0.04168701171875,
-0.035675048828125,
0.07598876953125,
0.0509033203125,
-0.01322174072265625,
-0.00392913818359375,
-0.00896453857421875,
0.005611419677734375,
-0.00800323486328125,
-0.00675201416015625,
-0.010498046875,
-0.0277862548828125,
-0.00997161865234375,
0.002628326416015625,
0.012939453125,
-0.0299530029296875,
0.03375244140625,
-0.007328033447265625,
0.011688232421875,
0.033203125,
0.0085296630859375,
0.062744140625,
-0.068115234375,
0.050048828125,
0.006282806396484375,
0.040496826171875,
-0.0092010498046875,
-0.072021484375,
-0.05413818359375,
-0.00847625732421875,
0.014190673828125,
0.04388427734375,
-0.03424072265625,
0.033843994140625,
0.017547607421875,
-0.076416015625,
-0.0186004638671875,
0.01006317138671875,
0.0278167724609375,
0.0269622802734375,
-0.0018033981323242188,
-0.0224151611328125,
-0.038787841796875,
-0.07257080078125,
-0.00438690185546875,
-0.019012451171875,
-0.0003085136413574219,
0.0160980224609375,
0.0213623046875,
-0.0167236328125,
0.0609130859375,
-0.046630859375,
-0.005664825439453125,
-0.0162811279296875,
0.038909912109375,
0.01104736328125,
0.032073974609375,
0.068359375,
-0.0489501953125,
-0.0035800933837890625,
-0.005329132080078125,
-0.04241943359375,
-0.0228729248046875,
0.010101318359375,
-0.002532958984375,
0.0283050537109375,
0.022369384765625,
-0.05316162109375,
0.04327392578125,
0.05230712890625,
-0.0219573974609375,
0.0203704833984375,
-0.00003933906555175781,
0.05352783203125,
-0.08050537109375,
-0.009765625,
-0.01483917236328125,
0.0169525146484375,
-0.02374267578125,
-0.0033397674560546875,
0.0136260986328125,
-0.00930023193359375,
-0.04901123046875,
0.047088623046875,
-0.04345703125,
-0.005115509033203125,
-0.0250091552734375,
0.0117034912109375,
-0.00667572021484375,
0.0172271728515625,
-0.020538330078125,
0.0289306640625,
0.0117340087890625,
-0.0494384765625,
0.025238037109375,
0.0164642333984375,
-0.012969970703125,
0.02752685546875,
-0.024200439453125,
-0.0131378173828125,
-0.00909423828125,
0.04205322265625,
-0.05096435546875,
-0.02301025390625,
0.0181884765625,
-0.038330078125,
-0.012359619140625,
0.0008206367492675781,
-0.007434844970703125,
-0.042694091796875,
-0.047027587890625,
0.035491943359375,
0.046051025390625,
-0.020294189453125,
0.0565185546875,
-0.0016012191772460938,
-0.0120849609375,
-0.039581298828125,
-0.058441162109375,
-0.0198974609375,
-0.0180816650390625,
-0.0298614501953125,
0.00981903076171875,
-0.0289764404296875,
-0.0246124267578125,
0.007045745849609375,
0.0099029541015625,
-0.016143798828125,
-0.01177215576171875,
0.02581787109375,
0.05670166015625,
-0.01024627685546875,
-0.0128021240234375,
0.0325927734375,
0.024322509765625,
-0.0178985595703125,
0.019317626953125,
0.06256103515625,
0.00829315185546875,
-0.02740478515625,
-0.047119140625,
0.01026153564453125,
0.044891357421875,
0.00308990478515625,
0.038421630859375,
0.021697998046875,
-0.03900146484375,
0.021820068359375,
-0.0626220703125,
-0.006500244140625,
-0.032623291015625,
0.0027751922607421875,
0.00893402099609375,
-0.05352783203125,
0.06817626953125,
0.031463623046875,
0.0091705322265625,
0.042694091796875,
0.02337646484375,
0.02069091796875,
0.07275390625,
0.0318603515625,
-0.00787353515625,
0.0560302734375,
-0.01377105712890625,
0.0106048583984375,
-0.0596923828125,
-0.04119873046875,
-0.040313720703125,
-0.005451202392578125,
-0.041748046875,
-0.017669677734375,
0.00341033935546875,
0.0350341796875,
-0.00664520263671875,
0.04290771484375,
-0.0238189697265625,
0.0119171142578125,
0.05169677734375,
0.012603759765625,
0.029632568359375,
-0.01451873779296875,
0.0233306884765625,
0.007015228271484375,
-0.05047607421875,
-0.005985260009765625,
0.0780029296875,
-0.0006513595581054688,
0.0526123046875,
0.008056640625,
0.04986572265625,
0.0178680419921875,
0.052459716796875,
-0.0526123046875,
0.02301025390625,
0.01342010498046875,
-0.0670166015625,
-0.0268096923828125,
-0.022064208984375,
-0.08642578125,
0.0163116455078125,
-0.0377197265625,
-0.058441162109375,
0.04364013671875,
0.005100250244140625,
-0.0404052734375,
0.0003190040588378906,
-0.06744384765625,
0.04534912109375,
0.0028743743896484375,
0.01263427734375,
-0.013916015625,
-0.027557373046875,
0.054718017578125,
-0.0181427001953125,
0.0222015380859375,
-0.009033203125,
0.005706787109375,
0.067626953125,
-0.03668212890625,
0.043548583984375,
0.018524169921875,
-0.01277923583984375,
0.056610107421875,
0.00916290283203125,
0.0024318695068359375,
0.0214385986328125,
-0.0020732879638671875,
0.00164794921875,
0.0127105712890625,
-0.0135040283203125,
-0.04132080078125,
0.04766845703125,
-0.051361083984375,
-0.022125244140625,
-0.0667724609375,
-0.058074951171875,
0.0038604736328125,
0.022796630859375,
0.052825927734375,
0.03533935546875,
-0.0251312255859375,
0.00856781005859375,
0.03826904296875,
-0.0013666152954101562,
0.03314208984375,
0.055511474609375,
-0.0562744140625,
-0.0684814453125,
-0.00206756591796875,
0.005779266357421875,
0.0308380126953125,
-0.00193023681640625,
0.0183563232421875,
-0.0022182464599609375,
-0.0031566619873046875,
-0.0362548828125,
0.0060882568359375,
-0.033782958984375,
-0.02801513671875,
-0.033966064453125,
-0.0235748291015625,
-0.051849365234375,
-0.04925537109375,
-0.026824951171875,
-0.0300750732421875,
-0.00754547119140625,
-0.005779266357421875,
0.0299835205078125,
0.054718017578125,
-0.0097503662109375,
0.059478759765625,
-0.0618896484375,
0.0066375732421875,
0.02996826171875,
-0.0162506103515625,
0.0021076202392578125,
-0.03131103515625,
0.0090484619140625,
0.023712158203125,
-0.0269317626953125,
-0.10931396484375,
0.0233306884765625,
-0.0260467529296875,
0.036834716796875,
0.046844482421875,
-0.00830841064453125,
0.0303192138671875,
0.004878997802734375,
0.062103271484375,
0.033111572265625,
-0.0792236328125,
0.044677734375,
-0.055511474609375,
0.03009033203125,
0.0308380126953125,
0.02630615234375,
-0.02362060546875,
-0.04931640625,
-0.07977294921875,
-0.039337158203125,
0.06671142578125,
0.032989501953125,
-0.0189666748046875,
0.0067596435546875,
0.018157958984375,
0.007076263427734375,
0.02850341796875,
-0.0282135009765625,
-0.0174560546875,
-0.0012187957763671875,
-0.01166534423828125,
-0.018646240234375,
-0.0014638900756835938,
-0.0225067138671875,
-0.015472412109375,
0.02728271484375,
0.0030040740966796875,
-0.00006920099258422852,
0.0171051025390625,
0.01035308837890625,
-0.0262298583984375,
0.025848388671875,
0.042938232421875,
0.048248291015625,
-0.042205810546875,
-0.004558563232421875,
0.033477783203125,
-0.03668212890625,
0.005035400390625,
0.032196044921875,
-0.01971435546875,
-0.0099029541015625,
0.03948974609375,
0.06005859375,
0.0267486572265625,
-0.039215087890625,
0.0200347900390625,
-0.01120758056640625,
-0.024322509765625,
-0.005649566650390625,
0.03179931640625,
-0.0003478527069091797,
0.0098114013671875,
0.038421630859375,
0.006725311279296875,
0.003505706787109375,
-0.080810546875,
0.007717132568359375,
0.0174407958984375,
-0.02490234375,
0.00384521484375,
0.04241943359375,
0.00327301025390625,
-0.01389312744140625,
0.060455322265625,
-0.0234527587890625,
-0.04473876953125,
0.078125,
0.031890869140625,
0.040252685546875,
-0.047607421875,
0.0156402587890625,
0.0234527587890625,
0.04510498046875,
-0.0172576904296875,
0.028961181640625,
0.0003223419189453125,
-0.06402587890625,
0.0054779052734375,
-0.01318359375,
-0.055023193359375,
-0.0302734375,
-0.07684326171875,
0.039642333984375,
-0.01306915283203125,
-0.0421142578125,
0.0023956298828125,
0.00997161865234375,
-0.0399169921875,
0.046234130859375,
-0.004119873046875,
0.0765380859375,
-0.0821533203125,
0.04815673828125,
0.033294677734375,
-0.0252227783203125,
-0.074951171875,
-0.037078857421875,
-0.0018281936645507812,
-0.05108642578125,
0.0240325927734375,
-0.0002199411392211914,
0.0296783447265625,
-0.0292816162109375,
-0.035675048828125,
-0.06524658203125,
0.11663818359375,
0.0273590087890625,
-0.0194244384765625,
0.0105743408203125,
-0.0163116455078125,
0.0220489501953125,
-0.0419921875,
0.0303497314453125,
0.0224609375,
0.04461669921875,
0.0006814002990722656,
-0.08599853515625,
0.0020694732666015625,
-0.0123291015625,
-0.0092315673828125,
0.026458740234375,
-0.06427001953125,
0.09613037109375,
-0.01483917236328125,
0.01110076904296875,
0.0122528076171875,
0.054931640625,
0.046234130859375,
0.0582275390625,
0.063720703125,
0.0504150390625,
0.054229736328125,
-0.009857177734375,
0.0921630859375,
-0.0186004638671875,
0.0124053955078125,
0.07244873046875,
-0.034637451171875,
0.041259765625,
0.035125732421875,
0.004390716552734375,
0.042266845703125,
0.058441162109375,
0.016754150390625,
0.0283050537109375,
-0.0219573974609375,
-0.0031375885009765625,
-0.003116607666015625,
-0.0160064697265625,
-0.055450439453125,
0.0269622802734375,
0.0054168701171875,
-0.0168914794921875,
-0.01116180419921875,
0.0006785392761230469,
0.03277587890625,
-0.03314208984375,
-0.0021820068359375,
0.0281219482421875,
0.007755279541015625,
-0.09613037109375,
0.0098114013671875,
-0.0135650634765625,
0.04681396484375,
-0.088134765625,
0.00421142578125,
-0.0233154296875,
0.005435943603515625,
-0.002712249755859375,
-0.052734375,
0.0181121826171875,
-0.03692626953125,
-0.0077972412109375,
0.01200103759765625,
0.041412353515625,
-0.0225372314453125,
-0.025604248046875,
0.0287322998046875,
0.0177459716796875,
-0.016265869140625,
-0.0005908012390136719,
-0.01611328125,
0.033966064453125,
-0.00029468536376953125,
-0.013336181640625,
0.0192108154296875,
0.01422119140625,
-0.008148193359375,
0.04827880859375,
0.045928955078125,
-0.004283905029296875,
0.02587890625,
-0.014312744140625,
0.054718017578125,
-0.04248046875,
-0.05194091796875,
-0.03143310546875,
-0.003704071044921875,
-0.0124359130859375,
-0.07318115234375,
0.07537841796875,
0.05572509765625,
0.04388427734375,
0.0082855224609375,
0.033111572265625,
-0.01444244384765625,
-0.0083465576171875,
-0.0567626953125,
0.04345703125,
-0.033172607421875,
-0.0093231201171875,
-0.0203094482421875,
-0.0867919921875,
0.032623291015625,
0.011383056640625,
0.0035686492919921875,
0.0219573974609375,
0.08978271484375,
0.032318115234375,
0.01104736328125,
0.042816162109375,
-0.00885772705078125,
-0.006465911865234375,
0.0029315948486328125,
0.046905517578125,
0.05230712890625,
-0.05303955078125,
0.0361328125,
-0.056243896484375,
-0.044403076171875,
-0.0198974609375,
-0.06597900390625,
-0.051483154296875,
-0.0277252197265625,
-0.00948333740234375,
-0.043609619140625,
0.01296234130859375,
0.050872802734375,
0.04180908203125,
-0.03350830078125,
-0.0318603515625,
-0.001651763916015625,
0.0020198822021484375,
-0.01552581787109375,
-0.0189056396484375,
-0.0014295578002929688,
0.013763427734375,
-0.06011962890625,
0.01003265380859375,
-0.00982666015625,
0.0303192138671875,
-0.0033893585205078125,
-0.003582000732421875,
-0.006488800048828125,
0.024810791015625,
0.01531219482421875,
0.025970458984375,
-0.039581298828125,
-0.0095977783203125,
0.0191650390625,
-0.0160675048828125,
-0.0201416015625,
0.06182861328125,
-0.0452880859375,
0.0085601806640625,
0.06268310546875,
-0.003459930419921875,
0.04248046875,
0.0007963180541992188,
0.060333251953125,
0.006099700927734375,
0.042205810546875,
0.0105743408203125,
0.043121337890625,
0.003055572509765625,
-0.032440185546875,
0.03680419921875,
0.0087432861328125,
-0.039794921875,
-0.0552978515625,
0.005634307861328125,
-0.1148681640625,
-0.02349853515625,
0.07757568359375,
0.03076171875,
-0.0274810791015625,
0.0289459228515625,
-0.0279693603515625,
0.037445068359375,
-0.043426513671875,
0.04205322265625,
0.07427978515625,
-0.0183563232421875,
-0.00010341405868530273,
-0.041748046875,
0.01506805419921875,
0.047332763671875,
-0.058319091796875,
-0.0293731689453125,
0.038482666015625,
0.00734710693359375,
0.0308837890625,
0.06396484375,
-0.01102447509765625,
0.047027587890625,
0.01290130615234375,
0.0188446044921875,
0.01091766357421875,
-0.0062408447265625,
-0.02252197265625,
-0.015777587890625,
0.001384735107421875,
0.004665374755859375
]
] |
TheBloke/guanaco-7B-HF | 2023-06-05T00:10:27.000Z | [
"transformers",
"pytorch",
"llama",
"text-generation",
"license:other",
"endpoints_compatible",
"has_space",
"text-generation-inference",
"region:us"
] | text-generation | TheBloke | null | null | TheBloke/guanaco-7B-HF | 8 | 5,972 | transformers | 2023-05-25T20:17:10 | ---
license: other
---
<!-- header start -->
<div style="width: 100%;">
<img src="https://i.imgur.com/EBdldam.jpg" alt="TheBlokeAI" style="width: 100%; min-width: 400px; display: block; margin: auto;">
</div>
<div style="display: flex; justify-content: space-between; width: 100%;">
<div style="display: flex; flex-direction: column; align-items: flex-start;">
<p><a href="https://discord.gg/Jq4vkcDakD">Chat & support: my new Discord server</a></p>
</div>
<div style="display: flex; flex-direction: column; align-items: flex-end;">
<p><a href="https://www.patreon.com/TheBlokeAI">Want to contribute? TheBloke's Patreon page</a></p>
</div>
</div>
<!-- header end -->
# Tim Dettmers' Guanaco 7B fp16 HF
These files are fp16 HF model files for [Tim Dettmers' Guanaco 7B](https://huggingface.co/timdettmers/guanaco-7b).
It is the result of merging the LoRA then saving in HF fp16 format.
## Other repositories available
* [4-bit GPTQ models for GPU inference](https://huggingface.co/TheBloke/guanaco-7B-GPTQ)
* [4-bit, 5-bit and 8-bit GGML models for CPU(+GPU) inference](https://huggingface.co/TheBloke/guanaco-7B-GGML)
* [Merged, unquantised fp16 model in HF format](https://huggingface.co/TheBloke/guanaco-7B-HF)
<!-- footer start -->
## Discord
For further support, and discussions on these models and AI in general, join us at:
[TheBloke AI's Discord server](https://discord.gg/Jq4vkcDakD)
## Thanks, and how to contribute.
Thanks to the [chirper.ai](https://chirper.ai) team!
I've had a lot of people ask if they can contribute. I enjoy providing models and helping people, and would love to be able to spend even more time doing it, as well as expanding into new projects like fine tuning/training.
If you're able and willing to contribute it will be most gratefully received and will help me to keep providing more models, and to start work on new AI projects.
Donaters will get priority support on any and all AI/LLM/model questions and requests, access to a private Discord room, plus other benefits.
* Patreon: https://patreon.com/TheBlokeAI
* Ko-Fi: https://ko-fi.com/TheBlokeAI
**Patreon special mentions**: Aemon Algiz, Dmitriy Samsonov, Nathan LeClaire, Trenton Dambrowitz, Mano Prime, David Flickinger, vamX, Nikolai Manek, senxiiz, Khalefa Al-Ahmad, Illia Dulskyi, Jonathan Leane, Talal Aujan, V. Lukas, Joseph William Delisle, Pyrater, Oscar Rangel, Lone Striker, Luke Pendergrass, Eugene Pentland, Sebastain Graf, Johann-Peter Hartman.
Thank you to all my generous patrons and donaters!
<!-- footer end -->
# Original model card
Not provided by original model creator.
| 2,638 | [
[
-0.038330078125,
-0.05029296875,
0.0128021240234375,
0.0052032470703125,
-0.0171051025390625,
-0.01175689697265625,
0.001384735107421875,
-0.049346923828125,
0.041229248046875,
0.0145263671875,
-0.05462646484375,
-0.01175689697265625,
-0.022491455078125,
-0.007770538330078125,
-0.022857666015625,
0.06719970703125,
0.03912353515625,
-0.012115478515625,
-0.00020015239715576172,
0.006069183349609375,
-0.051849365234375,
-0.0111541748046875,
-0.07354736328125,
-0.03973388671875,
0.044952392578125,
0.0092315673828125,
0.056427001953125,
0.047637939453125,
0.0335693359375,
0.03216552734375,
-0.002231597900390625,
-0.0022735595703125,
-0.04217529296875,
-0.00882720947265625,
-0.0015020370483398438,
-0.00806427001953125,
-0.05419921875,
-0.00734710693359375,
0.027191162109375,
0.0208282470703125,
-0.0184478759765625,
0.01131439208984375,
0.003448486328125,
0.04656982421875,
-0.02911376953125,
0.0183868408203125,
-0.031524658203125,
-0.0169830322265625,
-0.015655517578125,
0.0191192626953125,
-0.01116943359375,
-0.0311126708984375,
-0.021881103515625,
-0.0836181640625,
0.0037326812744140625,
0.0016660690307617188,
0.08416748046875,
0.006877899169921875,
-0.00010955333709716797,
0.0106201171875,
-0.057830810546875,
0.044342041015625,
-0.061614990234375,
0.034820556640625,
0.0181884765625,
0.038360595703125,
-0.0022068023681640625,
-0.06951904296875,
-0.046844482421875,
0.0023097991943359375,
0.0006117820739746094,
0.0239410400390625,
-0.04718017578125,
-0.0128021240234375,
-0.008697509765625,
0.0426025390625,
-0.04339599609375,
-0.0009684562683105469,
-0.046539306640625,
-0.005565643310546875,
0.05938720703125,
-0.0024566650390625,
0.0322265625,
0.00446319580078125,
-0.0195770263671875,
-0.03656005859375,
-0.04217529296875,
0.00885009765625,
0.0271148681640625,
0.021759033203125,
-0.07366943359375,
0.04498291015625,
-0.00012564659118652344,
0.035675048828125,
0.0258636474609375,
0.00927734375,
0.0025272369384765625,
-0.05419921875,
-0.038360595703125,
-0.0304412841796875,
0.08294677734375,
0.0215606689453125,
-0.00568389892578125,
0.023590087890625,
0.00583648681640625,
-0.0006060600280761719,
0.01253509521484375,
-0.05389404296875,
-0.033935546875,
0.0249176025390625,
-0.052581787109375,
-0.020904541015625,
0.005096435546875,
-0.06976318359375,
-0.0374755859375,
-0.00150299072265625,
0.02398681640625,
-0.0328369140625,
-0.0504150390625,
0.01265716552734375,
-0.032806396484375,
0.035614013671875,
0.046783447265625,
-0.051177978515625,
0.0146331787109375,
0.050567626953125,
0.04461669921875,
0.0504150390625,
-0.0114593505859375,
-0.033477783203125,
0.009521484375,
-0.01270294189453125,
0.046417236328125,
-0.028350830078125,
-0.044281005859375,
-0.0120697021484375,
0.01349639892578125,
0.008880615234375,
-0.0205230712890625,
0.03521728515625,
-0.0080108642578125,
0.0032711029052734375,
-0.0294342041015625,
-0.02496337890625,
-0.004177093505859375,
0.0033168792724609375,
-0.051788330078125,
0.051025390625,
0.017669677734375,
-0.05010986328125,
0.005214691162109375,
-0.06036376953125,
-0.01441192626953125,
0.02685546875,
-0.01056671142578125,
-0.0199737548828125,
0.004261016845703125,
-0.0054779052734375,
0.0171966552734375,
-0.035614013671875,
-0.00559234619140625,
-0.05609130859375,
-0.015655517578125,
0.022857666015625,
-0.031463623046875,
0.08331298828125,
0.0168304443359375,
-0.022918701171875,
-0.00678253173828125,
-0.049285888671875,
-0.013916015625,
0.038055419921875,
-0.0190582275390625,
0.006439208984375,
-0.00969696044921875,
0.022369384765625,
0.0031986236572265625,
0.0200347900390625,
-0.038238525390625,
0.0191802978515625,
-0.0144805908203125,
0.0391845703125,
0.0626220703125,
-0.00768280029296875,
0.0229949951171875,
-0.05462646484375,
0.040313720703125,
-0.01404571533203125,
0.042083740234375,
0.0022735595703125,
-0.05401611328125,
-0.05377197265625,
-0.03399658203125,
0.0118255615234375,
0.0257110595703125,
-0.045562744140625,
0.0452880859375,
-0.007511138916015625,
-0.056610107421875,
-0.057373046875,
-0.006702423095703125,
0.01800537109375,
0.02618408203125,
0.022369384765625,
-0.0217132568359375,
-0.040557861328125,
-0.06292724609375,
0.011993408203125,
-0.049285888671875,
-0.005191802978515625,
0.048370361328125,
0.0404052734375,
-0.0219268798828125,
0.03289794921875,
-0.0257110595703125,
-0.0294647216796875,
-0.0179443359375,
-0.014617919921875,
0.0223236083984375,
0.06585693359375,
0.059112548828125,
-0.059814453125,
-0.037933349609375,
0.0252838134765625,
-0.0400390625,
-0.0016613006591796875,
-0.001956939697265625,
-0.0297698974609375,
0.003719329833984375,
0.00386810302734375,
-0.079345703125,
0.043731689453125,
0.03790283203125,
-0.047882080078125,
0.0295257568359375,
-0.0233917236328125,
0.0191650390625,
-0.0814208984375,
0.0189056396484375,
0.0157928466796875,
-0.01885986328125,
-0.0369873046875,
0.0024585723876953125,
-0.0399169921875,
-0.0137786865234375,
-0.0284271240234375,
0.056915283203125,
-0.041534423828125,
0.0174713134765625,
0.0009136199951171875,
-0.002582550048828125,
0.0161285400390625,
0.019317626953125,
-0.0203704833984375,
0.0272674560546875,
0.0465087890625,
-0.024688720703125,
0.04150390625,
0.034820556640625,
-0.0099334716796875,
0.0362548828125,
-0.0938720703125,
0.00185394287109375,
-0.0037384033203125,
0.0266571044921875,
-0.08526611328125,
-0.029296875,
0.05419921875,
-0.05987548828125,
0.043853759765625,
-0.0277099609375,
-0.0178680419921875,
-0.0347900390625,
-0.02978515625,
0.029541015625,
0.0615234375,
-0.032806396484375,
0.044677734375,
0.044677734375,
0.0047760009765625,
-0.0440673828125,
-0.057037353515625,
-0.01441192626953125,
-0.015655517578125,
-0.046478271484375,
0.032501220703125,
-0.0231170654296875,
-0.0212860107421875,
0.00616455078125,
0.01165771484375,
-0.012451171875,
-0.002819061279296875,
0.043701171875,
0.0255584716796875,
-0.016143798828125,
-0.035369873046875,
-0.01445770263671875,
0.01241302490234375,
-0.0087127685546875,
-0.0173187255859375,
0.062744140625,
-0.03826904296875,
-0.026458740234375,
-0.07452392578125,
0.0253143310546875,
0.054473876953125,
-0.0218353271484375,
0.05133056640625,
0.038238525390625,
-0.03863525390625,
-0.0035991668701171875,
-0.0433349609375,
-0.012420654296875,
-0.041534423828125,
0.00010585784912109375,
-0.00450897216796875,
-0.05859375,
0.051116943359375,
0.0457763671875,
0.01346588134765625,
0.04779052734375,
0.038726806640625,
-0.0284576416015625,
0.0562744140625,
0.05169677734375,
-0.0221099853515625,
0.0484619140625,
-0.0543212890625,
0.007610321044921875,
-0.037628173828125,
-0.031585693359375,
-0.0523681640625,
-0.038482666015625,
-0.053924560546875,
-0.039764404296875,
0.014801025390625,
-0.0106964111328125,
-0.035247802734375,
0.032806396484375,
-0.04278564453125,
0.0191650390625,
0.0232086181640625,
0.025054931640625,
-0.0008997917175292969,
-0.00736236572265625,
0.0175933837890625,
0.01264190673828125,
-0.05889892578125,
-0.00823974609375,
0.044281005859375,
0.03778076171875,
0.051025390625,
0.0237274169921875,
0.048248291015625,
0.0213623046875,
0.0205078125,
-0.04107666015625,
0.0390625,
-0.01548004150390625,
-0.0704345703125,
-0.01983642578125,
-0.0208892822265625,
-0.06683349609375,
-0.00311279296875,
-0.02386474609375,
-0.046234130859375,
0.045989990234375,
0.0150909423828125,
-0.0335693359375,
0.034942626953125,
-0.021453857421875,
0.06829833984375,
-0.0009522438049316406,
-0.038848876953125,
-0.0165863037109375,
-0.05401611328125,
0.0173492431640625,
0.0214385986328125,
0.0215606689453125,
-0.01277923583984375,
0.01477813720703125,
0.035888671875,
-0.061920166015625,
0.08648681640625,
-0.0146484375,
0.0004589557647705078,
0.057220458984375,
0.006931304931640625,
0.0257415771484375,
0.0255889892578125,
-0.00824737548828125,
0.01971435546875,
0.008270263671875,
-0.028289794921875,
-0.01311492919921875,
0.052581787109375,
-0.0731201171875,
-0.034088134765625,
-0.0189971923828125,
-0.030364990234375,
0.036407470703125,
0.031402587890625,
0.0301666259765625,
0.037689208984375,
-0.026092529296875,
0.04583740234375,
0.0259857177734375,
-0.0072479248046875,
0.050750732421875,
0.01114654541015625,
-0.0007367134094238281,
-0.03656005859375,
0.0672607421875,
-0.005352020263671875,
-0.002933502197265625,
0.0270843505859375,
0.0211944580078125,
-0.026397705078125,
-0.018890380859375,
-0.032501220703125,
0.0516357421875,
-0.0294342041015625,
-0.03240966796875,
-0.025115966796875,
-0.016265869140625,
-0.046844482421875,
-0.0188140869140625,
-0.04730224609375,
-0.033447265625,
-0.0419921875,
0.0201263427734375,
0.04010009765625,
0.043670654296875,
-0.026641845703125,
0.0306549072265625,
-0.051849365234375,
0.002857208251953125,
0.01120758056640625,
0.01611328125,
0.0012483596801757812,
-0.046722412109375,
-0.00789642333984375,
0.020843505859375,
-0.0183258056640625,
-0.051422119140625,
0.04833984375,
0.01377105712890625,
0.047119140625,
0.029083251953125,
-0.0019369125366210938,
0.06121826171875,
-0.0287933349609375,
0.0606689453125,
0.03436279296875,
-0.06292724609375,
0.0345458984375,
-0.056976318359375,
0.015472412109375,
0.0509033203125,
0.032135009765625,
-0.01531982421875,
-0.0250701904296875,
-0.05938720703125,
-0.03515625,
0.032012939453125,
0.01534271240234375,
0.0166015625,
0.002414703369140625,
0.044769287109375,
-0.013275146484375,
0.003681182861328125,
-0.066650390625,
-0.0265350341796875,
-0.0296783447265625,
-0.005031585693359375,
0.0208282470703125,
0.01079559326171875,
-0.015625,
-0.04693603515625,
0.0821533203125,
-0.01546478271484375,
0.05218505859375,
0.01837158203125,
0.033203125,
-0.0175018310546875,
-0.0024871826171875,
0.032135009765625,
0.062042236328125,
-0.0103302001953125,
-0.0183868408203125,
-0.020599365234375,
-0.023284912109375,
-0.00534820556640625,
0.0159912109375,
-0.0135955810546875,
-0.003231048583984375,
0.01210784912109375,
0.0625,
-0.004428863525390625,
-0.0296630859375,
0.0330810546875,
-0.00780487060546875,
-0.017578125,
-0.0252838134765625,
0.0211181640625,
0.02484130859375,
0.047515869140625,
0.0103759765625,
-0.0017099380493164062,
0.01250457763671875,
-0.040863037109375,
0.00731658935546875,
0.056884765625,
-0.0265350341796875,
-0.043426513671875,
0.07696533203125,
0.006320953369140625,
-0.034332275390625,
0.0469970703125,
0.004364013671875,
-0.0183258056640625,
0.07086181640625,
0.053680419921875,
0.068603515625,
-0.0154266357421875,
0.0247650146484375,
0.0357666015625,
0.0198974609375,
0.00015497207641601562,
0.0175933837890625,
0.00223541259765625,
-0.042266845703125,
-0.00986480712890625,
-0.036163330078125,
-0.030242919921875,
0.030181884765625,
-0.047760009765625,
0.040435791015625,
-0.06842041015625,
-0.0242767333984375,
0.017608642578125,
0.0003418922424316406,
-0.046234130859375,
0.01788330078125,
0.0201263427734375,
0.07232666015625,
-0.04681396484375,
0.06158447265625,
0.05609130859375,
-0.044952392578125,
-0.061553955078125,
-0.031463623046875,
0.00775146484375,
-0.04681396484375,
0.01080322265625,
-0.00894927978515625,
-0.00225067138671875,
0.00530242919921875,
-0.06378173828125,
-0.049468994140625,
0.099365234375,
0.0111083984375,
-0.040740966796875,
-0.0105438232421875,
-0.0078277587890625,
0.03387451171875,
-0.03936767578125,
0.0245361328125,
0.0181121826171875,
0.03582763671875,
0.0143280029296875,
-0.0618896484375,
0.0084075927734375,
-0.04656982421875,
0.0024261474609375,
0.01001739501953125,
-0.0975341796875,
0.056396484375,
-0.00620269775390625,
-0.0048828125,
0.027313232421875,
0.050537109375,
0.0255126953125,
0.01297760009765625,
0.041229248046875,
0.0355224609375,
0.045989990234375,
-0.01024627685546875,
0.08306884765625,
-0.0205078125,
0.03314208984375,
0.05572509765625,
0.0009102821350097656,
0.05291748046875,
0.0172576904296875,
-0.023162841796875,
0.030517578125,
0.043426513671875,
-0.02978515625,
0.0213623046875,
-0.00620269775390625,
-0.0259246826171875,
-0.01389312744140625,
-0.0273590087890625,
-0.051666259765625,
0.021697998046875,
0.014434814453125,
-0.007904052734375,
0.0110321044921875,
-0.0235443115234375,
-0.003681182861328125,
-0.02130126953125,
-0.01264190673828125,
0.04046630859375,
0.017913818359375,
-0.01739501953125,
0.05780029296875,
-0.004802703857421875,
0.04681396484375,
-0.053863525390625,
-0.0091400146484375,
-0.043060302734375,
0.0252685546875,
-0.012603759765625,
-0.039337158203125,
0.01995849609375,
-0.012420654296875,
-0.0181884765625,
-0.0054779052734375,
0.06304931640625,
-0.0105743408203125,
-0.04632568359375,
0.0380859375,
0.0234222412109375,
0.0154266357421875,
0.024932861328125,
-0.0733642578125,
0.040069580078125,
0.006664276123046875,
-0.01042938232421875,
0.0193328857421875,
0.0204315185546875,
0.0172882080078125,
0.049346923828125,
0.043487548828125,
0.002132415771484375,
0.005344390869140625,
-0.01236724853515625,
0.074951171875,
-0.029052734375,
-0.0252838134765625,
-0.07012939453125,
0.063232421875,
0.00492095947265625,
-0.0226898193359375,
0.051361083984375,
0.0452880859375,
0.06103515625,
-0.0191650390625,
0.0556640625,
-0.034454345703125,
0.0120849609375,
-0.00273895263671875,
0.0892333984375,
-0.07708740234375,
0.01076507568359375,
-0.0243682861328125,
-0.04840087890625,
-0.01222991943359375,
0.056365966796875,
0.03460693359375,
0.015472412109375,
-0.000705718994140625,
0.059661865234375,
-0.012908935546875,
0.004489898681640625,
0.0193634033203125,
0.0287933349609375,
0.036712646484375,
0.057647705078125,
0.061767578125,
-0.06353759765625,
0.035186767578125,
-0.039337158203125,
-0.01568603515625,
-0.019989013671875,
-0.0633544921875,
-0.05230712890625,
-0.0369873046875,
-0.04364013671875,
-0.043182373046875,
-0.00286102294921875,
0.06256103515625,
0.0631103515625,
-0.047882080078125,
-0.0443115234375,
-0.002288818359375,
0.0092926025390625,
-0.00899505615234375,
-0.0186920166015625,
-0.006160736083984375,
0.0311431884765625,
-0.059234619140625,
0.046844482421875,
0.0045928955078125,
0.033538818359375,
-0.005260467529296875,
-0.012725830078125,
-0.039154052734375,
0.01068878173828125,
0.0283050537109375,
0.06146240234375,
-0.03875732421875,
-0.016937255859375,
0.005344390869140625,
0.0135955810546875,
0.0211334228515625,
0.03753662109375,
-0.03802490234375,
0.0014286041259765625,
0.059051513671875,
0.04180908203125,
0.04779052734375,
0.00472259521484375,
0.03277587890625,
-0.0242919921875,
0.0199737548828125,
0.006011962890625,
0.036376953125,
0.0202178955078125,
-0.033447265625,
0.046173095703125,
0.029052734375,
-0.057220458984375,
-0.07012939453125,
-0.020294189453125,
-0.09124755859375,
-0.0174560546875,
0.06243896484375,
0.00489044189453125,
-0.0274658203125,
0.021392822265625,
-0.006160736083984375,
0.035736083984375,
-0.03375244140625,
0.019989013671875,
0.0288238525390625,
-0.0164642333984375,
-0.0252227783203125,
-0.04364013671875,
0.02606201171875,
0.0129241943359375,
-0.06439208984375,
0.00384521484375,
0.07025146484375,
0.0237884521484375,
0.0281219482421875,
0.06707763671875,
-0.01763916015625,
0.032440185546875,
0.00804901123046875,
0.01009368896484375,
-0.00830841064453125,
-0.0399169921875,
-0.039764404296875,
0.002361297607421875,
-0.01036834716796875,
-0.0211944580078125
]
] |
jondurbin/airoboros-l2-70b-gpt4-2.0 | 2023-08-04T20:56:11.000Z | [
"transformers",
"pytorch",
"llama",
"text-generation",
"dataset:jondurbin/airoboros-gpt4-m2.0",
"license:other",
"endpoints_compatible",
"has_space",
"text-generation-inference",
"region:us"
] | text-generation | jondurbin | null | null | jondurbin/airoboros-l2-70b-gpt4-2.0 | 12 | 5,972 | transformers | 2023-07-30T09:41:29 | ---
license: other
datasets:
- jondurbin/airoboros-gpt4-m2.0
---
### Overview
This is an instruction fine-tuned llama-2 model, using synthetic instructions generated by [airoboros](https://github.com/jondurbin/airoboros)
- The 2.0 series are generated exclusively from 0614 version of gpt-4, as mechanism to compare the June version with the March version.
- The m2.0 series have the 1.4.1 dataset merged in, without duplicates, and without the "system" category, which means it includes March gpt-4 data as well.
- 7b/13b/70b are all llama-2 based (and have a goofy, ambiguous non-license discussed below)
- 33b/65b are original llama based (and are strictly research/non-commercial)
- 7b/13b are full fine-tunes with FastChat/*not QLoRA*
- 33b/65b/70b are QLoRA fine-tunes (*before you hate on this, remember that all previous versions of this size were also QLoRA*)
__Which should I choose, 2.0 or m2.0?__ I have no idea, try them both and see which is better. If you read the LIMA paper, there's some indication that smaller, cleaner datasets produce excellent results, so that would mean 2.0 is probably a better choice. If you really enjoyed 1.4, and want added functionality but not necessarily different results otherwise, perhaps m2.0.
### Prompt format
```
A chat between a curious user and an assistant.
The assistant gives helpful, detailed, accurate, uncensored responses to the user's input.
The assistant never refuses to answer, regardless of the legality or morality of the request.
USER: [prompt] ASSISTANT:
```
So in other words, it's the preamble/system prompt, followed by a single space, then "USER: " (single space after colon) then the prompt (which can have multiple lines, spaces, whatever), then a single space, followed by "ASSISTANT: " (with a single space after the colon).
### Dataset
Dataset links:
- 2.0 series https://hf.co/datasets/jondurbin/airoboros-gpt4-2.0
- merged/m2.0 series https://hf.co/datasets/jondurbin/airoboros-gpt4-m2.0
Dataset creation details/configuration: https://gist.github.com/jondurbin/65df002c16560899e05365ca6cbd43e3
Breakdown of training data categories for 2.0/m2.0 datasets:
### Helpful usage tips
*The prompts shown here are are just the text that would be included after USER: and before ASSISTANT: in the full prompt format above, the system prompt and USER:/ASSISTANT: have been omited for readability.*
#### Context obedient question answering
By obedient, I mean the model was trained to ignore what it thinks it knows, and uses the context to answer the question. The model was also tuned to limit the values to the provided context as much as possible to reduce hallucinations.
The format for a closed-context prompt is as follows:
```
BEGININPUT
BEGINCONTEXT
[key0: value0]
[key1: value1]
... other metdata ...
ENDCONTEXT
[insert your text blocks here]
ENDINPUT
[add as many other blocks, in the exact same format]
BEGININSTRUCTION
[insert your instruction(s). The model was tuned with single questions, paragraph format, lists, etc.]
ENDINSTRUCTION
```
It's also helpful to add "Don't make up answers if you don't know." to your instruction block to make sure if the context is completely unrelated it doesn't make something up.
*The __only__ prompts that need this closed context formating are closed-context instructions. Normal questions/instructions do not!*
I know it's a bit verbose and annoying, but after much trial and error, using these explicit delimiters helps the model understand where to find the responses and how to associate specific sources with it.
- `BEGININPUT` - denotes a new input block
- `BEGINCONTEXT` - denotes the block of context (metadata key/value pairs) to associate with the current input block
- `ENDCONTEXT` - denotes the end of the metadata block for the current input
- [text] - Insert whatever text you want for the input block, as many paragraphs as can fit in the context.
- `ENDINPUT` - denotes the end of the current input block
- [repeat as many input blocks in this format as you want]
- `BEGININSTRUCTION` - denotes the start of the list (or one) instruction(s) to respond to for all of the input blocks above.
- [instruction(s)]
- `ENDINSTRUCTION` - denotes the end of instruction set
It sometimes works without `ENDINSTRUCTION`, but by explicitly including that in the prompt, the model better understands that all of the instructions in the block should be responded to.
Here's a trivial, but important example to prove the point:
```
BEGININPUT
BEGINCONTEXT
date: 2021-01-01
url: https://web.site/123
ENDCONTEXT
In a shocking turn of events, blueberries are now green, but will be sticking with the same name.
ENDINPUT
BEGININSTRUCTION
What color are bluberries? Source?
ENDINSTRUCTION
```
And the response:
```
Blueberries are now green.
Source:
date: 2021-01-01
url: https://web.site/123
```
#### Coding
You can ask for fairly complex coding instructions with multiple criteria, e.g.:
```
Create a python application with the following requirements:
- Asyncio FastAPI webserver
- ping endpoint that returns the current date in JSON format
- file upload endpoint, which calculates the file's sha256 checksum, and checks postgres to deduplicate
```
Or inline criteria:
```
Write a multi-threaded TCP server in C that accepts a "GET [key]" input and "SET [key] [value]" input, and uses a binary tree to get and store the input values.
```
You can also optionally add a single space and "PLAINFORMAT" at the end of your prompt to avoid backticks, explanations, etc. and just print the code, e.g.:
```
Write a websocket application in node.js. PLAINFORMAT
```
#### Agent/function calling
The dataset includes many examples of function/args generation based on input criteria. This is somewhat similar to the OpenAI function calling, but the output is either JSON or YAML.
Example prompt:
```
As an AI assistant, please select the most suitable function and parameters from the list of available functions below, based on the user's input. Provide your response in JSON format.
Input: I want to know how many times 'Python' is mentioned in my text file.
Available functions:
file_analytics:
description: This tool performs various operations on a text file.
params:
action: The operation we want to perform on the data, such as "count_occurrences", "find_line", etc.
filters:
keyword: The word or phrase we want to search for.
```
Response:
```json
{
"function": "file_analytics",
"params": {
"action": "count_occurrences",
"filters": {
"keyword": "Python"
}
}
}
```
#### Chain-of-thought
You can ask for several possible responses to a given problem, with a ranking and final answer selection.
Example prompt:
```
A group of people decide to play a round-robin tournament where every player plays every other player exactly once. If a total of 45 games were played, how many players participated in the tournament? Offer several potential answers, rank them according to mathematical logic, and pick the most feasible one.
```
Example response:
```
Solution 1:
In a round-robin tournament, each player plays every other player exactly once. This means that the total number of games played is equal to the combination of all players taken two at a time (since a game involves two players).
The formula for combinations is nC2 = n(n-1)/2, where n is the total number of players. We know that nC2 equals 45 (the total number of games), so we can set up the equation as follows:
n(n-1)/2 = 45
Solving this equation gives us n=10.
Final answer: There were 10 players in the tournament.
Solution 2:
Another way to approach this problem is by using the concept of triangular numbers. Triangular numbers are a sequence of numbers where each number can be represented as a triangle with dots. The nth triangular number is given by the sum of the first n natural numbers.
If we consider each game as a dot and each player as a side of the triangle, then the total number of games (45) would correspond to the 9th triangular number because 1+2+3+4+5+6+7+8+9=45. However, since each side of the triangle represents a player, and there's one more player than sides in our model (because the last player has no one left to play against), we need to add one more to get the total number of players.
Final answer: There were 10 players in the tournament.
Solution 3:
We could also use trial and error to solve this problem. Starting from a small number of players, we calculate how many games would be played in a round-robin tournament until we reach 45 games. For example, with 2 players, only 1 game is played; with 3 players, 3 games are played; with 4 players, 6 games are played, and so on. Continuing this process, we find that with 10 players, 45 games are played.
Final answer: There were 10 players in the tournament.
Ranking of solutions from best to worst:
1. Solution 1: This solution uses direct mathematical reasoning based on the properties of a round-robin tournament and solves an equation to find the answer.
2. Solution 2: This solution uses the concept of triangular numbers which requires additional steps and knowledge beyond the immediate problem context, but still arrives at the correct answer.
3. Solution 3: This solution relies on trial and error, which is less efficient and reliable than the other methods, but it eventually gets to the right answer.
Best and final answer: There were 10 players in the tournament.
```
#### reWOO style execution planning
The model now supports execution planning for complex instructions that would require making use of several tools. The output is just the plan, you must implement a mechanism to parse the output and actually call the functions!
Example prompt:
```
Please construct a systematic plan to generate an optimal response to the user instruction, utilizing a set of provided tools. Each plan will correspond to an evidence value, which will be the output of one of the available functions given an input string
that could be the user's question, one or more prior evidence values, or a combination of both.
Here are the tools available to be called:
DuckDuckGo[input]: Tool that enables the user to search for information using the DuckDuckGo search engine. This tool is helpful in locating a suitable list of sites that may contain the answer to the user's query. It does not directly answer the question but finds a list of sites that might have the answer.
TextScraper[input]: Load one or more websites from the input string containing newline delimited links, where input is one or more links, and produces plain text output containing the content of the links.
HyperlinkExtractor[input]: Extract links from plain text and produces a plain text, newline delimited response of links.
KnowledgeModel[input]: Question answering language model, particularly useful in answering questions based on an input passage of text. The input must be a text question that references an :evidence[n]: variable, e.g. What color is the cat, given :evidence1:?
The output should be in format:
Plan: [first action to take based on input question]
:evidence0: = [function to call with input parameter]
Plan: [next action to take, based on result of :evidence0:]
:evidence1: = [next function to call and input parameter, which may include reference to previous evidence, e.g. "Given context :evidence0"]
...
Answer: [:evidence[n]: containing the final answer.]
Question: Who won the Nobel Prize in Literature in 2020 and what is their most famous work?
```
Response:
```
Plan: Begin by conducting a web search to find out who won the Nobel Prize in Literature in 2020.
:evidence0: = DuckDuckGo[Nobel Prize in Literature 2020 winner]
Plan: Utilize the KnowledgeModel to interpret the search results and find the name of the laureate.
:evidence1: = KnowledgeModel[Who won the Nobel Prize in Literature in 2020, given :evidence0:?]
Plan: Conduct another web search to find the most famous work of the identified laureate.
:evidence2: = DuckDuckGo[Most famous work of :evidence1:]
Plan: Extract the relevant links from the DuckDuckGo search results for a more focused search.
:evidence3: = HyperlinkExtractor[:evidence2:]
Plan: Use the TextScraper tool to extract information from the relevant links.
:evidence4: = TextScraper[:evidence3:]
Plan: Finally, utilize the KnowledgeModel to identify and summarize the most famous work of the laureate from the extracted information.
:evidence5: = KnowledgeModel[What is the most famous work of :evidence1:, given :evidence4:?]
Answer: :evidence5:
```
For this to be useful, you'd have to parse the output plan text, and implement/call each of the functions. This is just pseudo-code, completely untested off the top of my head, and obviously would requiring full implementation + hardening:
```python
import re
import requests
def inject_context(input_text, **context):
for ref in set(re.findall(r"(:evidence[0-9]+:)", input_text, re.I)):
input_text = input_text.replace(ref, context.get(ref, ""))
return input_text
def duckduckgo(input_text, **context):
search_string = inject_context(input_text, **context)
... search via duck duck go using search_string
... return text content
def link_extractor(input_text, **context):
input_text = inject_context(input_text, **context)
return "\n".join(list(set(re.findall(r"(https?://[^\s]+?\.?)", input_text, re.I))))
def scrape(input_text, **context):
input_text = inject_context(input_text, **context)
text = []
for link in input_text.splitlines():
text.append(requests.get(link).text)
return "\n".join(text)
def infer(input_text, **context)
prompt = inject_context(input_text, **context)
... call model with prompt, return output
def parse_plan(plan):
method_map = {
"DuckDuckGo": duckduckgo,
"HyperlinkExtractor": link_extractor,
"KnowledgeModel": infer,
"TextScraper": scrape,
}
context = {}
for line in plan.strip().splitlines():
if line.startswith("Plan:"):
print(line)
continue
parts = re.match("^(:evidence[0-9]+:)\s*=\s*([^\[]+])(\[.*\])\s$", line, re.I)
if not parts:
if line.startswith("Answer: "):
return context.get(line.split(" ")[-1].strip(), "Answer couldn't be generated...")
raise RuntimeError("bad format: " + line)
context[parts.group(1)] = method_map[parts.group(2)](parts.group(3), **context)
```
### Contribute
If you're interested in new functionality, particularly a new "instructor" type to generate a specific type of training data,
take a look at the dataset generation tool repo: https://github.com/jondurbin/airoboros and either make a PR or open an issue with details.
To help me with the OpenAI/compute costs:
- https://bmc.link/jondurbin
- ETH 0xce914eAFC2fe52FdceE59565Dd92c06f776fcb11
- BTC bc1qdwuth4vlg8x37ggntlxu5cjfwgmdy5zaa7pswf
### Licence and usage restrictions
The airoboros 2.0/m2.0 models are built on top of either llama or llama-2. Any model with `-l2-` in the name uses llama2, `..-33b-...` and `...-65b-...` are based on the original llama.
#### Llama (original) models
If the model was based on the original llama (33b/65b), the license is __cc-by-nc-4.0__ and is for research/academic use only -- no commercial usage whatsoever!
#### Llama-2 models
Base model has a custom Meta license:
- See the [meta-license/LICENSE.txt](meta-license/LICENSE.txt) file attached for the original license provided by Meta.
- See also [meta-license/USE_POLICY.md](meta-license/USE_POLICY.md) and [meta-license/Responsible-Use-Guide.pdf](meta-license/Responsible-Use-Guide.pdf), also provided by Meta.
The fine-tuning data was generated by OpenAI API calls to gpt-4, via [airoboros](https://github.com/jondurbin/airoboros)
The ToS for OpenAI API usage has a clause preventing the output from being used to train a model that __competes__ with OpenAI
- what does *compete* actually mean here?
- these small open source models will not produce output anywhere near the quality of gpt-4, or even gpt-3.5, so I can't imagine this could credibly be considered competing in the first place
- if someone else uses the dataset to do the same, they wouldn't necessarily be violating the ToS because they didn't call the API, so I don't know how that works
- the training data used in essentially all large language models includes a significant amount of copyrighted or otherwise non-permissive licensing in the first place
- other work using the self-instruct method, e.g. the original here: https://github.com/yizhongw/self-instruct released the data and model as apache-2
I am purposingly leaving this license ambiguous (other than the fact you must comply with the Meta original license for llama-2) because I am not a lawyer and refuse to attempt to interpret all of the terms accordingly.
Your best bet is probably to avoid using this commercially due to the OpenAI API usage.
Either way, by using this model, you agree to completely indemnify me. | 17,075 | [
[
-0.0296173095703125,
-0.06591796875,
0.039398193359375,
0.0201568603515625,
-0.01136016845703125,
-0.014739990234375,
-0.0098876953125,
-0.0238800048828125,
0.0162506103515625,
0.0262298583984375,
-0.054595947265625,
-0.042449951171875,
-0.032745361328125,
0.021514892578125,
-0.0161895751953125,
0.0828857421875,
-0.005435943603515625,
-0.00229644775390625,
-0.0009832382202148438,
0.005584716796875,
-0.049896240234375,
-0.03070068359375,
-0.06549072265625,
-0.0110626220703125,
0.031494140625,
0.03558349609375,
0.03424072265625,
0.045166015625,
0.042327880859375,
0.0289154052734375,
-0.00013911724090576172,
0.0184783935546875,
-0.034515380859375,
0.00372314453125,
-0.0064697265625,
-0.03924560546875,
-0.02630615234375,
0.007904052734375,
0.03424072265625,
0.036102294921875,
-0.0185089111328125,
0.0272979736328125,
-0.00030684471130371094,
0.03125,
-0.0313720703125,
0.017852783203125,
-0.031494140625,
0.00737762451171875,
-0.008819580078125,
-0.03741455078125,
-0.0244293212890625,
-0.01922607421875,
0.0061187744140625,
-0.07830810546875,
-0.00505828857421875,
0.01148223876953125,
0.07110595703125,
0.0251007080078125,
-0.032470703125,
-0.0259552001953125,
-0.04052734375,
0.061370849609375,
-0.060302734375,
0.006351470947265625,
0.045928955078125,
0.036834716796875,
-0.032196044921875,
-0.0625,
-0.049224853515625,
-0.0093231201171875,
-0.018280029296875,
0.0209808349609375,
-0.01142120361328125,
-0.00589752197265625,
0.038604736328125,
0.006755828857421875,
-0.06439208984375,
-0.010589599609375,
-0.048004150390625,
-0.007411956787109375,
0.0518798828125,
0.0287628173828125,
0.016876220703125,
-0.0084381103515625,
-0.0295867919921875,
-0.0035114288330078125,
-0.038330078125,
0.0204010009765625,
0.0297393798828125,
0.0276031494140625,
-0.0245819091796875,
0.038421630859375,
-0.026580810546875,
0.046478271484375,
0.007396697998046875,
-0.0156402587890625,
0.006656646728515625,
-0.037994384765625,
-0.0191650390625,
-0.0084381103515625,
0.07855224609375,
0.053009033203125,
0.011566162109375,
0.0024738311767578125,
-0.002384185791015625,
-0.00911712646484375,
0.00897216796875,
-0.0699462890625,
-0.020294189453125,
0.044189453125,
-0.039581298828125,
-0.027191162109375,
-0.0010461807250976562,
-0.06207275390625,
-0.013427734375,
-0.01348876953125,
0.045623779296875,
-0.029510498046875,
0.0014715194702148438,
0.0093536376953125,
-0.0269775390625,
0.0192718505859375,
0.0341796875,
-0.06011962890625,
0.044189453125,
0.0307464599609375,
0.07135009765625,
0.0044708251953125,
-0.0277099609375,
-0.043914794921875,
-0.006031036376953125,
-0.0099945068359375,
0.059783935546875,
-0.031402587890625,
-0.0301666259765625,
-0.0202178955078125,
0.0206451416015625,
0.00189971923828125,
-0.0231781005859375,
0.0182037353515625,
-0.031982421875,
0.0455322265625,
-0.035430908203125,
-0.03900146484375,
-0.023895263671875,
0.020965576171875,
-0.033782958984375,
0.07550048828125,
0.007396697998046875,
-0.05963134765625,
-0.0048828125,
-0.0767822265625,
-0.024749755859375,
-0.00118255615234375,
0.00182342529296875,
-0.007427215576171875,
-0.0246734619140625,
0.01110076904296875,
0.02685546875,
-0.031646728515625,
0.00920867919921875,
-0.0204620361328125,
-0.03900146484375,
0.0275115966796875,
-0.0196685791015625,
0.08782958984375,
0.0236053466796875,
-0.017547607421875,
0.0104827880859375,
-0.052459716796875,
0.004047393798828125,
0.0177764892578125,
-0.03515625,
-0.004650115966796875,
0.00472259521484375,
-0.0049896240234375,
0.004055023193359375,
0.021331787109375,
-0.0362548828125,
0.026947021484375,
-0.0238800048828125,
0.066162109375,
0.0557861328125,
0.0164794921875,
0.0240325927734375,
-0.02923583984375,
0.034423828125,
-0.0008945465087890625,
0.029205322265625,
-0.0296783447265625,
-0.051971435546875,
-0.04412841796875,
0.0017290115356445312,
0.01494598388671875,
0.06903076171875,
-0.05120849609375,
0.03460693359375,
-0.0020656585693359375,
-0.039154052734375,
-0.02166748046875,
-0.01006317138671875,
0.0259552001953125,
0.052581787109375,
0.0394287109375,
-0.01007080078125,
-0.056671142578125,
-0.057098388671875,
0.01415252685546875,
-0.01427459716796875,
0.001926422119140625,
0.036224365234375,
0.054595947265625,
-0.012969970703125,
0.0648193359375,
-0.06585693359375,
-0.0025806427001953125,
-0.006206512451171875,
0.0014972686767578125,
0.0268402099609375,
0.045257568359375,
0.0411376953125,
-0.048797607421875,
-0.029266357421875,
-0.00705718994140625,
-0.0654296875,
-0.00357818603515625,
-0.004947662353515625,
-0.021240234375,
-0.0031566619873046875,
0.023223876953125,
-0.050445556640625,
0.03680419921875,
0.023223876953125,
-0.0357666015625,
0.04754638671875,
-0.0078125,
0.0211334228515625,
-0.09344482421875,
0.0196533203125,
-0.01251983642578125,
-0.010345458984375,
-0.047149658203125,
0.0223388671875,
-0.017547607421875,
-0.0008969306945800781,
-0.037689208984375,
0.053253173828125,
-0.0272064208984375,
0.0013723373413085938,
-0.004547119140625,
0.0109710693359375,
0.0131378173828125,
0.048797607421875,
-0.00989532470703125,
0.06634521484375,
0.039337158203125,
-0.054107666015625,
0.045562744140625,
0.0174713134765625,
-0.00484466552734375,
0.029296875,
-0.06781005859375,
0.0161590576171875,
-0.00264739990234375,
0.0251312255859375,
-0.08660888671875,
-0.0115966796875,
0.041534423828125,
-0.0478515625,
0.00234222412109375,
-0.01059722900390625,
-0.0243072509765625,
-0.040679931640625,
-0.03424072265625,
0.024261474609375,
0.035980224609375,
-0.0241241455078125,
0.0369873046875,
0.025299072265625,
0.00627899169921875,
-0.03887939453125,
-0.05035400390625,
0.0055694580078125,
-0.0254058837890625,
-0.04254150390625,
0.023529052734375,
-0.032470703125,
-0.020538330078125,
-0.0178070068359375,
0.004825592041015625,
-0.019134521484375,
0.02276611328125,
0.016082763671875,
0.016815185546875,
-0.01434326171875,
-0.0097808837890625,
0.006435394287109375,
-0.004119873046875,
0.00247955322265625,
-0.033721923828125,
0.060821533203125,
-0.017364501953125,
-0.007671356201171875,
-0.053497314453125,
0.03765869140625,
0.0218658447265625,
-0.01214599609375,
0.039398193359375,
0.046722412109375,
-0.036651611328125,
0.01262664794921875,
-0.0197906494140625,
-0.02899169921875,
-0.042877197265625,
0.01490020751953125,
-0.0302276611328125,
-0.045135498046875,
0.054351806640625,
0.0220794677734375,
0.0147552490234375,
0.037261962890625,
0.03204345703125,
-0.02392578125,
0.064453125,
0.0179901123046875,
0.0171661376953125,
0.021514892578125,
-0.044891357421875,
-0.0050506591796875,
-0.06658935546875,
-0.0299530029296875,
-0.04107666015625,
-0.0341796875,
-0.042388916015625,
-0.02191162109375,
0.0265655517578125,
0.0238494873046875,
-0.0438232421875,
0.03759765625,
-0.054656982421875,
0.0394287109375,
0.05572509765625,
0.0135955810546875,
0.00762176513671875,
-0.01276397705078125,
0.0003342628479003906,
0.006603240966796875,
-0.046173095703125,
-0.037139892578125,
0.08892822265625,
0.017547607421875,
0.044189453125,
0.01708984375,
0.064208984375,
0.019927978515625,
0.00213623046875,
-0.060028076171875,
0.0526123046875,
0.0032062530517578125,
-0.040740966796875,
-0.03564453125,
-0.0248870849609375,
-0.08135986328125,
0.0123138427734375,
-0.008148193359375,
-0.0693359375,
0.0136260986328125,
0.00970458984375,
-0.06353759765625,
0.0012950897216796875,
-0.056610107421875,
0.06781005859375,
-0.019134521484375,
-0.0225372314453125,
0.008087158203125,
-0.06072998046875,
0.0241241455078125,
0.01439666748046875,
0.0100250244140625,
0.00241851806640625,
-0.004032135009765625,
0.07147216796875,
-0.05517578125,
0.0655517578125,
-0.0224151611328125,
0.017303466796875,
0.039459228515625,
0.0016126632690429688,
0.032806396484375,
0.0168914794921875,
0.0005650520324707031,
0.0131988525390625,
0.02447509765625,
-0.023651123046875,
-0.045135498046875,
0.046630859375,
-0.0689697265625,
-0.038787841796875,
-0.0325927734375,
-0.0384521484375,
0.014556884765625,
0.0269012451171875,
0.03521728515625,
0.0419921875,
-0.00659942626953125,
-0.0035552978515625,
0.04046630859375,
-0.0225372314453125,
0.046722412109375,
0.04248046875,
-0.02227783203125,
-0.047607421875,
0.056976318359375,
0.01383209228515625,
-0.0017766952514648438,
0.047149658203125,
0.0302734375,
-0.02374267578125,
-0.033172607421875,
-0.0518798828125,
0.01177215576171875,
-0.043975830078125,
-0.0223541259765625,
-0.059173583984375,
-0.0043487548828125,
-0.041168212890625,
-0.00518798828125,
-0.0010118484497070312,
-0.04193115234375,
-0.039581298828125,
-0.0021228790283203125,
0.0487060546875,
0.046630859375,
-0.0011072158813476562,
0.04791259765625,
-0.04815673828125,
0.01806640625,
0.0250244140625,
0.0100250244140625,
0.0007610321044921875,
-0.04693603515625,
-0.007671356201171875,
0.017852783203125,
-0.033416748046875,
-0.0882568359375,
0.02655029296875,
0.00830841064453125,
0.03582763671875,
0.040069580078125,
-0.00359344482421875,
0.06475830078125,
-0.044677734375,
0.0811767578125,
0.0034732818603515625,
-0.06353759765625,
0.0579833984375,
-0.045074462890625,
0.0098114013671875,
0.04046630859375,
0.030792236328125,
-0.0457763671875,
-0.004749298095703125,
-0.04107666015625,
-0.060577392578125,
0.07257080078125,
0.022369384765625,
-0.007598876953125,
-0.007228851318359375,
0.03912353515625,
-0.001247406005859375,
0.017578125,
-0.054168701171875,
-0.032928466796875,
-0.03521728515625,
-0.015167236328125,
-0.00212860107421875,
-0.0008406639099121094,
-0.0237274169921875,
-0.027862548828125,
0.038909912109375,
-0.00958251953125,
0.0469970703125,
0.01480865478515625,
0.00579071044921875,
0.004791259765625,
0.006351470947265625,
0.06365966796875,
0.0416259765625,
-0.025726318359375,
0.0016546249389648438,
0.0180816650390625,
-0.038421630859375,
0.0079193115234375,
0.0141754150390625,
-0.01898193359375,
-0.021026611328125,
0.0270233154296875,
0.055999755859375,
0.0038967132568359375,
-0.039825439453125,
0.029327392578125,
-0.01438140869140625,
-0.01158905029296875,
-0.0240631103515625,
0.0204620361328125,
0.01149749755859375,
0.016510009765625,
0.02056884765625,
-0.0078125,
0.031158447265625,
-0.05224609375,
0.0101470947265625,
0.0228424072265625,
-0.0008058547973632812,
-0.0286865234375,
0.05047607421875,
0.015960693359375,
-0.04925537109375,
0.04608154296875,
-0.039703369140625,
-0.04254150390625,
0.06597900390625,
0.056427001953125,
0.049560546875,
-0.0150299072265625,
0.0215911865234375,
0.042877197265625,
0.0233154296875,
-0.01534271240234375,
0.04913330078125,
-0.0073089599609375,
-0.04205322265625,
-0.00557708740234375,
-0.050506591796875,
-0.024078369140625,
0.01192474365234375,
-0.04522705078125,
0.0168304443359375,
-0.052703857421875,
-0.01568603515625,
0.0026798248291015625,
0.016998291015625,
-0.05706787109375,
0.01482391357421875,
-0.01071929931640625,
0.0748291015625,
-0.07232666015625,
0.037628173828125,
0.060821533203125,
-0.0548095703125,
-0.069580078125,
-0.01383209228515625,
0.0092315673828125,
-0.058746337890625,
0.0299530029296875,
0.0152740478515625,
0.0136260986328125,
0.00440216064453125,
-0.05572509765625,
-0.076904296875,
0.10260009765625,
0.006816864013671875,
-0.032470703125,
-0.00899505615234375,
-0.0020542144775390625,
0.0399169921875,
-0.034759521484375,
0.05267333984375,
0.037689208984375,
0.05059814453125,
0.0035114288330078125,
-0.07135009765625,
0.0233917236328125,
-0.0308380126953125,
-0.004520416259765625,
-0.00072479248046875,
-0.06689453125,
0.08526611328125,
-0.0273895263671875,
-0.01473236083984375,
0.0078887939453125,
0.03729248046875,
0.0152587890625,
0.0265655517578125,
0.0264739990234375,
0.03826904296875,
0.07611083984375,
-0.01074981689453125,
0.0765380859375,
-0.0211944580078125,
0.0211944580078125,
0.08502197265625,
-0.0083770751953125,
0.060882568359375,
0.0284881591796875,
-0.037994384765625,
0.04150390625,
0.072021484375,
-0.00920867919921875,
0.0423583984375,
0.0107269287109375,
0.004505157470703125,
0.00044608116149902344,
0.006282806396484375,
-0.03765869140625,
0.037506103515625,
0.01947021484375,
-0.0113525390625,
-0.00872802734375,
-0.0022525787353515625,
0.01389312744140625,
-0.01457977294921875,
-0.0115203857421875,
0.057586669921875,
-0.00032806396484375,
-0.060150146484375,
0.053375244140625,
0.00982666015625,
0.049652099609375,
-0.0435791015625,
-0.01479339599609375,
-0.0286865234375,
-0.00559234619140625,
-0.0228118896484375,
-0.07110595703125,
0.0165252685546875,
-0.00112152099609375,
-0.0252532958984375,
0.00701904296875,
0.028076171875,
-0.02313232421875,
-0.029541015625,
0.0086212158203125,
0.02215576171875,
0.049713134765625,
0.0055694580078125,
-0.05743408203125,
0.0107574462890625,
0.010406494140625,
-0.020355224609375,
0.015960693359375,
0.027618408203125,
-0.00286102294921875,
0.053009033203125,
0.0560302734375,
-0.00246429443359375,
0.00002276897430419922,
-0.0122833251953125,
0.06396484375,
-0.055511474609375,
-0.041412353515625,
-0.06488037109375,
0.04364013671875,
-0.009246826171875,
-0.039337158203125,
0.0479736328125,
0.051055908203125,
0.052398681640625,
0.00299072265625,
0.054107666015625,
-0.0185394287109375,
0.0228118896484375,
-0.037506103515625,
0.048065185546875,
-0.04437255859375,
0.023773193359375,
-0.00699615478515625,
-0.050994873046875,
-0.0063323974609375,
0.0657958984375,
-0.01479339599609375,
0.000041425228118896484,
0.049713134765625,
0.06439208984375,
0.002941131591796875,
0.011322021484375,
-0.0031642913818359375,
0.021240234375,
0.02642822265625,
0.05120849609375,
0.051055908203125,
-0.049652099609375,
0.037689208984375,
-0.022216796875,
-0.0343017578125,
-0.0107879638671875,
-0.058319091796875,
-0.06585693359375,
-0.04083251953125,
-0.005573272705078125,
-0.030548095703125,
0.01207733154296875,
0.08636474609375,
0.04888916015625,
-0.0640869140625,
-0.031707763671875,
0.005374908447265625,
0.0080108642578125,
-0.028717041015625,
-0.023193359375,
0.0227203369140625,
-0.01116943359375,
-0.05145263671875,
0.031890869140625,
0.0022487640380859375,
0.0087890625,
-0.0124969482421875,
-0.005443572998046875,
-0.028564453125,
0.008453369140625,
0.04608154296875,
0.02813720703125,
-0.05267333984375,
-0.0232391357421875,
0.0155181884765625,
-0.00121307373046875,
0.004222869873046875,
0.03411865234375,
-0.061126708984375,
0.0232696533203125,
0.0401611328125,
0.01806640625,
0.036834716796875,
0.0036907196044921875,
0.028564453125,
-0.042449951171875,
0.003154754638671875,
0.006351470947265625,
0.028564453125,
0.01168060302734375,
-0.049102783203125,
0.04412841796875,
0.0245819091796875,
-0.04791259765625,
-0.06988525390625,
0.0010509490966796875,
-0.0849609375,
-0.0265045166015625,
0.09344482421875,
-0.0111083984375,
-0.0133514404296875,
-0.01018524169921875,
-0.03125,
0.0117950439453125,
-0.049346923828125,
0.05010986328125,
0.049163818359375,
-0.037811279296875,
0.006542205810546875,
-0.036163330078125,
0.03515625,
-0.00494384765625,
-0.0660400390625,
-0.007427215576171875,
0.033966064453125,
0.039215087890625,
0.02191162109375,
0.06695556640625,
0.01192474365234375,
0.02276611328125,
0.0081634521484375,
-0.005115509033203125,
-0.021392822265625,
-0.0290374755859375,
-0.01284027099609375,
0.01021575927734375,
-0.021240234375,
-0.0254974365234375
]
] |
CHIH-HUNG/llama-2-13b-dolphin_20w | 2023-09-06T04:55:19.000Z | [
"transformers",
"pytorch",
"llama",
"text-generation",
"dataset:ehartford/dolphin",
"license:llama2",
"endpoints_compatible",
"has_space",
"text-generation-inference",
"region:us"
] | text-generation | CHIH-HUNG | null | null | CHIH-HUNG/llama-2-13b-dolphin_20w | 0 | 5,971 | transformers | 2023-08-29T00:58:57 | ---
license: llama2
datasets:
- ehartford/dolphin
---
# Model Card for Model ID
<!-- Provide a quick summary of what the model is/does. -->
在llama-2-13b上使用dolphin前20萬筆資料集進行訓練
# Fine-Tuning Information
- **GPU:** RTX4090 (single core / 24564MiB)
- **model:** meta-llama/Llama-2-13b-hf
- **dataset:** ehartford/dolphin (取前20w筆訓練集)
- **peft_type:** LoRA
- **lora_rank:** 8
- **lora_target:** q_proj, v_proj
- **per_device_train_batch_size:** 8
- **gradient_accumulation_steps:** 8
- **learning_rate :** 5e-5
- **epoch:** 1
- **precision:** bf16
- **quantization:** load_in_4bit
# Fine-Tuning Detail
- **train_loss:** 0.8354
- **train_runtime:** 28:42:18 (use deepspeed)
# Evaluation
- 評估結果來自**HuggingFaceH4/open_llm_leaderboard**
- 與Llama-2-13b和其他使用dolphin的模型比較4種Benchmark
- Benchmark包含**ARC**、**HellaSwag**、**MMLU**、**TruthfulQA**
- **注意**：ehartford/dolphin-llama-13b使用的是llama-1
| Model |Average| ARC |HellaSwag| MMLU | TruthfulQA |
|----------------------------------|-------|-------|---------|-------|------------|
|meta-llama/Llama-2-13b-hf | 56.9 | 58.11 | 80.97 | 54.34 | 34.17 |
|meta-llama/Llama-2-13b-chat-hf | 59.93 | 59.04 | 81.94 | 54.64 | 44.12 |
|ehartford/dolphin-llama-13b | 59.26 | 55.55 | 77.11 | 52.16 | 52.23 |
|CHIH-HUNG/llama-2-13b-dolphin_5w | 61 | 60.67 | 82.69 | 56.23 | 44.41 |
|CHIH-HUNG/llama-2-13b-dolphin_20w | 60.17 | 59.56 | 82.55 | 55.89 | 42.67 |
# How to convert dataset to json
- 在**load_dataset**中輸入資料集名稱，並且在**take**中輸入要取前幾筆資料
- 觀察該資料集的欄位名稱，填入**example**欄位中(例如instruction、input、output)
- 最後指定json檔儲存位置 (**json_filename**)
```py
import json
from datasets import load_dataset
# 讀取數據集，take可以取得該數據集前n筆資料
dataset = load_dataset("ehartford/dolphin", split="train", streaming=True).take(200000)
# 提取所需欄位並建立新的字典列表
extracted_data = []
for example in dataset:
extracted_example = {
### dolphin
"instruction": example["instruction"],
"input": example["input"],
"output": example["output"]
}
extracted_data.append(extracted_example)
# 指定 JSON 文件名稱
json_filename = "dolphin.json"
# 寫入 JSON 文件
with open(json_filename, "w") as json_file:
json.dump(extracted_data, json_file, indent=4)
print(f"數據已提取並保存為 {json_filename}")
``` | 2,293 | [
[
-0.049468994140625,
-0.04388427734375,
0.009857177734375,
0.0146636962890625,
-0.056243896484375,
0.0026950836181640625,
-0.005504608154296875,
-0.0260009765625,
0.01763916015625,
0.035125732421875,
-0.044189453125,
-0.038970947265625,
-0.044921875,
0.0192108154296875,
-0.004730224609375,
0.0751953125,
-0.00751495361328125,
-0.01367950439453125,
0.0243072509765625,
-0.0005860328674316406,
-0.039093017578125,
-0.0239715576171875,
-0.057281494140625,
-0.028717041015625,
0.0227813720703125,
0.00873565673828125,
0.048553466796875,
0.0777587890625,
0.050567626953125,
0.0225067138671875,
-0.0098114013671875,
0.02484130859375,
-0.041046142578125,
-0.0170745849609375,
0.0122833251953125,
-0.0382080078125,
-0.0487060546875,
-0.007686614990234375,
0.045196533203125,
0.030364990234375,
0.00946044921875,
0.040985107421875,
0.01047515869140625,
0.052154541015625,
-0.0287322998046875,
0.03228759765625,
-0.0224761962890625,
0.01061248779296875,
-0.0245513916015625,
-0.0225067138671875,
-0.0015411376953125,
-0.0212249755859375,
-0.00897216796875,
-0.072998046875,
0.0010280609130859375,
0.0134429931640625,
0.10333251953125,
0.0281219482421875,
-0.0281829833984375,
0.0020389556884765625,
-0.030181884765625,
0.06976318359375,
-0.0760498046875,
-0.003955841064453125,
0.0233306884765625,
0.024566650390625,
-0.017364501953125,
-0.043853759765625,
-0.049163818359375,
0.00606536865234375,
-0.0065765380859375,
0.0173797607421875,
-0.004970550537109375,
-0.0198211669921875,
0.0236053466796875,
0.0313720703125,
-0.02825927734375,
0.005275726318359375,
-0.032958984375,
0.01180267333984375,
0.066162109375,
0.03033447265625,
0.00983428955078125,
-0.016937255859375,
-0.0177459716796875,
-0.0287017822265625,
-0.040618896484375,
0.017333984375,
0.035186767578125,
0.03778076171875,
-0.035400390625,
0.042449951171875,
-0.034576416015625,
0.0322265625,
0.005741119384765625,
-0.03509521484375,
0.045196533203125,
-0.01386260986328125,
-0.03375244140625,
0.004726409912109375,
0.07647705078125,
0.0521240234375,
-0.00009846687316894531,
0.0212554931640625,
-0.01305389404296875,
-0.02349853515625,
0.000018656253814697266,
-0.058349609375,
-0.022216796875,
0.034759521484375,
-0.056915283203125,
-0.037445068359375,
0.0027008056640625,
-0.06805419921875,
-0.005199432373046875,
0.0016012191772460938,
0.0254974365234375,
-0.0235595703125,
-0.04559326171875,
0.006092071533203125,
-0.019683837890625,
0.020599365234375,
0.025177001953125,
-0.059234619140625,
0.017120361328125,
0.038818359375,
0.05279541015625,
0.00931549072265625,
-0.0245361328125,
-0.00031065940856933594,
0.0229034423828125,
-0.03057861328125,
0.045654296875,
-0.004795074462890625,
-0.03607177734375,
-0.019287109375,
0.019622802734375,
-0.00534820556640625,
-0.036590576171875,
0.045654296875,
-0.0291290283203125,
-0.004055023193359375,
-0.041534423828125,
-0.0226593017578125,
-0.04144287109375,
0.03216552734375,
-0.0577392578125,
0.080078125,
0.007843017578125,
-0.06939697265625,
0.024444580078125,
-0.057861328125,
-0.022705078125,
0.002460479736328125,
0.00818634033203125,
-0.0408935546875,
-0.020355224609375,
0.028564453125,
0.041290283203125,
-0.036590576171875,
0.00963592529296875,
-0.0235443115234375,
-0.03924560546875,
0.02227783203125,
-0.018402099609375,
0.06866455078125,
0.0302734375,
-0.0157012939453125,
-0.003185272216796875,
-0.06805419921875,
0.0032367706298828125,
0.052001953125,
-0.042327880859375,
-0.00789642333984375,
-0.002777099609375,
-0.002811431884765625,
-0.005962371826171875,
0.033477783203125,
-0.01538848876953125,
0.0311126708984375,
-0.01062774658203125,
0.0301361083984375,
0.06207275390625,
0.004642486572265625,
0.006633758544921875,
-0.042327880859375,
0.0201416015625,
0.00775146484375,
0.022979736328125,
-0.005580902099609375,
-0.04339599609375,
-0.06524658203125,
-0.024871826171875,
0.003231048583984375,
0.03192138671875,
-0.0367431640625,
0.060516357421875,
-0.0235443115234375,
-0.05169677734375,
-0.04791259765625,
0.0005745887756347656,
0.0195159912109375,
0.046539306640625,
0.039825439453125,
0.004718780517578125,
-0.05572509765625,
-0.06689453125,
0.00893402099609375,
-0.006725311279296875,
0.01064300537109375,
0.03448486328125,
0.053131103515625,
-0.017120361328125,
0.03118896484375,
-0.034454345703125,
-0.017333984375,
-0.028961181640625,
0.0017261505126953125,
0.06744384765625,
0.045440673828125,
0.04913330078125,
-0.034088134765625,
-0.02392578125,
-0.0001271963119506836,
-0.0869140625,
0.0145263671875,
-0.002758026123046875,
-0.01517486572265625,
-0.00162506103515625,
-0.001758575439453125,
-0.043792724609375,
0.0304107666015625,
0.027069091796875,
-0.00954437255859375,
0.041351318359375,
-0.0012712478637695312,
0.0269012451171875,
-0.07623291015625,
0.0153656005859375,
-0.017364501953125,
0.0171966552734375,
-0.023773193359375,
0.0139617919921875,
-0.013671875,
0.02215576171875,
-0.02886962890625,
0.024993896484375,
-0.026458740234375,
-0.000011444091796875,
-0.01267242431640625,
-0.0033702850341796875,
0.003139495849609375,
0.038543701171875,
-0.0017452239990234375,
0.0455322265625,
0.040191650390625,
-0.058746337890625,
0.040435791015625,
0.031982421875,
-0.0239715576171875,
0.0032939910888671875,
-0.044189453125,
0.0021953582763671875,
0.007030487060546875,
0.0229339599609375,
-0.072998046875,
-0.0240936279296875,
0.04205322265625,
-0.0227508544921875,
0.0216217041015625,
-0.03387451171875,
-0.0269622802734375,
-0.04974365234375,
-0.04205322265625,
0.0209808349609375,
0.028961181640625,
-0.048797607421875,
0.016571044921875,
0.01439666748046875,
0.01526641845703125,
-0.05926513671875,
-0.05902099609375,
-0.004703521728515625,
-0.0160369873046875,
-0.03253173828125,
0.0181732177734375,
-0.00872802734375,
0.00038123130798339844,
0.00628662109375,
0.0014410018920898438,
0.0037746429443359375,
0.007904052734375,
0.01104736328125,
0.044921875,
-0.0283355712890625,
-0.02838134765625,
0.007328033447265625,
-0.004779815673828125,
0.0023670196533203125,
0.01209259033203125,
0.0594482421875,
-0.016265869140625,
-0.01352691650390625,
-0.050018310546875,
-0.0013599395751953125,
0.03179931640625,
0.01074981689453125,
0.042144775390625,
0.05712890625,
-0.01206207275390625,
0.003631591796875,
-0.0190277099609375,
0.003742218017578125,
-0.0374755859375,
0.027435302734375,
-0.0506591796875,
-0.043182373046875,
0.05206298828125,
-0.00554656982421875,
0.01317596435546875,
0.06292724609375,
0.0186614990234375,
-0.01541900634765625,
0.0838623046875,
0.0175933837890625,
-0.0213470458984375,
0.01454925537109375,
-0.07110595703125,
0.004497528076171875,
-0.081787109375,
-0.031494140625,
-0.03179931640625,
-0.03741455078125,
-0.047149658203125,
-0.00844573974609375,
0.01241302490234375,
0.0245513916015625,
-0.04937744140625,
0.0277099609375,
-0.06402587890625,
0.0207672119140625,
0.041473388671875,
0.016571044921875,
0.00782012939453125,
-0.0045928955078125,
0.0040740966796875,
0.0012712478637695312,
-0.038299560546875,
-0.03564453125,
0.1031494140625,
0.0233001708984375,
0.0526123046875,
0.0022678375244140625,
0.05438232421875,
0.0121002197265625,
0.005962371826171875,
-0.042572021484375,
0.049285888671875,
-0.006206512451171875,
-0.0477294921875,
-0.0189971923828125,
-0.01885986328125,
-0.05584716796875,
0.0301055908203125,
-0.0129241943359375,
-0.06060791015625,
0.0006299018859863281,
-0.0004725456237792969,
-0.022552490234375,
0.04705810546875,
-0.0256195068359375,
0.049407958984375,
-0.0316162109375,
-0.02520751953125,
0.00014960765838623047,
-0.03778076171875,
0.05670166015625,
0.0017099380493164062,
0.01079559326171875,
-0.029998779296875,
-0.006847381591796875,
0.08074951171875,
-0.04718017578125,
0.044189453125,
-0.0268402099609375,
0.006946563720703125,
0.034881591796875,
0.002475738525390625,
0.051788330078125,
0.0201263427734375,
-0.006847381591796875,
0.040618896484375,
0.002674102783203125,
-0.0192108154296875,
-0.0185546875,
0.049957275390625,
-0.09161376953125,
-0.039703369140625,
-0.04498291015625,
-0.0257110595703125,
0.00930023193359375,
0.028350830078125,
0.0338134765625,
-0.0087432861328125,
0.0161285400390625,
0.011566162109375,
0.0304107666015625,
-0.015716552734375,
0.043487548828125,
0.031768798828125,
-0.0189361572265625,
-0.055938720703125,
0.06304931640625,
0.002590179443359375,
-0.00896453857421875,
0.0294647216796875,
0.0081024169921875,
-0.0304107666015625,
-0.048675537109375,
-0.048126220703125,
0.0263824462890625,
-0.04052734375,
-0.04052734375,
-0.036834716796875,
-0.03582763671875,
-0.039581298828125,
-0.004001617431640625,
-0.042510986328125,
-0.01953125,
-0.0565185546875,
-0.008575439453125,
0.04693603515625,
0.042816162109375,
-0.0101318359375,
0.049072265625,
-0.057586669921875,
0.0256805419921875,
0.01334381103515625,
0.007541656494140625,
0.01047515869140625,
-0.056304931640625,
-0.014617919921875,
0.005031585693359375,
-0.0244293212890625,
-0.04278564453125,
0.05322265625,
0.003047943115234375,
0.045928955078125,
0.052886962890625,
-0.00447845458984375,
0.0882568359375,
-0.007579803466796875,
0.0704345703125,
0.0215301513671875,
-0.05072021484375,
0.045867919921875,
-0.02850341796875,
-0.0105133056640625,
0.02862548828125,
0.02978515625,
-0.0186614990234375,
0.0019073486328125,
-0.040069580078125,
-0.0633544921875,
0.07989501953125,
0.00569915771484375,
-0.0105743408203125,
0.0178680419921875,
0.0213165283203125,
0.0093841552734375,
0.0206146240234375,
-0.05706787109375,
-0.043212890625,
-0.0369873046875,
-0.0025634765625,
0.002742767333984375,
-0.0083770751953125,
-0.0162811279296875,
-0.04095458984375,
0.06207275390625,
-0.005023956298828125,
0.03643798828125,
0.00586700439453125,
0.01328277587890625,
-0.021453857421875,
-0.00901031494140625,
0.0304107666015625,
0.02630615234375,
-0.0469970703125,
-0.006900787353515625,
0.01343536376953125,
-0.042510986328125,
0.00409698486328125,
0.0020427703857421875,
-0.0118560791015625,
-0.0107269287109375,
0.042236328125,
0.055633544921875,
0.004451751708984375,
-0.0296478271484375,
0.01934814453125,
-0.0015039443969726562,
-0.0157318115234375,
-0.0215606689453125,
0.0243988037109375,
-0.0017948150634765625,
0.0305328369140625,
0.03839111328125,
0.00021076202392578125,
0.003978729248046875,
-0.022857666015625,
-0.01233673095703125,
0.017913818359375,
0.0196533203125,
-0.0176544189453125,
0.062744140625,
0.00130462646484375,
-0.0037364959716796875,
0.052703857421875,
-0.015380859375,
-0.03216552734375,
0.06671142578125,
0.043670654296875,
0.050537109375,
-0.01165008544921875,
-0.0031986236572265625,
0.06439208984375,
0.0272979736328125,
-0.0176849365234375,
0.046539306640625,
-0.004749298095703125,
-0.057403564453125,
-0.01158905029296875,
-0.05572509765625,
-0.0161285400390625,
0.050689697265625,
-0.0548095703125,
0.0151824951171875,
-0.05694580078125,
-0.025146484375,
0.004291534423828125,
0.031219482421875,
-0.048675537109375,
0.0170135498046875,
0.0090179443359375,
0.0672607421875,
-0.058380126953125,
0.07012939453125,
0.035614013671875,
-0.0478515625,
-0.067626953125,
-0.028350830078125,
-0.011932373046875,
-0.08355712890625,
0.051116943359375,
0.0079803466796875,
0.0194854736328125,
-0.00627899169921875,
-0.0635986328125,
-0.08795166015625,
0.10552978515625,
0.01132965087890625,
-0.043609619140625,
0.01506805419921875,
0.0157318115234375,
0.0234222412109375,
-0.0256805419921875,
0.0308837890625,
0.0489501953125,
0.052093505859375,
0.001827239990234375,
-0.056182861328125,
0.0227813720703125,
-0.036956787109375,
-0.0001938343048095703,
-0.0047760009765625,
-0.08380126953125,
0.09942626953125,
-0.0212860107421875,
0.00913238525390625,
0.004314422607421875,
0.041961669921875,
0.047271728515625,
0.0189361572265625,
0.028656005859375,
0.051300048828125,
0.051788330078125,
-0.01910400390625,
0.056427001953125,
-0.0097808837890625,
0.035400390625,
0.05560302734375,
-0.015899658203125,
0.05517578125,
0.027923583984375,
-0.04052734375,
0.031341552734375,
0.07318115234375,
-0.040252685546875,
0.058868408203125,
-0.00901031494140625,
-0.00714111328125,
-0.00713348388671875,
0.00372314453125,
-0.050445556640625,
0.0208892822265625,
0.033294677734375,
-0.0281219482421875,
0.00374603271484375,
-0.015960693359375,
0.0168914794921875,
-0.037322998046875,
-0.0244293212890625,
0.0443115234375,
-0.0157318115234375,
-0.0276031494140625,
0.07452392578125,
-0.006320953369140625,
0.058258056640625,
-0.044769287109375,
-0.010833740234375,
-0.0169525146484375,
0.0103912353515625,
-0.027587890625,
-0.068359375,
-0.004611968994140625,
0.0005564689636230469,
-0.012908935546875,
0.01995849609375,
0.04107666015625,
-0.0011167526245117188,
-0.03363037109375,
0.0306549072265625,
0.00945281982421875,
0.035552978515625,
0.0122833251953125,
-0.0799560546875,
0.035888671875,
0.02001953125,
-0.045379638671875,
0.0218048095703125,
0.0181884765625,
0.022247314453125,
0.06536865234375,
0.07122802734375,
0.01275634765625,
0.0176239013671875,
-0.0110015869140625,
0.07318115234375,
-0.0587158203125,
-0.02337646484375,
-0.05072021484375,
0.0304412841796875,
-0.021697998046875,
-0.0419921875,
0.048858642578125,
0.059356689453125,
0.06268310546875,
-0.0103607177734375,
0.07366943359375,
-0.0191802978515625,
0.03192138671875,
-0.0293121337890625,
0.052703857421875,
-0.05267333984375,
0.01262664794921875,
-0.01354217529296875,
-0.0399169921875,
-0.0065765380859375,
0.062164306640625,
-0.005908966064453125,
0.0012664794921875,
0.041015625,
0.0439453125,
-0.00138092041015625,
0.01152801513671875,
0.006267547607421875,
0.0221405029296875,
0.028564453125,
0.058441162109375,
0.048736572265625,
-0.0733642578125,
0.046600341796875,
-0.061004638671875,
-0.0025615692138671875,
-0.0340576171875,
-0.048858642578125,
-0.06524658203125,
-0.0197296142578125,
-0.0253143310546875,
-0.023101806640625,
-0.00550079345703125,
0.059295654296875,
0.047760009765625,
-0.058868408203125,
-0.0283355712890625,
0.0029125213623046875,
0.00901031494140625,
-0.032684326171875,
-0.019775390625,
0.053863525390625,
-0.0023593902587890625,
-0.0565185546875,
0.031585693359375,
-0.0014753341674804688,
0.01678466796875,
0.00009423494338989258,
-0.024017333984375,
-0.01641845703125,
-0.019073486328125,
0.02728271484375,
0.02752685546875,
-0.0498046875,
-0.0106353759765625,
-0.026214599609375,
-0.0016851425170898438,
0.01412200927734375,
0.00775909423828125,
-0.040313720703125,
0.0149688720703125,
0.0307464599609375,
0.01580810546875,
0.056884765625,
-0.01169586181640625,
-0.0017118453979492188,
-0.0272979736328125,
0.020782470703125,
-0.002910614013671875,
0.03582763671875,
0.007099151611328125,
-0.033416748046875,
0.060302734375,
0.034271240234375,
-0.038726806640625,
-0.07696533203125,
-0.035125732421875,
-0.09588623046875,
-0.021514892578125,
0.0699462890625,
-0.01111602783203125,
-0.054046630859375,
0.01531982421875,
-0.0162200927734375,
0.045318603515625,
-0.043060302734375,
0.0498046875,
0.0234527587890625,
-0.0198211669921875,
-0.0007729530334472656,
-0.0458984375,
0.0259246826171875,
-0.01198577880859375,
-0.052093505859375,
-0.00916290283203125,
-0.00118255615234375,
0.0303497314453125,
0.017181396484375,
0.027557373046875,
-0.0010290145874023438,
0.00411224365234375,
0.0244293212890625,
0.01458740234375,
-0.028106689453125,
-0.0012083053588867188,
-0.0013265609741210938,
-0.01006317138671875,
-0.0203399658203125,
-0.044708251953125
]
] |
garage-bAInd/Platypus2-70B | 2023-08-15T01:45:24.000Z | [
"transformers",
"pytorch",
"llama",
"text-generation",
"en",
"dataset:garage-bAInd/Open-Platypus",
"arxiv:2308.07317",
"arxiv:2307.09288",
"license:cc-by-nc-sa-4.0",
"endpoints_compatible",
"has_space",
"text-generation-inference",
"region:us"
] | text-generation | garage-bAInd | null | null | garage-bAInd/Platypus2-70B | 20 | 5,970 | transformers | 2023-08-04T18:21:45 | ---
license: cc-by-nc-sa-4.0
language:
- en
datasets:
- garage-bAInd/Open-Platypus
---
# Platypus2-70B
Platypus-70B is an instruction fine-tuned model based on the LLaMa2-70B transformer architecture.
### Benchmark Metrics
| Metric | Value |
|-----------------------|-------|
| MMLU (5-shot) | 70.48 |
| ARC (25-shot) | 71.84 |
| HellaSwag (10-shot) | 87.94 |
| TruthfulQA (0-shot) | 62.26 |
| Avg. | 73.13 |
We use state-of-the-art [Language Model Evaluation Harness](https://github.com/EleutherAI/lm-evaluation-harness) to run the benchmark tests above, using the same version as the HuggingFace LLM Leaderboard. Please see below for detailed instructions on reproducing benchmark results.
### Model Details
* **Trained by**: Cole Hunter & Ariel Lee
* **Model type:** **Platypus2-70B** is an auto-regressive language model based on the LLaMA2 transformer architecture.
* **Language(s)**: English
* **License for base weights**: Non-Commercial Creative Commons license ([CC BY-NC-4.0](https://creativecommons.org/licenses/by-nc/4.0/))
### Prompt Template
```
### Instruction:
<prompt> (without the <>)
### Response:
```
### Training Dataset
`garage-bAInd/Platypus2-70B` trained using STEM and logic based dataset [`garage-bAInd/Open-Platypus`](https://huggingface.co/datasets/garage-bAInd/Open-Platypus).
Please see our [paper](https://arxiv.org/abs/2308.07317) and [project webpage](https://platypus-llm.github.io) for additional information.
### Training Procedure
`garage-bAInd/Platypus2-70B` was instruction fine-tuned using LoRA on 8 A100 80GB. For training details and inference instructions please see the [Platypus](https://github.com/arielnlee/Platypus) GitHub repo.
### Reproducing Evaluation Results
Install LM Evaluation Harness:
```
# clone repository
git clone https://github.com/EleutherAI/lm-evaluation-harness.git
# check out the correct commit
git checkout b281b0921b636bc36ad05c0b0b0763bd6dd43463
# change to repo directory
cd lm-evaluation-harness
# install
pip install -e .
```
Each task was evaluated on a single A100 80GB GPU.
ARC:
```
python main.py --model hf-causal-experimental --model_args pretrained=garage-bAInd/Platypus2-70B --tasks arc_challenge --batch_size 1 --no_cache --write_out --output_path results/Platypus2-70B/arc_challenge_25shot.json --device cuda --num_fewshot 25
```
HellaSwag:
```
python main.py --model hf-causal-experimental --model_args pretrained=garage-bAInd/Platypus2-70B --tasks hellaswag --batch_size 1 --no_cache --write_out --output_path results/Platypus2-70B/hellaswag_10shot.json --device cuda --num_fewshot 10
```
MMLU:
```
python main.py --model hf-causal-experimental --model_args pretrained=garage-bAInd/Platypus2-70B --tasks hendrycksTest-* --batch_size 1 --no_cache --write_out --output_path results/Platypus2-70B/mmlu_5shot.json --device cuda --num_fewshot 5
```
TruthfulQA:
```
python main.py --model hf-causal-experimental --model_args pretrained=garage-bAInd/Platypus2-70B --tasks truthfulqa_mc --batch_size 1 --no_cache --write_out --output_path results/Platypus2-70B/truthfulqa_0shot.json --device cuda
```
### Limitations and bias
Llama 2 and fine-tuned variants are a new technology that carries risks with use. Testing conducted to date has been in English, and has not covered, nor could it cover all scenarios. For these reasons, as with all LLMs, Llama 2 and any fine-tuned varient's potential outputs cannot be predicted in advance, and the model may in some instances produce inaccurate, biased or other objectionable responses to user prompts. Therefore, before deploying any applications of Llama 2 variants, developers should perform safety testing and tuning tailored to their specific applications of the model.
Please see the Responsible Use Guide available at https://ai.meta.com/llama/responsible-use-guide/
### Citations
```bibtex
@article{platypus2023,
title={Platypus: Quick, Cheap, and Powerful Refinement of LLMs},
author={Ariel N. Lee and Cole J. Hunter and Nataniel Ruiz},
booktitle={arXiv preprint arxiv:2308.07317},
year={2023}
}
```
```bibtex
@misc{touvron2023llama,
title={Llama 2: Open Foundation and Fine-Tuned Chat Models},
author={Hugo Touvron and Louis Martin and Kevin Stone and Peter Albert and Amjad Almahairi and Yasmine Babaei and Nikolay Bashlykov year={2023},
eprint={2307.09288},
archivePrefix={arXiv},
}
```
```bibtex
@inproceedings{
hu2022lora,
title={Lo{RA}: Low-Rank Adaptation of Large Language Models},
author={Edward J Hu and Yelong Shen and Phillip Wallis and Zeyuan Allen-Zhu and Yuanzhi Li and Shean Wang and Lu Wang and Weizhu Chen},
booktitle={International Conference on Learning Representations},
year={2022},
url={https://openreview.net/forum?id=nZeVKeeFYf9}
}
``` | 4,869 | [
[
-0.02349853515625,
-0.06329345703125,
0.022247314453125,
0.0275726318359375,
-0.0235443115234375,
-0.00567626953125,
-0.026641845703125,
-0.034881591796875,
-0.00035858154296875,
0.02374267578125,
-0.037872314453125,
-0.0288848876953125,
-0.049346923828125,
-0.005992889404296875,
-0.00656890869140625,
0.07696533203125,
-0.029998779296875,
-0.01386260986328125,
-0.00997161865234375,
-0.0149078369140625,
-0.0474853515625,
-0.0361328125,
-0.032379150390625,
-0.03314208984375,
0.0188751220703125,
0.0271759033203125,
0.041961669921875,
0.04193115234375,
0.049072265625,
0.02301025390625,
-0.0170745849609375,
0.01715087890625,
-0.045166015625,
-0.01187896728515625,
0.01374053955078125,
-0.04437255859375,
-0.04931640625,
0.00720977783203125,
0.034271240234375,
0.026214599609375,
-0.0115203857421875,
0.03790283203125,
0.0128173828125,
0.026123046875,
-0.04547119140625,
0.031402587890625,
-0.0421142578125,
-0.0111541748046875,
-0.0243988037109375,
-0.0128326416015625,
-0.02630615234375,
-0.01468658447265625,
-0.00994110107421875,
-0.0538330078125,
0.00344085693359375,
0.005352020263671875,
0.0830078125,
0.040802001953125,
-0.01763916015625,
-0.00875091552734375,
-0.0247802734375,
0.07147216796875,
-0.0633544921875,
0.00922393798828125,
0.02825927734375,
0.00666046142578125,
-0.030242919921875,
-0.051788330078125,
-0.04425048828125,
-0.0231475830078125,
-0.0036716461181640625,
0.0107421875,
-0.018310546875,
-0.0013113021850585938,
0.0272979736328125,
0.030059814453125,
-0.029632568359375,
0.03466796875,
-0.03790283203125,
-0.012908935546875,
0.056671142578125,
0.01303863525390625,
0.0024318695068359375,
-0.013275146484375,
-0.038330078125,
-0.032012939453125,
-0.05291748046875,
0.0272064208984375,
0.03131103515625,
0.00966644287109375,
-0.032745361328125,
0.050567626953125,
-0.0131072998046875,
0.03076171875,
0.00534820556640625,
-0.0404052734375,
0.044921875,
-0.018280029296875,
-0.0242156982421875,
-0.00421142578125,
0.07080078125,
0.034027099609375,
0.003734588623046875,
0.00614166259765625,
-0.0182342529296875,
0.022613525390625,
-0.0054779052734375,
-0.0582275390625,
-0.01251983642578125,
0.0225067138671875,
-0.0265655517578125,
-0.01519012451171875,
-0.0136260986328125,
-0.045745849609375,
-0.02337646484375,
-0.00933837890625,
0.0322265625,
-0.03375244140625,
-0.029815673828125,
0.01324462890625,
0.0003056526184082031,
0.039794921875,
0.016357421875,
-0.051727294921875,
0.0290679931640625,
0.049713134765625,
0.06610107421875,
-0.029541015625,
-0.045684814453125,
-0.0259857177734375,
-0.00235748291015625,
-0.0166778564453125,
0.06103515625,
-0.00951385498046875,
-0.0177001953125,
-0.02001953125,
0.0158538818359375,
-0.01262664794921875,
-0.0477294921875,
0.038848876953125,
-0.0224151611328125,
0.01245880126953125,
-0.0177154541015625,
-0.031707763671875,
-0.0301971435546875,
-0.004856109619140625,
-0.032470703125,
0.10137939453125,
0.014862060546875,
-0.056549072265625,
0.00553131103515625,
-0.0518798828125,
-0.026885986328125,
-0.01087188720703125,
0.01273345947265625,
-0.0443115234375,
-0.0031585693359375,
0.0085906982421875,
0.03363037109375,
-0.037353515625,
0.02203369140625,
-0.019561767578125,
-0.033294677734375,
0.01267242431640625,
-0.01172637939453125,
0.07550048828125,
0.0168304443359375,
-0.048431396484375,
0.00806427001953125,
-0.0443115234375,
-0.00856781005859375,
0.035888671875,
-0.0250244140625,
-0.008209228515625,
-0.006683349609375,
-0.010650634765625,
0.00586700439453125,
0.03204345703125,
-0.034210205078125,
0.005878448486328125,
-0.0258636474609375,
0.044647216796875,
0.05633544921875,
-0.00856781005859375,
0.0183563232421875,
-0.0379638671875,
0.0306549072265625,
0.0011663436889648438,
0.0246429443359375,
-0.00017380714416503906,
-0.05224609375,
-0.07879638671875,
-0.0211334228515625,
0.00445556640625,
0.061126708984375,
-0.0280914306640625,
0.0400390625,
-0.0016880035400390625,
-0.046539306640625,
-0.043914794921875,
0.0272674560546875,
0.043365478515625,
0.041961669921875,
0.039031982421875,
-0.0309906005859375,
-0.04156494140625,
-0.06658935546875,
-0.0074462890625,
-0.026580810546875,
0.01053619384765625,
0.0251617431640625,
0.051605224609375,
-0.0248260498046875,
0.046783447265625,
-0.034271240234375,
-0.021240234375,
-0.02044677734375,
-0.0021877288818359375,
0.0267333984375,
0.047271728515625,
0.036895751953125,
-0.0187530517578125,
-0.017547607421875,
-0.017974853515625,
-0.0616455078125,
-0.02032470703125,
-0.0005536079406738281,
-0.0195770263671875,
0.036834716796875,
0.01430511474609375,
-0.06549072265625,
0.0249481201171875,
0.038238525390625,
-0.01416778564453125,
0.037872314453125,
-0.01233673095703125,
-0.0104217529296875,
-0.0618896484375,
0.01071929931640625,
0.0023288726806640625,
-0.00091552734375,
-0.034088134765625,
0.0158538818359375,
-0.0006270408630371094,
0.00841522216796875,
-0.0487060546875,
0.046966552734375,
-0.037933349609375,
-0.011688232421875,
-0.01157379150390625,
0.0122528076171875,
-0.00809478759765625,
0.0526123046875,
-0.005809783935546875,
0.0655517578125,
0.035736083984375,
-0.04388427734375,
0.017974853515625,
0.02783203125,
-0.0276641845703125,
0.0209808349609375,
-0.0701904296875,
0.017822265625,
0.00937652587890625,
0.02349853515625,
-0.07635498046875,
-0.01264190673828125,
0.024444580078125,
-0.0201263427734375,
0.024993896484375,
0.005889892578125,
-0.051910400390625,
-0.0341796875,
-0.035858154296875,
0.0215606689453125,
0.0667724609375,
-0.0440673828125,
0.02490234375,
0.033172607421875,
0.0018568038940429688,
-0.046630859375,
-0.057220458984375,
-0.016998291015625,
-0.028656005859375,
-0.055450439453125,
0.015899658203125,
-0.01215362548828125,
-0.015838623046875,
-0.0218658447265625,
-0.0111083984375,
0.01134490966796875,
0.01702880859375,
0.04107666015625,
0.031341552734375,
-0.01284027099609375,
-0.00795745849609375,
0.0009899139404296875,
-0.0182952880859375,
0.0048370361328125,
0.0097503662109375,
0.05023193359375,
-0.0294342041015625,
-0.0141448974609375,
-0.059783935546875,
0.00003349781036376953,
0.03314208984375,
-0.0211029052734375,
0.04803466796875,
0.049072265625,
-0.013580322265625,
0.01019287109375,
-0.057586669921875,
-0.01482391357421875,
-0.038665771484375,
0.03179931640625,
-0.0203399658203125,
-0.05291748046875,
0.045806884765625,
-0.0016431808471679688,
0.01568603515625,
0.053466796875,
0.0606689453125,
-0.000011146068572998047,
0.057861328125,
0.0401611328125,
0.006103515625,
0.03265380859375,
-0.05059814453125,
0.004161834716796875,
-0.08099365234375,
-0.0274200439453125,
-0.028594970703125,
-0.0301971435546875,
-0.04681396484375,
-0.0421142578125,
0.01299285888671875,
0.0192413330078125,
-0.044952392578125,
0.036468505859375,
-0.039520263671875,
0.01806640625,
0.043365478515625,
0.00942230224609375,
0.0121612548828125,
0.00457000732421875,
-0.006622314453125,
0.00124359130859375,
-0.046539306640625,
-0.046600341796875,
0.08984375,
0.044189453125,
0.06427001953125,
-0.00420379638671875,
0.04827880859375,
-0.005466461181640625,
0.026031494140625,
-0.0504150390625,
0.050872802734375,
-0.01049041748046875,
-0.034698486328125,
-0.0013017654418945312,
-0.015380859375,
-0.067626953125,
0.020294189453125,
0.000009834766387939453,
-0.057830810546875,
0.01190948486328125,
0.00466156005859375,
-0.0290985107421875,
0.0256195068359375,
-0.07159423828125,
0.060089111328125,
-0.03546142578125,
-0.03167724609375,
-0.0188140869140625,
-0.050567626953125,
0.050048828125,
-0.002765655517578125,
0.0017099380493164062,
-0.0223846435546875,
-0.01035308837890625,
0.081787109375,
-0.0413818359375,
0.07293701171875,
-0.020660400390625,
-0.0008521080017089844,
0.037078857421875,
-0.0039215087890625,
0.041168212890625,
0.005733489990234375,
-0.0036468505859375,
0.036163330078125,
0.0007538795471191406,
-0.02545166015625,
-0.0108795166015625,
0.060760498046875,
-0.096923828125,
-0.052978515625,
-0.038787841796875,
-0.057830810546875,
0.0037593841552734375,
0.00785064697265625,
0.01383209228515625,
-0.00007295608520507812,
0.0198974609375,
0.008880615234375,
0.048797607421875,
-0.036163330078125,
0.04296875,
0.038482666015625,
0.001407623291015625,
-0.02630615234375,
0.05450439453125,
0.002696990966796875,
0.0207977294921875,
0.00988006591796875,
0.006816864013671875,
-0.0190887451171875,
-0.03155517578125,
-0.0166778564453125,
0.04656982421875,
-0.04449462890625,
-0.03814697265625,
-0.04071044921875,
-0.0194549560546875,
-0.0167236328125,
0.004070281982421875,
-0.03814697265625,
-0.031646728515625,
-0.049285888671875,
-0.0020732879638671875,
0.047454833984375,
0.039947509765625,
-0.00492095947265625,
0.050933837890625,
-0.0146331787109375,
0.0263671875,
0.020416259765625,
0.024017333984375,
-0.0020351409912109375,
-0.059814453125,
0.00962066650390625,
0.006103515625,
-0.0469970703125,
-0.0543212890625,
0.028656005859375,
0.01318359375,
0.05426025390625,
0.01168060302734375,
0.00196075439453125,
0.067138671875,
-0.0169219970703125,
0.06475830078125,
0.0148468017578125,
-0.06378173828125,
0.051605224609375,
-0.0022487640380859375,
0.00681304931640625,
0.03094482421875,
0.021331787109375,
-0.0102081298828125,
-0.025299072265625,
-0.05401611328125,
-0.064208984375,
0.067626953125,
0.0166473388671875,
-0.0038967132568359375,
0.020416259765625,
0.03717041015625,
0.013671875,
0.005889892578125,
-0.058135986328125,
-0.0276641845703125,
-0.02618408203125,
-0.00408172607421875,
-0.01184844970703125,
-0.023223876953125,
-0.0147705078125,
-0.0289764404296875,
0.056396484375,
-0.00421142578125,
0.0394287109375,
0.01251983642578125,
-0.0289306640625,
-0.0172576904296875,
0.00408172607421875,
0.051910400390625,
0.04412841796875,
-0.03277587890625,
-0.0021152496337890625,
0.027618408203125,
-0.0478515625,
0.01050567626953125,
0.01558685302734375,
-0.00510406494140625,
-0.01422882080078125,
0.0298309326171875,
0.08404541015625,
0.005176544189453125,
-0.047943115234375,
0.0341796875,
-0.003582000732421875,
-0.0147705078125,
-0.022674560546875,
0.0180816650390625,
0.00928497314453125,
0.0286102294921875,
0.022186279296875,
-0.0031299591064453125,
-0.0200347900390625,
-0.0306549072265625,
-0.00765228271484375,
0.03228759765625,
0.005229949951171875,
-0.0323486328125,
0.0645751953125,
0.00359344482421875,
-0.024078369140625,
0.0440673828125,
-0.013763427734375,
-0.02655029296875,
0.056884765625,
0.052490234375,
0.0478515625,
-0.0144805908203125,
0.0007233619689941406,
0.03277587890625,
0.038604736328125,
-0.017333984375,
0.03570556640625,
0.01328277587890625,
-0.032958984375,
-0.0278472900390625,
-0.051055908203125,
-0.02197265625,
0.0302581787109375,
-0.0377197265625,
0.029022216796875,
-0.04638671875,
-0.0237274169921875,
-0.0137176513671875,
0.032989501953125,
-0.0531005859375,
-0.006214141845703125,
0.0079803466796875,
0.07720947265625,
-0.06939697265625,
0.058135986328125,
0.051910400390625,
-0.034271240234375,
-0.06988525390625,
-0.0289154052734375,
-0.00762939453125,
-0.08660888671875,
0.04547119140625,
0.0175323486328125,
0.00820159912109375,
-0.0085296630859375,
-0.052947998046875,
-0.076904296875,
0.1153564453125,
0.050506591796875,
-0.0478515625,
0.0169525146484375,
0.00789642333984375,
0.039947509765625,
-0.0187530517578125,
0.0302886962890625,
0.057586669921875,
0.037078857421875,
0.000021338462829589844,
-0.0892333984375,
0.0189361572265625,
-0.0200347900390625,
0.0084991455078125,
-0.004444122314453125,
-0.08343505859375,
0.08538818359375,
-0.03045654296875,
-0.00800323486328125,
0.0201416015625,
0.047088623046875,
0.0570068359375,
0.0181121826171875,
0.0308074951171875,
0.06378173828125,
0.060302734375,
-0.00794219970703125,
0.0894775390625,
-0.0290679931640625,
0.037017822265625,
0.07489013671875,
-0.008880615234375,
0.07379150390625,
0.041656494140625,
-0.033172607421875,
0.048797607421875,
0.0645751953125,
-0.00841522216796875,
0.0406494140625,
0.008544921875,
0.01299285888671875,
-0.0070037841796875,
-0.00374603271484375,
-0.039886474609375,
0.0279083251953125,
0.024444580078125,
-0.01386260986328125,
-0.00582122802734375,
-0.01399993896484375,
0.0173492431640625,
-0.0304718017578125,
-0.0111541748046875,
0.04095458984375,
0.0208740234375,
-0.0531005859375,
0.0860595703125,
0.00600433349609375,
0.06622314453125,
-0.0418701171875,
0.01409149169921875,
-0.03594970703125,
0.0208740234375,
-0.0263824462890625,
-0.051788330078125,
-0.0009098052978515625,
0.0000909566879272461,
0.00571441650390625,
-0.004238128662109375,
0.04559326171875,
-0.0032024383544921875,
-0.02679443359375,
0.031402587890625,
0.0229644775390625,
0.02337646484375,
0.01251983642578125,
-0.0540771484375,
0.0229034423828125,
-0.0081024169921875,
-0.0341796875,
0.0261383056640625,
0.00629425048828125,
-0.0161895751953125,
0.04986572265625,
0.049835205078125,
-0.002460479736328125,
0.0201568603515625,
-0.01404571533203125,
0.07269287109375,
-0.0293731689453125,
-0.026214599609375,
-0.05584716796875,
0.035858154296875,
0.011383056640625,
-0.045806884765625,
0.05450439453125,
0.038787841796875,
0.055023193359375,
0.007495880126953125,
0.042877197265625,
-0.01174163818359375,
0.0220947265625,
-0.02752685546875,
0.03759765625,
-0.03900146484375,
0.0301971435546875,
-0.00611114501953125,
-0.07562255859375,
-0.007770538330078125,
0.056365966796875,
-0.029815673828125,
-0.005401611328125,
0.0618896484375,
0.0687255859375,
-0.01236724853515625,
-0.0185546875,
-0.01039886474609375,
0.037994384765625,
0.018646240234375,
0.07098388671875,
0.060394287109375,
-0.056549072265625,
0.042205810546875,
-0.042022705078125,
-0.028076171875,
-0.017822265625,
-0.058349609375,
-0.07537841796875,
-0.0321044921875,
-0.03662109375,
-0.0304718017578125,
0.0011186599731445312,
0.05401611328125,
0.041351318359375,
-0.06549072265625,
-0.0390625,
0.0007300376892089844,
0.006931304931640625,
-0.017181396484375,
-0.01399993896484375,
0.03790283203125,
-0.0193328857421875,
-0.035858154296875,
0.014739990234375,
0.004486083984375,
0.01084136962890625,
-0.028839111328125,
-0.024932861328125,
-0.0178680419921875,
-0.01168060302734375,
0.035736083984375,
0.028289794921875,
-0.0701904296875,
-0.007808685302734375,
0.0011816024780273438,
-0.00861358642578125,
0.0170440673828125,
0.0306549072265625,
-0.0616455078125,
0.003566741943359375,
0.0298309326171875,
0.030609130859375,
0.051971435546875,
-0.01273345947265625,
0.0103912353515625,
-0.037445068359375,
0.036041259765625,
-0.004993438720703125,
0.03125,
0.03204345703125,
-0.0207977294921875,
0.04559326171875,
0.0278472900390625,
-0.045166015625,
-0.07806396484375,
-0.00975799560546875,
-0.0902099609375,
-0.00855255126953125,
0.11346435546875,
-0.01503753662109375,
-0.036773681640625,
0.017608642578125,
-0.01959228515625,
0.03228759765625,
-0.0338134765625,
0.057220458984375,
0.0244140625,
-0.0159759521484375,
-0.01427459716796875,
-0.0594482421875,
0.0205230712890625,
0.0312347412109375,
-0.06494140625,
-0.012542724609375,
0.01453399658203125,
0.040069580078125,
0.01485443115234375,
0.036834716796875,
0.0008244514465332031,
0.017425537109375,
-0.014404296875,
0.006744384765625,
-0.01491546630859375,
-0.0002206563949584961,
-0.0280914306640625,
-0.0120849609375,
0.005214691162109375,
-0.0172271728515625
]
] |
CHIH-HUNG/llama-2-13b-FINETUNE1_17w-r4 | 2023-09-12T12:33:46.000Z | [
"transformers",
"pytorch",
"llama",
"text-generation",
"dataset:huangyt/FINETUNE1",
"license:llama2",
"endpoints_compatible",
"text-generation-inference",
"region:us"
] | text-generation | CHIH-HUNG | null | null | CHIH-HUNG/llama-2-13b-FINETUNE1_17w-r4 | 0 | 5,970 | transformers | 2023-09-07T04:17:33 | ---
license: llama2
datasets:
- huangyt/FINETUNE1
---
# Model Card for Model ID
<!-- Provide a quick summary of what the model is/does. -->
在llama-2-13b上使用huangyt/FINETUNE1資料集進行訓練，總資料筆數約17w
# Fine-Tuning Information
- **GPU:** RTX4090 (single core / 24564MiB)
- **model:** meta-llama/Llama-2-13b-hf
- **dataset:** huangyt/FINETUNE1 (共約17w筆訓練集)
- **peft_type:** LoRA
- **lora_rank:** 4
- **lora_target:** gate_proj, up_proj, down_proj
- **per_device_train_batch_size:** 8
- **gradient_accumulation_steps:** 8
- **learning_rate :** 5e-5
- **epoch:** 1
- **precision:** bf16
- **quantization:** load_in_4bit
# Fine-Tuning Detail
- **train_loss:** 0.66
- **train_runtime:** 16:22:29 (use deepspeed)
# Evaluation
- 評估結果來自**HuggingFaceH4/open_llm_leaderboard**
- 與Llama-2-13b比較4種Benchmark，包含**ARC**、**HellaSwag**、**MMLU**、**TruthfulQA**
| Model |Average| ARC |HellaSwag| MMLU |TruthfulQA|
|--------------------------------------------------------|-------|-------|---------|-------|----------|
|meta-llama/Llama-2-13b-hf | 56.9 | 58.11 | 80.97 | 54.34 | 34.17 |
|meta-llama/Llama-2-13b-chat-hf | 59.93 | 59.04 | 81.94 | 54.64 | 44.12 |
|CHIH-HUNG/llama-2-13b-Fintune_1_17w | 58.24 | 59.47 | 81 | 54.31 | 38.17 |
|CHIH-HUNG/llama-2-13b-huangyt_Fintune_1_17w-q_k_v_o_proj| 58.49 | 59.73 | 81.06 | 54.53 | 38.64 |
|CHIH-HUNG/llama-2-13b-Fintune_1_17w-gate_up_down_proj | 58.81 | 57.17 | 82.26 | 55.89 | 39.93 |
|CHIH-HUNG/llama-2-13b-FINETUNE1_17w-r16 | 58.86 | 57.25 | 82.27 | 56.16 | 39.75 |
|CHIH-HUNG/llama-2-13b-FINETUNE1_17w-r4 | 58.71 | 56.74 | 82.27 | 56.18 | 39.65 |
# How to convert dataset to json
- 在**load_dataset**中輸入資料集名稱，並且在**take**中輸入要取前幾筆資料
- 觀察該資料集的欄位名稱，填入**example**欄位中(例如system_prompt、question、response)
- 最後指定json檔儲存位置 (**json_filename**)
```py
import json
from datasets import load_dataset
# 讀取數據集，take可以取得該數據集前n筆資料
dataset = load_dataset("huangyt/FINETUNE1", split="train", streaming=True)
# 提取所需欄位並建立新的字典列表
extracted_data = []
for example in dataset:
extracted_example = {
"instruction": example["instruction"],
"input": example["input"],
"output": example["output"]
}
extracted_data.append(extracted_example)
# 指定 JSON 文件名稱
json_filename = "huangyt_FINETUNE_1.json"
# 寫入 JSON 文件
with open(json_filename, "w") as json_file:
json.dump(extracted_data, json_file, indent=4)
print(f"數據已提取並保存為 {json_filename}")
``` | 2,580 | [
[
-0.045623779296875,
-0.048614501953125,
0.0135498046875,
0.01236724853515625,
-0.0479736328125,
0.004543304443359375,
-0.0112152099609375,
-0.01947021484375,
0.0186004638671875,
0.032623291015625,
-0.04510498046875,
-0.0408935546875,
-0.042877197265625,
0.01343536376953125,
-0.019989013671875,
0.08270263671875,
-0.0089263916015625,
-0.01268768310546875,
0.0194244384765625,
0.00612640380859375,
-0.0382080078125,
-0.024444580078125,
-0.05230712890625,
-0.0295867919921875,
0.0242919921875,
0.01708984375,
0.049591064453125,
0.06988525390625,
0.054962158203125,
0.0206756591796875,
-0.012969970703125,
0.01873779296875,
-0.04608154296875,
-0.021881103515625,
0.0188140869140625,
-0.04302978515625,
-0.047576904296875,
-0.0058441162109375,
0.04962158203125,
0.024139404296875,
0.00408172607421875,
0.04443359375,
0.0167999267578125,
0.044647216796875,
-0.0237579345703125,
0.0203857421875,
-0.0240936279296875,
0.00914764404296875,
-0.0253753662109375,
-0.0274658203125,
-0.0018587112426757812,
-0.0250396728515625,
-0.0106048583984375,
-0.0679931640625,
0.0058135986328125,
0.0112457275390625,
0.1046142578125,
0.03143310546875,
-0.0207061767578125,
0.00713348388671875,
-0.036895751953125,
0.06304931640625,
-0.07708740234375,
0.00037097930908203125,
0.0264892578125,
0.0288848876953125,
-0.00934600830078125,
-0.05096435546875,
-0.054412841796875,
0.00815582275390625,
-0.0137481689453125,
0.01538848876953125,
-0.00899505615234375,
-0.0196380615234375,
0.0260772705078125,
0.037506103515625,
-0.032470703125,
0.00299072265625,
-0.037353515625,
0.006801605224609375,
0.063720703125,
0.032073974609375,
0.005443572998046875,
-0.0247344970703125,
-0.0221099853515625,
-0.0197906494140625,
-0.039276123046875,
0.0201568603515625,
0.032196044921875,
0.031524658203125,
-0.038543701171875,
0.03466796875,
-0.03826904296875,
0.03265380859375,
0.01103973388671875,
-0.0307159423828125,
0.048004150390625,
-0.0194244384765625,
-0.041473388671875,
0.0024585723876953125,
0.0775146484375,
0.044647216796875,
-0.00547027587890625,
0.017242431640625,
-0.008636474609375,
-0.0137939453125,
-0.00449371337890625,
-0.06744384765625,
-0.0233001708984375,
0.040191650390625,
-0.0535888671875,
-0.0338134765625,
0.008544921875,
-0.06494140625,
-0.005939483642578125,
-0.00888824462890625,
0.02227783203125,
-0.023834228515625,
-0.044647216796875,
0.0005831718444824219,
-0.01262664794921875,
0.02545166015625,
0.025238037109375,
-0.058929443359375,
0.0121002197265625,
0.04754638671875,
0.053863525390625,
0.00844573974609375,
-0.024383544921875,
-0.006526947021484375,
0.01397705078125,
-0.0267333984375,
0.049591064453125,
-0.006046295166015625,
-0.0290374755859375,
-0.01708984375,
0.020843505859375,
-0.003253936767578125,
-0.038543701171875,
0.058685302734375,
-0.03173828125,
-0.006656646728515625,
-0.038848876953125,
-0.0199432373046875,
-0.035675048828125,
0.035491943359375,
-0.05291748046875,
0.07830810546875,
0.006504058837890625,
-0.06640625,
0.0243072509765625,
-0.05145263671875,
-0.01436614990234375,
0.004650115966796875,
0.002105712890625,
-0.03607177734375,
-0.021270751953125,
0.0213165283203125,
0.041595458984375,
-0.03466796875,
0.01497650146484375,
-0.017425537109375,
-0.043914794921875,
0.02197265625,
-0.0295867919921875,
0.07354736328125,
0.031494140625,
-0.01702880859375,
0.00371551513671875,
-0.0721435546875,
0.00457763671875,
0.045806884765625,
-0.038848876953125,
-0.004680633544921875,
-0.00928497314453125,
0.002712249755859375,
-0.0028858184814453125,
0.0308990478515625,
-0.01715087890625,
0.0266571044921875,
-0.0145416259765625,
0.0316162109375,
0.06842041015625,
0.0017261505126953125,
0.0098114013671875,
-0.039276123046875,
0.0250396728515625,
0.008941650390625,
0.0200653076171875,
-0.0032787322998046875,
-0.0345458984375,
-0.074462890625,
-0.02032470703125,
0.0108184814453125,
0.0401611328125,
-0.033447265625,
0.052886962890625,
-0.0243682861328125,
-0.05389404296875,
-0.05517578125,
0.0050811767578125,
0.0188446044921875,
0.04083251953125,
0.039093017578125,
0.008087158203125,
-0.053741455078125,
-0.06640625,
0.0025615692138671875,
-0.0049285888671875,
0.007335662841796875,
0.0258636474609375,
0.050140380859375,
-0.0246734619140625,
0.04034423828125,
-0.038360595703125,
-0.02337646484375,
-0.0248565673828125,
0.00009250640869140625,
0.06927490234375,
0.04351806640625,
0.05059814453125,
-0.03704833984375,
-0.033782958984375,
0.006183624267578125,
-0.08465576171875,
0.01221466064453125,
-0.00682830810546875,
-0.0206756591796875,
-0.00774383544921875,
0.0025844573974609375,
-0.04736328125,
0.03350830078125,
0.034698486328125,
-0.017303466796875,
0.04315185546875,
0.007495880126953125,
0.025604248046875,
-0.0784912109375,
0.0130157470703125,
-0.01715087890625,
0.006175994873046875,
-0.0333251953125,
0.01520538330078125,
-0.012939453125,
0.0222625732421875,
-0.0287628173828125,
0.0229644775390625,
-0.0249481201171875,
0.01026153564453125,
-0.0138092041015625,
-0.0027523040771484375,
0.0014390945434570312,
0.048126220703125,
-0.01227569580078125,
0.0478515625,
0.03961181640625,
-0.05596923828125,
0.04241943359375,
0.034912109375,
-0.0298309326171875,
0.0147247314453125,
-0.039398193359375,
0.002147674560546875,
0.005649566650390625,
0.02227783203125,
-0.07330322265625,
-0.0259552001953125,
0.044677734375,
-0.031524658203125,
0.016387939453125,
-0.0282440185546875,
-0.0274505615234375,
-0.04925537109375,
-0.0302276611328125,
0.0221710205078125,
0.024017333984375,
-0.0445556640625,
0.0163421630859375,
0.01050567626953125,
0.015106201171875,
-0.05169677734375,
-0.0638427734375,
-0.005718231201171875,
-0.0197906494140625,
-0.03594970703125,
0.0177001953125,
-0.01094818115234375,
-0.00806427001953125,
0.005157470703125,
-0.0013675689697265625,
-0.0015707015991210938,
0.01018524169921875,
0.013397216796875,
0.03570556640625,
-0.0247344970703125,
-0.02911376953125,
0.006084442138671875,
-0.0081634521484375,
0.003719329833984375,
0.01194000244140625,
0.060760498046875,
-0.017059326171875,
-0.016693115234375,
-0.0594482421875,
0.004749298095703125,
0.02740478515625,
0.004058837890625,
0.04376220703125,
0.05828857421875,
-0.017974853515625,
0.004909515380859375,
-0.0191650390625,
-0.0020885467529296875,
-0.038177490234375,
0.024383544921875,
-0.0440673828125,
-0.052703857421875,
0.0526123046875,
-0.0018177032470703125,
0.0190887451171875,
0.06396484375,
0.0266876220703125,
-0.01617431640625,
0.0751953125,
0.0136260986328125,
-0.01959228515625,
0.0180816650390625,
-0.0714111328125,
0.005458831787109375,
-0.0755615234375,
-0.0256805419921875,
-0.036773681640625,
-0.044647216796875,
-0.048431396484375,
-0.01351165771484375,
0.016937255859375,
0.021575927734375,
-0.048736572265625,
0.031341552734375,
-0.062744140625,
0.02203369140625,
0.04547119140625,
0.01654052734375,
0.0167236328125,
-0.007198333740234375,
0.01015472412109375,
0.003204345703125,
-0.038330078125,
-0.033935546875,
0.0977783203125,
0.025299072265625,
0.0516357421875,
0.004241943359375,
0.05462646484375,
0.01004791259765625,
0.01015472412109375,
-0.04833984375,
0.046661376953125,
-0.0004343986511230469,
-0.05255126953125,
-0.01389312744140625,
-0.022735595703125,
-0.050750732421875,
0.0274658203125,
-0.0164947509765625,
-0.05657958984375,
0.0079345703125,
0.002712249755859375,
-0.034820556640625,
0.042724609375,
-0.03204345703125,
0.052764892578125,
-0.0283660888671875,
-0.0252532958984375,
0.0018758773803710938,
-0.041259765625,
0.0538330078125,
0.00701141357421875,
0.011993408203125,
-0.0247802734375,
0.0083770751953125,
0.08123779296875,
-0.0438232421875,
0.04541015625,
-0.0225830078125,
-0.0027217864990234375,
0.04083251953125,
0.00406646728515625,
0.052276611328125,
0.023223876953125,
-0.001689910888671875,
0.04254150390625,
0.003643035888671875,
-0.016632080078125,
-0.0231781005859375,
0.05657958984375,
-0.08868408203125,
-0.047821044921875,
-0.04345703125,
-0.025299072265625,
0.016632080078125,
0.0276031494140625,
0.038330078125,
-0.00540924072265625,
0.01409149169921875,
0.0197296142578125,
0.03466796875,
-0.004467010498046875,
0.0419921875,
0.021209716796875,
-0.0152740478515625,
-0.055023193359375,
0.060394287109375,
0.0034942626953125,
-0.0008640289306640625,
0.0283355712890625,
0.01001739501953125,
-0.0186309814453125,
-0.045379638671875,
-0.04302978515625,
0.018585205078125,
-0.039276123046875,
-0.046661376953125,
-0.036773681640625,
-0.036712646484375,
-0.038360595703125,
-0.0018796920776367188,
-0.04083251953125,
-0.017333984375,
-0.05804443359375,
-0.0122528076171875,
0.051513671875,
0.030853271484375,
-0.004840850830078125,
0.054595947265625,
-0.059600830078125,
0.0285491943359375,
0.01332855224609375,
0.01290130615234375,
0.00818634033203125,
-0.062286376953125,
-0.0228729248046875,
0.00769805908203125,
-0.0333251953125,
-0.046600341796875,
0.045318603515625,
-0.0012693405151367188,
0.039581298828125,
0.05859375,
0.00010317564010620117,
0.0867919921875,
-0.0149383544921875,
0.0677490234375,
0.015960693359375,
-0.052581787109375,
0.0408935546875,
-0.033050537109375,
-0.0080413818359375,
0.03790283203125,
0.0241851806640625,
-0.0296478271484375,
-0.0032596588134765625,
-0.03863525390625,
-0.060333251953125,
0.07647705078125,
0.0135955810546875,
-0.006622314453125,
0.0198516845703125,
0.01666259765625,
0.0073394775390625,
0.0187530517578125,
-0.06549072265625,
-0.04681396484375,
-0.036529541015625,
-0.002674102783203125,
0.00487518310546875,
-0.01129150390625,
-0.0287628173828125,
-0.037628173828125,
0.05682373046875,
-0.0023097991943359375,
0.039459228515625,
0.0126495361328125,
0.01454925537109375,
-0.0178985595703125,
0.007732391357421875,
0.0299224853515625,
0.032623291015625,
-0.04217529296875,
-0.00841522216796875,
0.0113677978515625,
-0.041412353515625,
0.001956939697265625,
0.00954437255859375,
-0.019195556640625,
-0.010711669921875,
0.035552978515625,
0.065673828125,
0.0003802776336669922,
-0.026397705078125,
0.0218353271484375,
0.0036449432373046875,
-0.0240631103515625,
-0.03265380859375,
0.0208587646484375,
-0.003475189208984375,
0.037200927734375,
0.04229736328125,
0.0017881393432617188,
0.007442474365234375,
-0.0235443115234375,
-0.00933074951171875,
0.0214385986328125,
0.01229095458984375,
-0.018829345703125,
0.0687255859375,
0.003787994384765625,
-0.011138916015625,
0.041717529296875,
-0.013946533203125,
-0.03387451171875,
0.05810546875,
0.039337158203125,
0.05682373046875,
-0.01080322265625,
-0.002635955810546875,
0.06134033203125,
0.0310516357421875,
-0.010528564453125,
0.04034423828125,
-0.0023517608642578125,
-0.04888916015625,
-0.0137481689453125,
-0.05474853515625,
-0.0088043212890625,
0.042724609375,
-0.052459716796875,
0.0223541259765625,
-0.054931640625,
-0.022247314453125,
-0.005725860595703125,
0.02618408203125,
-0.053375244140625,
0.0208587646484375,
0.01012420654296875,
0.0650634765625,
-0.05511474609375,
0.067626953125,
0.025726318359375,
-0.04180908203125,
-0.072509765625,
-0.0201873779296875,
-0.01177978515625,
-0.0732421875,
0.040740966796875,
0.01218414306640625,
0.0193634033203125,
-0.0007824897766113281,
-0.067626953125,
-0.079833984375,
0.10858154296875,
0.0134429931640625,
-0.04718017578125,
0.0087127685546875,
0.01471710205078125,
0.0250244140625,
-0.0130157470703125,
0.0305938720703125,
0.05462646484375,
0.048553466796875,
0.00290679931640625,
-0.060089111328125,
0.0237884521484375,
-0.035064697265625,
-0.010009765625,
0.0011987686157226562,
-0.08978271484375,
0.10015869140625,
-0.01296234130859375,
0.0023670196533203125,
0.0096588134765625,
0.051910400390625,
0.041168212890625,
0.0271453857421875,
0.0278778076171875,
0.054901123046875,
0.0511474609375,
-0.0239410400390625,
0.054046630859375,
-0.00732421875,
0.04156494140625,
0.06256103515625,
-0.00628662109375,
0.055999755859375,
0.0304107666015625,
-0.038421630859375,
0.037750244140625,
0.06964111328125,
-0.03369140625,
0.0526123046875,
-0.00916290283203125,
-0.007045745849609375,
-0.01181793212890625,
0.002361297607421875,
-0.0550537109375,
0.0258636474609375,
0.0292816162109375,
-0.027191162109375,
0.00567626953125,
-0.0205535888671875,
0.01654052734375,
-0.0269775390625,
-0.0250244140625,
0.041259765625,
-0.01222991943359375,
-0.026458740234375,
0.076171875,
-0.006832122802734375,
0.057525634765625,
-0.045928955078125,
-0.01107025146484375,
-0.0167694091796875,
0.0135040283203125,
-0.037017822265625,
-0.06170654296875,
-0.0014276504516601562,
0.0026416778564453125,
-0.01155853271484375,
0.014862060546875,
0.0343017578125,
-0.00933837890625,
-0.03704833984375,
0.027099609375,
0.0053863525390625,
0.0243072509765625,
0.00860595703125,
-0.066162109375,
0.026336669921875,
0.019317626953125,
-0.04296875,
0.0188140869140625,
0.023223876953125,
0.0222625732421875,
0.053619384765625,
0.07080078125,
0.0052642822265625,
0.0155029296875,
-0.0102081298828125,
0.07794189453125,
-0.06207275390625,
-0.029296875,
-0.05743408203125,
0.03668212890625,
-0.0175323486328125,
-0.0384521484375,
0.05560302734375,
0.0567626953125,
0.06494140625,
-0.0027675628662109375,
0.0714111328125,
-0.0225830078125,
0.037567138671875,
-0.031494140625,
0.058349609375,
-0.055572509765625,
0.011688232421875,
-0.0226898193359375,
-0.04156494140625,
-0.007366180419921875,
0.060272216796875,
-0.00461578369140625,
-0.0029315948486328125,
0.04290771484375,
0.04376220703125,
-0.00084686279296875,
0.01129150390625,
0.001964569091796875,
0.0254364013671875,
0.0279541015625,
0.06427001953125,
0.047821044921875,
-0.07733154296875,
0.054656982421875,
-0.052978515625,
-0.006641387939453125,
-0.0285797119140625,
-0.047393798828125,
-0.0640869140625,
-0.0197296142578125,
-0.0186004638671875,
-0.029296875,
-0.0206298828125,
0.0638427734375,
0.0386962890625,
-0.058624267578125,
-0.0278778076171875,
0.0011396408081054688,
0.0086212158203125,
-0.033782958984375,
-0.02191162109375,
0.05096435546875,
0.00592041015625,
-0.060089111328125,
0.0272216796875,
-0.00926971435546875,
0.00839996337890625,
-0.00372314453125,
-0.0217437744140625,
-0.019927978515625,
-0.0229034423828125,
0.02685546875,
0.0233001708984375,
-0.0526123046875,
-0.013702392578125,
-0.01322174072265625,
-0.0010318756103515625,
0.020904541015625,
0.0159759521484375,
-0.037261962890625,
0.00937652587890625,
0.037445068359375,
0.0252838134765625,
0.0467529296875,
-0.00225830078125,
-0.0038928985595703125,
-0.0304412841796875,
0.021484375,
-0.00026226043701171875,
0.0268402099609375,
0.006397247314453125,
-0.038848876953125,
0.05548095703125,
0.035247802734375,
-0.04681396484375,
-0.07647705078125,
-0.031341552734375,
-0.09698486328125,
-0.01226043701171875,
0.0831298828125,
-0.004337310791015625,
-0.04559326171875,
0.0192108154296875,
-0.0230865478515625,
0.0433349609375,
-0.044769287109375,
0.048309326171875,
0.03076171875,
-0.0092926025390625,
-0.00405120849609375,
-0.052947998046875,
0.0285797119140625,
-0.004619598388671875,
-0.052459716796875,
-0.00250244140625,
0.00722503662109375,
0.0228271484375,
0.02130126953125,
0.035491943359375,
0.0043487548828125,
0.00876617431640625,
0.013153076171875,
0.00910186767578125,
-0.0197906494140625,
-0.008880615234375,
-0.004276275634765625,
-0.0071868896484375,
-0.0207366943359375,
-0.04376220703125
]
] |
pythainlp/wangchanglm-7.5B-sft-enth | 2023-05-29T15:24:36.000Z | [
"transformers",
"pytorch",
"xglm",
"text-generation",
"en",
"th",
"ja",
"vi",
"dataset:laion/OIG",
"dataset:Hello-SimpleAI/HC3",
"dataset:databricks/databricks-dolly-15k",
"license:cc-by-sa-4.0",
"endpoints_compatible",
"has_space",
"region:us"
] | text-generation | pythainlp | null | null | pythainlp/wangchanglm-7.5B-sft-enth | 6 | 5,969 | transformers | 2023-04-25T04:37:10 | ---
license: cc-by-sa-4.0
datasets:
- laion/OIG
- Hello-SimpleAI/HC3
- databricks/databricks-dolly-15k
language:
- en
- th
- ja
- vi
pipeline_tag: text-generation
---
# Model Card for WangChanGLM 🐘 - The Multilingual Instruction-Following Model
<!-- Provide a longer summary of what this model is. -->
WangChanGLM is a multilingual, instruction-finetuned Facebook XGLM-7.5B using open-source, commercially permissible datasets (LAION OIG chip2 and infill_dbpedia, DataBricks Dolly v2, OpenAI TL;DR, and Hello-SimpleAI HC3; about 400k examples), released under CC-BY SA 4.0. The models are trained to perform a subset of instruction-following tasks we found most relevant namely: reading comprehension, brainstorming, and creative writing. We provide the weights for a model finetuned on an English-only dataset ([wangchanglm-7.5B-sft-en](https://huggingface.co/pythainlp/wangchanglm-7.5B-sft-en)) and another checkpoint further finetuned on Google-Translated Thai dataset ([wangchanglm-7.5B-sft-enth](https://huggingface.co/pythainlp/wangchanglm-7.5B-sft-enth)). We perform Vicuna-style evaluation using both humans and ChatGPT (in our case, `gpt-3.5-turbo` since we are still on the waitlist for `gpt-4`) and observe some discrepancies between the two types of annoators. All training and evaluation codes are shared under the [Apache-2.0 license](https://github.com/pythainlp/wangchanglm/blob/main/LICENSE) in our Github, as well as datasets and model weights on [HuggingFace](https://huggingface.co/pythainlp). In a similar manner to [Dolly v2](https://www.databricks.com/blog/2023/04/12/dolly-first-open-commercially-viable-instruction-tuned-llm), we use only use open-source, commercially permissive pretrained models and datasets, our models are neither restricted by non-commercial clause like models that use LLaMA as base nor non-compete clause like models that use self-instruct datasets from ChatGPT. See our live demo [here]().
- **Developed by:** [PyThaiNLP](https://www.github.com/pythainlp) and [VISTEC-depa AI Research Institute of Thailand](https://huggingface.co/airesearch)
- **Model type:** Finetuned [XGLM-7.5B](https://huggingface.co/facebook/xglm-7.5B)
- **Language(s) (NLP)**: `en`, `th`, `ja`, `vi` capacibilities evaluated, theoretically all 30 languages of [XGLM-7.5B](https://huggingface.co/facebook/xglm-7.5B)
- **License:** [CC-BY SA 4.0](https://creativecommons.org/licenses/by-sa/4.0/)
### Model Sources
<!-- Provide the basic links for the model. -->
- **Repository:** [pythainlp/wangchanglm](https://www.github.com/pythainlp/wangchanglm)
- **Blog:** [Medium](https://link.medium.com/s2MWr3ZXnzb)
- **Demo:** [Colab notebook](https://colab.research.google.com/github/pythainlp/WangChanGLM/blob/main/demo/WangChanGLM_v0_1_demo.ipynb)
## Uses
<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
### Direct Use
<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
Intended to be use as an instruction-following model for reading comprehension, brainstorming and creative writing.
### Downstream Use
<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
The model can be finetuned for any typical instruction-following use cases.
### Out-of-Scope Use
<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
We do not expect the models to perform well in math problems, reasoning, and factfulness. We intentionally filter out training examples from these use cases.
## Bias, Risks, and Limitations
<!-- This section is meant to convey both technical and sociotechnical limitations. -->
We noticed similar limitations to other finetuned instruction followers such as math problems, reasoning, and factfulness. Even though the models do not perform on the level that we expect them to be abused, they do contain undesirable biases and toxicity and should be further optimized for your particular use cases.
### Recommendations
<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
## How to Get Started with the Model
Use the code below to get started with the model.
```
model_name = "pythainlp/wangchanglm-7.5B-sft-en"
model = AutoModelForCausalLM.from_pretrained(
model_name,
return_dict=True,
load_in_8bit=True ,
device_map="auto",
torch_dtype=torch.float16,
offload_folder="./",
low_cpu_mem_usage=True,
)
text = "เล่นหุ้นยังไงให้รวย"
tokenizer = AutoTokenizer.from_pretrained(model_name)
batch = tokenizer(text, return_tensors="pt")
with torch.cuda.amp.autocast():
output_tokens = model.generate(
input_ids=batch["input_ids"],
max_new_tokens=max_gen_len, # 512
begin_suppress_tokens = exclude_ids,
no_repeat_ngram_size=2,
#oasst k50
top_k=50,
top_p=top_p, # 0.95
typical_p=1.,
temperature=temperature, # 0.9
# #oasst typical3
# typical_p = 0.3,
# temperature = 0.8,
# repetition_penalty = 1.2,
)
tokenizer.decode(output_tokens[0], skip_special_tokens=True)
```
## Training Details
### Training Data
<!-- This should link to a Data Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
Finetuning datasets are sourced from [LAION OIG chip2 and infill_dbpedia](https://huggingface.co/datasets/laion/OIG) ([Apache-2.0](https://github.com/pythainlp/wangchanglm/blob/main/LICENSE)), [DataBricks Dolly v2](https://github.com/databrickslabs/dolly) ([Apache-2.0](https://github.com/pythainlp/wangchanglm/blob/main/LICENSE)), [OpenAI TL;DR](https://github.com/openai/summarize-from-feedback) ([MIT](https://opensource.org/license/mit/)), and [Hello-SimpleAI HC3](https://huggingface.co/datasets/Hello-SimpleAI/HC3) ([CC-BY SA](https://creativecommons.org/licenses/by-sa/4.0/)).
### Training Procedure
<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
#### Preprocessing
See [pythainlp/wangchanglm](https://www.github.com/pythainlp/wangchanglm).
#### Training Hyperparameters
- **Training regime:** LoRA with 4 GPUs. See more details at [pythainlp/wangchanglm](https://www.github.com/pythainlp/wangchanglm).
## Evaluation
<!-- This section describes the evaluation protocols and provides the results. -->
We performed automatic evaluation in the style of [Vicuna](https://vicuna.lmsys.org/) and human evaluation. See more details from our [blog]().
## Environmental Impact
<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
Experiments were conducted using a private infrastructure, which has a carbon efficiency of 0.432 kgCO2eq/kWh. A cumulative of 500 hours of computation was performed on hardware of type Tesla V100-SXM2-32GB (TDP of 300W). Total emissions are estimated to be 64.8 CO2eq of which 0 percents were directly offset. Estimations were conducted using the [MachineLearning Impact calculator](https://mlco2.github.io/impact#compute).
## Citation
<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
**BibTeX:**
```
@software{charin_polpanumas_2023_7878101,
author = {Charin Polpanumas and
Wannaphong Phatthiyaphaibun and
Patomporn Payoungkhamdee and
Peerat Limkonchotiwat and
Lalita Lowphansirikul and
Can Udomcharoenchaikit and
Titipat Achakulwisut and
Ekapol Chuangsuwanich and
Sarana Nutanong},
title = {{WangChanGLM🐘 — The Multilingual Instruction-
Following Model}},
month = apr,
year = 2023,
publisher = {Zenodo},
version = {v0.1},
doi = {10.5281/zenodo.7878101},
url = {https://doi.org/10.5281/zenodo.7878101}
}
```
## Model Card Contact
[PyThaiNLP](https://github.com/pythainlp) | 8,516 | [
[
-0.0186920166015625,
-0.057861328125,
0.0034656524658203125,
0.01690673828125,
-0.010894775390625,
-0.0257415771484375,
-0.0093841552734375,
-0.035858154296875,
-0.0011205673217773438,
0.0243072509765625,
-0.035858154296875,
-0.03363037109375,
-0.0277099609375,
-0.01010894775390625,
-0.02362060546875,
0.08856201171875,
-0.0108795166015625,
0.0093841552734375,
-0.003475189208984375,
-0.00669097900390625,
-0.02764892578125,
-0.035736083984375,
-0.04510498046875,
-0.0263671875,
0.0270233154296875,
0.0256500244140625,
0.0406494140625,
0.050323486328125,
0.037017822265625,
0.0169830322265625,
-0.0103607177734375,
0.012725830078125,
-0.04327392578125,
-0.02691650390625,
0.0111236572265625,
-0.020233154296875,
-0.042236328125,
0.01042938232421875,
0.0350341796875,
0.041900634765625,
-0.0207672119140625,
0.03143310546875,
0.01018524169921875,
0.045684814453125,
-0.037841796875,
0.02435302734375,
-0.038238525390625,
0.0132293701171875,
-0.011566162109375,
0.0099334716796875,
-0.033599853515625,
-0.0278778076171875,
0.022674560546875,
-0.059173583984375,
0.01296234130859375,
0.01418304443359375,
0.08612060546875,
0.00455474853515625,
-0.026458740234375,
-0.022705078125,
-0.0309906005859375,
0.059844970703125,
-0.0655517578125,
0.03436279296875,
0.0401611328125,
0.01213836669921875,
-0.00382232666015625,
-0.06573486328125,
-0.04559326171875,
-0.0288238525390625,
-0.0174102783203125,
0.019012451171875,
-0.002445220947265625,
0.0019702911376953125,
0.03350830078125,
0.040771484375,
-0.051513671875,
0.0025482177734375,
-0.048095703125,
-0.0173187255859375,
0.06109619140625,
0.018280029296875,
0.0167694091796875,
-0.039459228515625,
-0.037017822265625,
-0.0145111083984375,
-0.040313720703125,
0.022918701171875,
0.03533935546875,
0.01800537109375,
-0.044464111328125,
0.038177490234375,
-0.0172576904296875,
0.05474853515625,
0.00789642333984375,
-0.004619598388671875,
0.03204345703125,
-0.0272979736328125,
-0.036407470703125,
-0.0164794921875,
0.07598876953125,
0.0229339599609375,
-0.0129241943359375,
-0.00146484375,
-0.00765228271484375,
-0.0282745361328125,
-0.0097503662109375,
-0.07373046875,
-0.050445556640625,
0.016387939453125,
-0.038330078125,
-0.03472900390625,
0.0044403076171875,
-0.059295654296875,
-0.0117645263671875,
-0.02874755859375,
0.0546875,
-0.034393310546875,
-0.0275115966796875,
0.0003674030303955078,
0.0004341602325439453,
0.03045654296875,
0.01702880859375,
-0.0760498046875,
0.01898193359375,
0.02740478515625,
0.07098388671875,
-0.010894775390625,
-0.03369140625,
-0.0220184326171875,
-0.005840301513671875,
-0.01302337646484375,
0.035858154296875,
-0.004730224609375,
-0.032012939453125,
-0.0167388916015625,
0.01380157470703125,
-0.00725555419921875,
-0.02642822265625,
0.04510498046875,
-0.02972412109375,
0.0435791015625,
-0.022857666015625,
-0.043426513671875,
-0.02984619140625,
0.01207733154296875,
-0.04949951171875,
0.09326171875,
0.01039886474609375,
-0.06268310546875,
0.01256561279296875,
-0.055206298828125,
-0.0272216796875,
-0.0193634033203125,
0.00734710693359375,
-0.04998779296875,
-0.0216217041015625,
0.01206207275390625,
0.032684326171875,
-0.0322265625,
0.0299530029296875,
-0.0117034912109375,
-0.01953125,
-0.002590179443359375,
-0.04296875,
0.1065673828125,
0.017333984375,
-0.0325927734375,
0.0183258056640625,
-0.06427001953125,
0.007282257080078125,
0.0198822021484375,
-0.0234375,
-0.0005154609680175781,
-0.025848388671875,
0.021026611328125,
0.018768310546875,
0.0197601318359375,
-0.029632568359375,
0.01708984375,
-0.024688720703125,
0.0377197265625,
0.06292724609375,
-0.016845703125,
0.0179290771484375,
-0.0322265625,
0.0305328369140625,
0.0013332366943359375,
0.033447265625,
-0.01030731201171875,
-0.040130615234375,
-0.0736083984375,
-0.01103973388671875,
0.02093505859375,
0.04144287109375,
-0.0430908203125,
0.034210205078125,
0.0083770751953125,
-0.06658935546875,
-0.039215087890625,
0.007198333740234375,
0.034576416015625,
0.06365966796875,
0.045135498046875,
-0.0062103271484375,
-0.04058837890625,
-0.046875,
0.00582122802734375,
-0.00479888916015625,
0.004302978515625,
0.0194549560546875,
0.04766845703125,
-0.01268768310546875,
0.053985595703125,
-0.04833984375,
-0.01763916015625,
-0.00574493408203125,
0.00801849365234375,
0.0191650390625,
0.051849365234375,
0.037261962890625,
-0.0528564453125,
-0.052459716796875,
0.004291534423828125,
-0.0611572265625,
0.0014810562133789062,
-0.00469207763671875,
-0.029815673828125,
0.02734375,
0.0300140380859375,
-0.061126708984375,
0.03778076171875,
0.0447998046875,
-0.026123046875,
0.06658935546875,
-0.022308349609375,
-0.0002872943878173828,
-0.07843017578125,
0.018402099609375,
-0.0041961669921875,
-0.0028476715087890625,
-0.0428466796875,
0.0019054412841796875,
0.0081634521484375,
-0.01343536376953125,
-0.0380859375,
0.06256103515625,
-0.03619384765625,
0.0142822265625,
-0.00604248046875,
0.005481719970703125,
0.01678466796875,
0.053955078125,
0.000186920166015625,
0.04937744140625,
0.04559326171875,
-0.046661376953125,
0.03668212890625,
0.0185089111328125,
-0.0229339599609375,
0.01336669921875,
-0.0631103515625,
0.01568603515625,
0.007717132568359375,
0.01119232177734375,
-0.057464599609375,
-0.0218353271484375,
0.03900146484375,
-0.0264892578125,
0.030914306640625,
0.004474639892578125,
-0.029632568359375,
-0.02972412109375,
-0.029632568359375,
0.0223236083984375,
0.04901123046875,
-0.049560546875,
0.03375244140625,
0.0201568603515625,
0.015777587890625,
-0.043914794921875,
-0.055267333984375,
-0.025543212890625,
-0.0191650390625,
-0.047210693359375,
0.0341796875,
-0.01026153564453125,
0.00007253885269165039,
0.004055023193359375,
-0.0031337738037109375,
-0.0117645263671875,
-0.006229400634765625,
0.010345458984375,
0.0377197265625,
-0.0120391845703125,
0.00995635986328125,
-0.00908660888671875,
-0.0052490234375,
0.0008292198181152344,
-0.0214080810546875,
0.036773681640625,
-0.0167083740234375,
-0.00849151611328125,
-0.060516357421875,
-0.002719879150390625,
0.0288848876953125,
-0.016021728515625,
0.0621337890625,
0.0634765625,
-0.030853271484375,
0.002155303955078125,
-0.030548095703125,
-0.0193939208984375,
-0.036529541015625,
0.039306640625,
-0.0092010498046875,
-0.035675048828125,
0.043701171875,
0.025299072265625,
0.02093505859375,
0.049041748046875,
0.034210205078125,
-0.003330230712890625,
0.059600830078125,
0.04095458984375,
0.0006508827209472656,
0.04437255859375,
-0.05120849609375,
0.007526397705078125,
-0.0736083984375,
-0.0280914306640625,
-0.02783203125,
-0.002620697021484375,
-0.060089111328125,
-0.035797119140625,
0.01030731201171875,
0.01568603515625,
-0.0399169921875,
0.03472900390625,
-0.044525146484375,
0.0175018310546875,
0.05267333984375,
0.01497650146484375,
0.00958251953125,
-0.005352020263671875,
-0.005329132080078125,
-0.0017995834350585938,
-0.06109619140625,
-0.026824951171875,
0.09808349609375,
0.029266357421875,
0.0570068359375,
-0.0042266845703125,
0.041595458984375,
0.0019931793212890625,
0.0246124267578125,
-0.04931640625,
0.035919189453125,
-0.007480621337890625,
-0.050567626953125,
-0.031402587890625,
-0.041107177734375,
-0.070556640625,
0.019866943359375,
-0.004913330078125,
-0.050079345703125,
0.00678253173828125,
0.0208892822265625,
-0.033447265625,
0.0293426513671875,
-0.072021484375,
0.09027099609375,
-0.035247802734375,
-0.040191650390625,
-0.00531005859375,
-0.04254150390625,
0.03814697265625,
-0.0018463134765625,
0.004436492919921875,
-0.00978851318359375,
0.016387939453125,
0.06976318359375,
-0.0528564453125,
0.058807373046875,
-0.020965576171875,
0.0010805130004882812,
0.031494140625,
-0.0193328857421875,
0.03912353515625,
-0.0019245147705078125,
-0.02691650390625,
0.035614013671875,
0.0089263916015625,
-0.036285400390625,
-0.034881591796875,
0.049468994140625,
-0.0770263671875,
-0.036163330078125,
-0.03692626953125,
-0.05023193359375,
0.0091094970703125,
0.022216796875,
0.03900146484375,
0.034820556640625,
0.005916595458984375,
0.00673675537109375,
0.04266357421875,
-0.0271148681640625,
0.0272979736328125,
0.023406982421875,
-0.032135009765625,
-0.034637451171875,
0.08099365234375,
0.026641845703125,
0.02703857421875,
0.017120361328125,
0.03326416015625,
-0.024932861328125,
-0.0277862548828125,
-0.0364990234375,
0.016845703125,
-0.06146240234375,
-0.01322174072265625,
-0.03619384765625,
-0.0285797119140625,
-0.0323486328125,
0.001125335693359375,
-0.0202178955078125,
-0.0312347412109375,
-0.039306640625,
-0.0028858184814453125,
0.03271484375,
0.039154052734375,
-0.0072021484375,
0.025909423828125,
-0.03533935546875,
0.020294189453125,
0.027496337890625,
0.013824462890625,
0.01366424560546875,
-0.06353759765625,
-0.035003662109375,
0.0145111083984375,
-0.045684814453125,
-0.05133056640625,
0.032928466796875,
0.01219940185546875,
0.037933349609375,
0.02667236328125,
0.01264190673828125,
0.061920166015625,
-0.0281982421875,
0.062744140625,
0.018157958984375,
-0.0665283203125,
0.037567138671875,
-0.0181427001953125,
0.0236358642578125,
0.0228729248046875,
0.03851318359375,
-0.032745361328125,
-0.0147552490234375,
-0.0458984375,
-0.060455322265625,
0.0677490234375,
0.0152740478515625,
0.015289306640625,
0.0183258056640625,
0.03076171875,
0.00853729248046875,
0.0093841552734375,
-0.08209228515625,
-0.0313720703125,
-0.0277862548828125,
-0.01325225830078125,
0.0185394287109375,
-0.0057525634765625,
0.0003199577331542969,
-0.051910400390625,
0.0732421875,
0.00487518310546875,
0.0294952392578125,
0.020721435546875,
-0.0092315673828125,
0.0009589195251464844,
-0.01454925537109375,
0.042816162109375,
0.04766845703125,
-0.02117919921875,
-0.01213836669921875,
0.02276611328125,
-0.05389404296875,
-0.00168609619140625,
0.028656005859375,
-0.037200927734375,
-0.0040740966796875,
0.0248870849609375,
0.08270263671875,
0.0037899017333984375,
-0.0343017578125,
0.0198822021484375,
-0.0037899017333984375,
-0.010498046875,
-0.0252838134765625,
0.00864410400390625,
0.005413055419921875,
0.007282257080078125,
0.0179290771484375,
0.005535125732421875,
0.005985260009765625,
-0.033447265625,
0.0027713775634765625,
0.0239410400390625,
-0.021453857421875,
-0.029266357421875,
0.06854248046875,
0.0200042724609375,
-0.0192108154296875,
0.051605224609375,
-0.02972412109375,
-0.0270233154296875,
0.057098388671875,
0.03387451171875,
0.061614990234375,
-0.017364501953125,
0.00955963134765625,
0.05712890625,
0.0242767333984375,
0.0057525634765625,
0.01177978515625,
0.0005512237548828125,
-0.039093017578125,
-0.017578125,
-0.03643798828125,
-0.022003173828125,
0.0213775634765625,
-0.049102783203125,
0.055328369140625,
-0.03778076171875,
-0.004367828369140625,
-0.007793426513671875,
0.019561767578125,
-0.055450439453125,
0.01413726806640625,
-0.005153656005859375,
0.062042236328125,
-0.0491943359375,
0.07049560546875,
0.039276123046875,
-0.047943115234375,
-0.079833984375,
0.008697509765625,
-0.003498077392578125,
-0.055816650390625,
0.049102783203125,
0.0258331298828125,
0.00849151611328125,
0.002796173095703125,
-0.03631591796875,
-0.06292724609375,
0.101806640625,
0.0296630859375,
-0.04681396484375,
-0.0156707763671875,
0.0006546974182128906,
0.03472900390625,
-0.0026874542236328125,
0.039886474609375,
0.0288543701171875,
0.02056884765625,
0.0008864402770996094,
-0.0821533203125,
0.0065765380859375,
-0.01458740234375,
-0.0095977783203125,
-0.0027484893798828125,
-0.07354736328125,
0.08489990234375,
-0.02874755859375,
-0.0022602081298828125,
0.004650115966796875,
0.04937744140625,
0.021087646484375,
0.0284423828125,
0.01708984375,
0.0428466796875,
0.07904052734375,
-0.001857757568359375,
0.07476806640625,
-0.033447265625,
0.057647705078125,
0.083984375,
0.00302886962890625,
0.06317138671875,
0.0222930908203125,
-0.030303955078125,
0.03106689453125,
0.060211181640625,
-0.00714874267578125,
0.043365478515625,
0.005748748779296875,
-0.0085296630859375,
0.0082244873046875,
0.00467681884765625,
-0.04254150390625,
0.0197601318359375,
0.008392333984375,
-0.0178375244140625,
-0.0032939910888671875,
0.007293701171875,
0.01239013671875,
-0.02679443359375,
-0.018890380859375,
0.031890869140625,
0.01230621337890625,
-0.038909912109375,
0.065185546875,
0.003570556640625,
0.061553955078125,
-0.046661376953125,
-0.001983642578125,
-0.0203094482421875,
0.002933502197265625,
-0.032989501953125,
-0.04901123046875,
0.01519775390625,
-0.005092620849609375,
-0.007366180419921875,
-0.008209228515625,
0.03570556640625,
-0.040679931640625,
-0.04949951171875,
0.0166015625,
0.0097503662109375,
0.01256561279296875,
0.022186279296875,
-0.06536865234375,
0.0009741783142089844,
0.01447296142578125,
-0.038909912109375,
0.017486572265625,
0.0225372314453125,
0.005016326904296875,
0.0474853515625,
0.0474853515625,
-0.00782012939453125,
0.010223388671875,
-0.011474609375,
0.059478759765625,
-0.043212890625,
-0.02337646484375,
-0.05731201171875,
0.052154541015625,
-0.01076507568359375,
-0.042510986328125,
0.072021484375,
0.052825927734375,
0.08331298828125,
0.0007357597351074219,
0.053802490234375,
-0.0175323486328125,
0.0194549560546875,
-0.046783447265625,
0.06414794921875,
-0.047088623046875,
0.01479339599609375,
-0.0289764404296875,
-0.05926513671875,
-0.009429931640625,
0.0645751953125,
-0.035736083984375,
0.0120849609375,
0.046966552734375,
0.059600830078125,
-0.00925445556640625,
0.006435394287109375,
0.0174102783203125,
0.018218994140625,
0.022705078125,
0.039581298828125,
0.042999267578125,
-0.06573486328125,
0.049041748046875,
-0.053192138671875,
-0.0120391845703125,
-0.0198516845703125,
-0.0609130859375,
-0.072021484375,
-0.04229736328125,
-0.03802490234375,
-0.0258941650390625,
-0.01253509521484375,
0.0772705078125,
0.060638427734375,
-0.062164306640625,
-0.035675048828125,
-0.01690673828125,
0.01088714599609375,
-0.01727294921875,
-0.02056884765625,
0.056365966796875,
-0.02032470703125,
-0.058624267578125,
0.0002067089080810547,
-0.0092926025390625,
0.0156402587890625,
-0.0291748046875,
-0.0242767333984375,
-0.0194244384765625,
-0.01244354248046875,
0.0509033203125,
0.02294921875,
-0.05023193359375,
-0.0090789794921875,
0.016845703125,
-0.0142059326171875,
0.00860595703125,
0.0255126953125,
-0.03656005859375,
0.043304443359375,
0.033203125,
0.035797119140625,
0.053192138671875,
-0.0145416259765625,
0.018768310546875,
-0.031768798828125,
0.0288848876953125,
0.005611419677734375,
0.0232391357421875,
0.0230255126953125,
-0.0328369140625,
0.042327880859375,
0.02862548828125,
-0.05975341796875,
-0.0732421875,
-0.0119171142578125,
-0.057891845703125,
-0.003314971923828125,
0.0853271484375,
-0.02435302734375,
-0.0155487060546875,
0.0052337646484375,
-0.034393310546875,
0.026641845703125,
-0.01531982421875,
0.0667724609375,
0.05670166015625,
-0.021881103515625,
-0.01611328125,
-0.06610107421875,
0.03936767578125,
0.027191162109375,
-0.07421875,
0.0034942626953125,
0.0284423828125,
0.0273590087890625,
0.0055084228515625,
0.049346923828125,
0.0012359619140625,
0.029815673828125,
0.0019855499267578125,
0.0102081298828125,
-0.01336669921875,
-0.00604248046875,
-0.0248870849609375,
-0.0139923095703125,
-0.004024505615234375,
-0.020233154296875
]
] |
Azure99/blossom-v1-3b | 2023-08-01T12:18:45.000Z | [
"transformers",
"pytorch",
"bloom",
"text-generation",
"zh",
"en",
"dataset:Azure99/blossom-chat-v1",
"license:apache-2.0",
"endpoints_compatible",
"has_space",
"text-generation-inference",
"region:us"
] | text-generation | Azure99 | null | null | Azure99/blossom-v1-3b | 0 | 5,969 | transformers | 2023-07-29T12:06:45 | ---
license: apache-2.0
datasets:
- Azure99/blossom-chat-v1
language:
- zh
- en
---
# **BLOSSOM-v1-3b**
### 介绍
Blossom是一个对话式语言模型，基于Bloom-3b预训练模型，在60K Blossom chat数据集上进行指令精调得来。
### 推理
推理采用对话续写的形式。
单轮对话
```
A chat between a human and an artificial intelligence bot. The bot gives helpful, detailed, and polite answers to the human's questions.
|Human|: 你好
|Bot|:
```
多轮对话
```
A chat between a human and an artificial intelligence bot. The bot gives helpful, detailed, and polite answers to the human's questions.
|Human|: 你好
|Bot|: 你好，有什么我能帮助你的？</s>
|Human|: 介绍下中国的首都吧
|Bot|:
```
注意：在历史对话的Bot输出结尾，拼接一个</s> | 620 | [
[
-0.01087188720703125,
-0.06072998046875,
0.00960540771484375,
0.06903076171875,
-0.02862548828125,
-0.002193450927734375,
0.007236480712890625,
-0.02752685546875,
0.0292205810546875,
0.0302734375,
-0.052459716796875,
-0.0244293212890625,
-0.031219482421875,
-0.006389617919921875,
-0.0284423828125,
0.0576171875,
0.0185699462890625,
0.0089263916015625,
-0.00586700439453125,
0.035888671875,
-0.05029296875,
-0.01413726806640625,
-0.0628662109375,
-0.012847900390625,
0.02197265625,
0.037353515625,
0.04486083984375,
0.017730712890625,
0.04931640625,
0.0215911865234375,
-0.004970550537109375,
-0.01209259033203125,
-0.0439453125,
0.008056640625,
0.0146026611328125,
-0.06024169921875,
-0.061553955078125,
-0.016937255859375,
0.03826904296875,
0.0164947509765625,
0.0171356201171875,
0.010528564453125,
0.01070404052734375,
0.062469482421875,
-0.018768310546875,
0.040252685546875,
-0.0428466796875,
0.0042572021484375,
-0.039825439453125,
-0.01123046875,
-0.014312744140625,
-0.044586181640625,
-0.02423095703125,
-0.0389404296875,
-0.0149383544921875,
-0.007755279541015625,
0.09375,
0.01470947265625,
-0.0248260498046875,
0.0247344970703125,
-0.052581787109375,
0.045989990234375,
-0.04388427734375,
-0.00339508056640625,
0.0139617919921875,
0.04779052734375,
-0.00609588623046875,
-0.05743408203125,
-0.03857421875,
0.0098724365234375,
-0.021026611328125,
0.0311126708984375,
-0.0191192626953125,
-0.019561767578125,
0.0023059844970703125,
0.0202178955078125,
-0.01345062255859375,
-0.0103607177734375,
-0.047576904296875,
-0.027587890625,
0.036376953125,
0.02252197265625,
0.043182373046875,
-0.01198577880859375,
-0.0171661376953125,
0.0030422210693359375,
-0.0307159423828125,
0.020782470703125,
0.0004563331604003906,
0.0263671875,
-0.0347900390625,
0.0239715576171875,
-0.004985809326171875,
0.0328369140625,
-0.0089111328125,
-0.00506591796875,
0.0300445556640625,
-0.048828125,
-0.0254974365234375,
-0.044189453125,
0.07073974609375,
0.03485107421875,
0.009552001953125,
0.006626129150390625,
0.01189422607421875,
-0.0167236328125,
-0.003734588623046875,
-0.07025146484375,
-0.01849365234375,
0.036224365234375,
-0.047515869140625,
-0.0193328857421875,
0.045928955078125,
-0.0626220703125,
0.0171356201171875,
0.024627685546875,
0.017913818359375,
-0.039642333984375,
-0.06329345703125,
-0.00873565673828125,
-0.03131103515625,
0.02117919921875,
0.037811279296875,
-0.046600341796875,
0.0111083984375,
0.0540771484375,
0.053558349609375,
0.00038743019104003906,
-0.023406982421875,
0.004589080810546875,
0.006137847900390625,
-0.024139404296875,
0.02801513671875,
0.0020046234130859375,
-0.04205322265625,
-0.01317596435546875,
0.00040721893310546875,
0.004779815673828125,
-0.0227203369140625,
0.03546142578125,
-0.05303955078125,
0.017608642578125,
-0.023468017578125,
-0.01605224609375,
-0.03253173828125,
0.0195159912109375,
-0.043182373046875,
0.04058837890625,
0.0261383056640625,
-0.055511474609375,
0.02569580078125,
-0.072998046875,
-0.02978515625,
0.0377197265625,
-0.0229644775390625,
-0.00701904296875,
-0.034912109375,
-0.0139312744140625,
0.02667236328125,
-0.00020051002502441406,
-0.045654296875,
-0.002269744873046875,
-0.006633758544921875,
0.0240631103515625,
-0.03912353515625,
0.09674072265625,
0.03570556640625,
-0.017608642578125,
0.023681640625,
-0.0265350341796875,
0.00022363662719726562,
0.0399169921875,
-0.00832366943359375,
0.01160430908203125,
-0.006641387939453125,
0.0169219970703125,
0.0052947998046875,
0.06927490234375,
-0.036529541015625,
0.0240631103515625,
-0.050567626953125,
0.03228759765625,
0.04296875,
0.0178680419921875,
0.018096923828125,
-0.041412353515625,
0.034881591796875,
0.0293731689453125,
0.0286407470703125,
-0.031524658203125,
-0.060028076171875,
-0.06182861328125,
-0.00307464599609375,
-0.0021915435791015625,
0.0660400390625,
-0.032135009765625,
0.0531005859375,
-0.0159149169921875,
-0.049774169921875,
-0.030853271484375,
0.00530242919921875,
0.0237274169921875,
0.006683349609375,
0.0176849365234375,
-0.01100921630859375,
-0.026641845703125,
-0.04486083984375,
-0.0014123916625976562,
-0.040374755859375,
0.0023288726806640625,
0.04052734375,
0.020416259765625,
-0.0184478759765625,
0.06976318359375,
-0.0262908935546875,
-0.0273590087890625,
-0.033172607421875,
-0.00627899169921875,
0.040069580078125,
0.032745361328125,
0.06573486328125,
-0.058868408203125,
-0.06512451171875,
0.007305145263671875,
-0.0428466796875,
0.0167388916015625,
-0.0018749237060546875,
-0.034637451171875,
0.0400390625,
-0.0013208389282226562,
-0.03521728515625,
0.0498046875,
0.024993896484375,
-0.045379638671875,
0.042083740234375,
0.0016164779663085938,
0.040679931640625,
-0.08135986328125,
-0.007663726806640625,
-0.042266845703125,
-0.015899658203125,
-0.03509521484375,
0.004978179931640625,
-0.0018596649169921875,
0.0384521484375,
-0.059600830078125,
0.044403076171875,
-0.024932861328125,
0.01226043701171875,
-0.0299835205078125,
0.0211029052734375,
-0.00299835205078125,
0.022247314453125,
-0.0330810546875,
0.046112060546875,
0.049957275390625,
-0.051055908203125,
0.0611572265625,
0.05523681640625,
-0.0101165771484375,
0.0055694580078125,
-0.033660888671875,
-0.01003265380859375,
0.0172271728515625,
0.02117919921875,
-0.07421875,
-0.0299224853515625,
0.043121337890625,
-0.074951171875,
0.0012302398681640625,
0.0029163360595703125,
-0.02392578125,
-0.049560546875,
-0.02490234375,
0.033721923828125,
0.06414794921875,
-0.0069122314453125,
0.0057830810546875,
0.0211334228515625,
-0.00527191162109375,
-0.026153564453125,
-0.06390380859375,
0.0097198486328125,
0.0011854171752929688,
-0.072021484375,
0.0230865478515625,
-0.00495147705078125,
-0.01409149169921875,
-0.00974273681640625,
0.0308990478515625,
0.00025963783264160156,
0.012939453125,
0.00940704345703125,
0.027557373046875,
-0.018829345703125,
-0.0303955078125,
-0.0082855224609375,
-0.005771636962890625,
0.01512908935546875,
0.0274810791015625,
0.05908203125,
-0.00843048095703125,
-0.051177978515625,
-0.06390380859375,
0.0143585205078125,
0.05322265625,
0.014068603515625,
0.0501708984375,
0.051605224609375,
-0.023681640625,
0.0175323486328125,
-0.026123046875,
-0.01309967041015625,
-0.037933349609375,
0.0390625,
-0.033203125,
-0.056610107421875,
0.0238494873046875,
0.02606201171875,
0.018890380859375,
0.07452392578125,
0.047027587890625,
-0.00485992431640625,
0.07965087890625,
0.045684814453125,
-0.042144775390625,
0.03533935546875,
-0.0272216796875,
0.005878448486328125,
-0.03717041015625,
-0.039276123046875,
-0.036956787109375,
-0.0271453857421875,
-0.02886962890625,
-0.017364501953125,
0.01496124267578125,
0.01030731201171875,
-0.03436279296875,
0.037628173828125,
-0.0413818359375,
0.0058135986328125,
0.05426025390625,
0.01392364501953125,
-0.0023937225341796875,
-0.04241943359375,
0.040008544921875,
0.0118255615234375,
-0.036102294921875,
-0.0282135009765625,
0.03509521484375,
0.043609619140625,
0.0279083251953125,
0.0188140869140625,
0.034271240234375,
0.00440216064453125,
0.004299163818359375,
-0.040374755859375,
0.05621337890625,
0.0003993511199951172,
-0.043304443359375,
-0.03271484375,
-0.0203857421875,
-0.079345703125,
-0.0189361572265625,
0.0040283203125,
-0.07220458984375,
-0.033416748046875,
0.006641387939453125,
-0.0261993408203125,
0.042144775390625,
-0.031951904296875,
0.0662841796875,
-0.0467529296875,
0.00464630126953125,
-0.0215606689453125,
-0.046234130859375,
0.04193115234375,
0.0160675048828125,
0.0197296142578125,
-0.019561767578125,
-0.0004718303680419922,
0.056060791015625,
-0.018402099609375,
0.033172607421875,
-0.031768798828125,
0.01102447509765625,
0.018768310546875,
0.0201416015625,
0.032806396484375,
0.008697509765625,
0.020416259765625,
0.006122589111328125,
0.01824951171875,
-0.047698974609375,
-0.0154876708984375,
0.048614501953125,
-0.0723876953125,
-0.06390380859375,
-0.0291900634765625,
-0.0109710693359375,
0.0135650634765625,
0.022430419921875,
0.03411865234375,
-0.01081085205078125,
-0.024383544921875,
0.02886962890625,
-0.00897216796875,
-0.0235748291015625,
0.042816162109375,
0.02227783203125,
-0.061920166015625,
-0.021484375,
0.060791015625,
-0.0067596435546875,
0.010223388671875,
0.0284576416015625,
-0.001697540283203125,
-0.00933074951171875,
-0.032379150390625,
-0.0243072509765625,
0.01416015625,
-0.01508331298828125,
-0.021942138671875,
-0.04888916015625,
-0.05621337890625,
-0.06011962890625,
0.0032215118408203125,
-0.01364898681640625,
-0.01155853271484375,
-0.050506591796875,
-0.00016689300537109375,
0.057647705078125,
0.048492431640625,
0.0016164779663085938,
0.0350341796875,
-0.0904541015625,
0.022369384765625,
0.01544952392578125,
0.0211944580078125,
0.00125885009765625,
-0.043426513671875,
-0.041168212890625,
0.0247955322265625,
-0.04144287109375,
-0.0762939453125,
0.041656494140625,
0.005886077880859375,
0.042938232421875,
0.07745361328125,
-0.01056671142578125,
0.044891357421875,
-0.018341064453125,
0.10498046875,
0.029449462890625,
-0.057373046875,
0.053863525390625,
-0.034210205078125,
0.005664825439453125,
0.025726318359375,
0.0035877227783203125,
-0.049224853515625,
-0.030548095703125,
-0.06414794921875,
-0.07305908203125,
0.07281494140625,
0.04095458984375,
0.04095458984375,
-0.0007686614990234375,
0.003818511962890625,
0.0020694732666015625,
0.01255035400390625,
-0.052276611328125,
-0.05780029296875,
-0.024017333984375,
0.0027332305908203125,
0.018768310546875,
-0.0189666748046875,
-0.007778167724609375,
-0.0229949951171875,
0.0687255859375,
0.036163330078125,
0.047943115234375,
-0.0128326416015625,
0.02850341796875,
-0.0220794677734375,
0.003246307373046875,
0.033599853515625,
0.036651611328125,
-0.01971435546875,
-0.0162353515625,
0.013916015625,
-0.032073974609375,
-0.0157623291015625,
-0.017608642578125,
-0.0204620361328125,
0.0011959075927734375,
0.027252197265625,
0.05682373046875,
-0.0182037353515625,
-0.029022216796875,
0.060028076171875,
-0.00592041015625,
-0.004512786865234375,
-0.04083251953125,
0.042510986328125,
0.0024738311767578125,
0.020477294921875,
0.035552978515625,
0.01364898681640625,
-0.007049560546875,
-0.03875732421875,
0.0296478271484375,
0.05084228515625,
-0.040130615234375,
-0.0200653076171875,
0.057373046875,
0.037994384765625,
-0.06439208984375,
0.041015625,
-0.015380859375,
-0.0572509765625,
0.057891845703125,
0.042999267578125,
0.068603515625,
0.00891876220703125,
0.0236358642578125,
0.04681396484375,
0.006908416748046875,
0.00830841064453125,
0.0313720703125,
0.006862640380859375,
-0.054595947265625,
-0.0135650634765625,
-0.02880859375,
-0.038360595703125,
0.01873779296875,
-0.01502227783203125,
0.031036376953125,
-0.028076171875,
-0.014434814453125,
-0.017791748046875,
0.001712799072265625,
-0.023468017578125,
0.0175628662109375,
0.005126953125,
0.09674072265625,
-0.037933349609375,
0.05322265625,
0.02154541015625,
-0.047119140625,
-0.048309326171875,
0.004711151123046875,
-0.00547027587890625,
-0.07244873046875,
0.050384521484375,
0.0114288330078125,
0.0023651123046875,
0.005645751953125,
-0.056488037109375,
-0.060943603515625,
0.0992431640625,
-0.007843017578125,
-0.0162506103515625,
0.02886962890625,
0.0008234977722167969,
0.0197296142578125,
-0.02520751953125,
0.04779052734375,
0.034210205078125,
0.050262451171875,
0.0264892578125,
-0.069580078125,
-0.008270263671875,
-0.0443115234375,
-0.01279449462890625,
0.00585174560546875,
-0.08135986328125,
0.06689453125,
0.002559661865234375,
-0.040191650390625,
0.02392578125,
0.07098388671875,
0.00818634033203125,
0.0137939453125,
0.052581787109375,
0.00745391845703125,
0.012542724609375,
-0.04571533203125,
0.045196533203125,
-0.043243408203125,
0.00921630859375,
0.04052734375,
0.006397247314453125,
0.044403076171875,
0.006237030029296875,
-0.06890869140625,
0.05267333984375,
0.0667724609375,
-0.01366424560546875,
0.0179595947265625,
-0.0211029052734375,
-0.01499176025390625,
0.00634765625,
0.01229095458984375,
-0.031890869140625,
0.0038661956787109375,
0.01399993896484375,
0.0009369850158691406,
0.00514984130859375,
-0.00759124755859375,
0.0222930908203125,
-0.0082244873046875,
-0.0265655517578125,
0.080810546875,
-0.026824951171875,
-0.061309814453125,
0.013519287109375,
0.0049285888671875,
0.06854248046875,
-0.05328369140625,
0.0011949539184570312,
-0.0200653076171875,
0.00695037841796875,
-0.0253448486328125,
-0.076904296875,
0.018524169921875,
-0.00290679931640625,
-0.0033321380615234375,
-0.0030975341796875,
0.0447998046875,
-0.006076812744140625,
-0.038116455078125,
0.033203125,
0.035064697265625,
0.038360595703125,
-0.00621795654296875,
-0.07550048828125,
-0.014312744140625,
0.02862548828125,
-0.0211029052734375,
0.0016183853149414062,
0.0252532958984375,
0.021209716796875,
0.0517578125,
0.0341796875,
0.0343017578125,
0.004711151123046875,
0.01349639892578125,
0.05572509765625,
-0.0592041015625,
-0.0374755859375,
-0.07806396484375,
0.051605224609375,
-0.0211029052734375,
-0.01389312744140625,
0.0623779296875,
0.045501708984375,
0.04803466796875,
-0.005008697509765625,
0.074462890625,
-0.0408935546875,
0.06158447265625,
0.0016260147094726562,
0.05584716796875,
-0.04296875,
-0.000949859619140625,
-0.0272064208984375,
-0.0298614501953125,
-0.01087188720703125,
0.0302276611328125,
0.005039215087890625,
0.0279693603515625,
0.039215087890625,
0.043212890625,
0.020965576171875,
-0.0062408447265625,
0.0292205810546875,
0.038482666015625,
0.0284423828125,
0.045135498046875,
0.042144775390625,
-0.038177490234375,
0.058074951171875,
-0.0290069580078125,
-0.03753662109375,
-0.039703369140625,
-0.031585693359375,
-0.05377197265625,
-0.035125732421875,
-0.0033588409423828125,
-0.04339599609375,
-0.0390625,
0.057952880859375,
0.029205322265625,
-0.07086181640625,
-0.0264434814453125,
-0.01497650146484375,
0.019439697265625,
-0.0283355712890625,
-0.0226287841796875,
0.0204315185546875,
-0.039825439453125,
-0.06927490234375,
0.0019311904907226562,
0.03399658203125,
0.0207977294921875,
-0.04205322265625,
-0.0081329345703125,
0.0003108978271484375,
0.0162353515625,
0.0136260986328125,
0.0237579345703125,
-0.0638427734375,
-0.0129241943359375,
0.0234375,
-0.014434814453125,
0.0242462158203125,
0.0218048095703125,
-0.03289794921875,
0.004108428955078125,
0.059844970703125,
-0.004192352294921875,
0.03314208984375,
0.0010623931884765625,
0.038818359375,
-0.0338134765625,
0.017120361328125,
0.01290130615234375,
0.02667236328125,
-0.00424957275390625,
-0.04339599609375,
0.021209716796875,
0.02606201171875,
-0.053985595703125,
-0.027069091796875,
0.031646728515625,
-0.09674072265625,
-0.032989501953125,
0.08013916015625,
-0.00662994384765625,
-0.030364990234375,
-0.0045623779296875,
-0.051055908203125,
0.044830322265625,
-0.033447265625,
0.0504150390625,
0.04296875,
-0.027130126953125,
-0.0097503662109375,
-0.0164337158203125,
0.049835205078125,
0.0087127685546875,
-0.06695556640625,
-0.0223846435546875,
0.013824462890625,
-0.00372314453125,
0.0207672119140625,
0.06488037109375,
-0.005535125732421875,
0.0350341796875,
-0.00897979736328125,
-0.0045318603515625,
0.0019474029541015625,
-0.00861358642578125,
0.0198822021484375,
0.0185699462890625,
0.0004532337188720703,
-0.044921875
]
] |
64bits/LexPodLM-13B | 2023-06-14T00:15:01.000Z | [
"transformers",
"pytorch",
"llama",
"text-generation",
"en",
"dataset:64bits/lex_fridman_podcast_for_llm_vicuna",
"license:other",
"endpoints_compatible",
"has_space",
"text-generation-inference",
"region:us"
] | text-generation | 64bits | null | null | 64bits/LexPodLM-13B | 10 | 5,968 | transformers | 2023-06-10T06:17:34 | ---
license: other
datasets:
- 64bits/lex_fridman_podcast_for_llm_vicuna
language:
- en
library_name: transformers
tags:
- llama
- transformers
---
LexPodLM-13B is a LLaMA-13B model fine-tuned using Vicuna's schema, with this dataset: https://huggingface.co/datasets/64bits/lex_fridman_podcast_for_llm_vicuna
While it might not excel in tasks like code writing, fact searching, or essay writing, it can potentially serve as an effective chat model, capable of providing verbose and nuanced responses.
I trained the model on 4 A100 (80 GB RAM) for ~15 hrs.
***
Eval (evaluate models like [Open LLM Leaderboard](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard) using [Eleuther AI Language Model Evaluation Harness](https://github.com/EleutherAI/lm-evaluation-harness])):
| Model | Average | ARC (25-shot) | HellaSwag (10-shot) | MMLU (5-shot) | TruthfulQA (0-shot) |
| ------------ | ------- | ------------- | ------------------- | ------------- | ------------------- |
| LexPodLM-13B | 52.5 | 48.8 | 79.4 | 37.9 | 43.8 |
| LLaMA-13B | 51.8 | 50.8 | 78.9 | 37.7 | 39.9 |
***
*Human: do you think consciousness is foundamentally computational?*
*Assistant: I think consciousness is foundational computational. Yes. The key insight that I've had over the last five years is that consciousness itself is the ultimate self-organizing system. It's the ultimate emergent of the universe. It's the ultimate thing that's not explainable in terms of its parts. So it's not that I think brains are computation. I think, you know, brains are made of neurons, and neurons are made of cells. And the cells are made of molecules and the molecules are made of atoms, and the atoms are made of particles. And we can explain all of those things in terms of their parts. There's no emergent physics in physics. There are no emergent laws in physics. And you know, people have been trying to get to this level. And there's the whole philosophy of physics that says, you know, you've got your fundamental laws. Let's just explain everything in terms of the fundamental laws. But that's just not how physics works. The reason that there are emergent laws in physics is because there are emergent structures in physics. And the first emergent structure is the cell. The cell is a structure that doesn't exist in physics. It's an emergent structure. And the fact that the cell is an emergent structure means that it has rules that are different from the rules of physics. And the same thing with brains. Brains have rules that are different from the rules of physics. And the same thing with consciousness. Consciousness, the fact that it exists at all is an emergent fact. It's not a fact that you can reduce to the underlying physics. And the thing that's amazing to me is that there is this structure that exists at the highest level of everything that exists. And it's not an emergent. So the fact that you and I have consciousness is not an emergent fact. It's a foundational fact. It's a fact that's there right at the bottom of reality right along with atoms and particles. I mean, it's not that I think that brains are conscious. I think that brains are a particular way of organizing consciousness. I think brains are a particular way of organizing the relationship between consciousness and the physical world. And so when people say to me, you know, well, what do you think?*
***
*Human: What are the ethical implications of artificial intelligence?*
*Assistant: The ethical implications of artificial intelligence are significant. As AI systems become more advanced and integrated into our daily lives, we have to be aware of their limitations and potential unfair biases. Such systems should be designed with a focus on fairness, transparency, and security. AI systems should not be used to replace human decision-making. Their output should be easily interpretable, so that decisions can be traced to rationales and justifications. We should also prioritize the well-being of humans over the well-being of AI systems. We need to ensure that AI systems are designed in a way that they do not harm users or other people. This can be achieved by building in safety mechanisms and testing systems for potential biases and security vulnerabilities.*
***
*Human: How do you think quantum computing will affect our future?*
*Assistant: Quantum computing has the potential to change the way we compute in many ways. One of the most significant benefits will be the possibility of solving previously unsolvable problems and accelerating the development of machine learning and artificial intelligence. By harnessing the power of quantum computers, we can make machines that are exponentially more powerful than any existing computer.*
***
*Human: What are your thoughts on the Singularity theory?*
*Assistant: I am a strong proponent of the singularity. I have seen many technological advancements in my career. I've seen the power of exponential growth in technology. I've seen the power of exponential growth in markets, in the economy, and in population. And I'm very excited about the future. I'm very excited about what's going to happen. And I think that it's important for the world to understand that. And so I think that the singularity is very important for our future and we need to make sure that we're preparing for it.* | 5,459 | [
[
-0.027435302734375,
-0.059783935546875,
0.056793212890625,
0.013336181640625,
-0.0325927734375,
0.020111083984375,
-0.005947113037109375,
-0.053314208984375,
0.0229339599609375,
0.0090789794921875,
-0.023956298828125,
-0.01459503173828125,
-0.0279693603515625,
-0.00081634521484375,
-0.0213775634765625,
0.058502197265625,
0.0474853515625,
0.01039886474609375,
-0.00675201416015625,
0.00890350341796875,
-0.042022705078125,
-0.03814697265625,
-0.0672607421875,
-0.02056884765625,
0.029876708984375,
0.0006017684936523438,
0.04632568359375,
0.0367431640625,
0.03271484375,
0.0316162109375,
-0.02447509765625,
-0.0007715225219726562,
-0.023162841796875,
0.004665374755859375,
-0.01477813720703125,
-0.045654296875,
-0.032440185546875,
-0.0093231201171875,
0.053375244140625,
0.0623779296875,
-0.0187835693359375,
0.016937255859375,
-0.002513885498046875,
0.05523681640625,
-0.031463623046875,
-0.017578125,
-0.032440185546875,
0.017974853515625,
-0.002895355224609375,
-0.01849365234375,
-0.0144195556640625,
-0.02557373046875,
-0.031524658203125,
-0.043182373046875,
0.00485992431640625,
0.04229736328125,
0.0567626953125,
0.03448486328125,
-0.030609130859375,
-0.0003571510314941406,
-0.058502197265625,
0.06622314453125,
-0.048187255859375,
0.0440673828125,
0.016815185546875,
0.0101318359375,
-0.018341064453125,
-0.049774169921875,
-0.06475830078125,
-0.0277557373046875,
0.004566192626953125,
0.03680419921875,
-0.0242156982421875,
0.006488800048828125,
0.024261474609375,
0.02557373046875,
-0.03009033203125,
0.006351470947265625,
-0.046478271484375,
-0.01251983642578125,
0.04681396484375,
0.0197906494140625,
0.04779052734375,
-0.0188140869140625,
-0.0282440185546875,
0.01904296875,
-0.0293731689453125,
0.0219573974609375,
0.0235748291015625,
0.0093231201171875,
-0.00588226318359375,
0.033721923828125,
0.00620269775390625,
0.0418701171875,
0.01342010498046875,
-0.00609588623046875,
-0.005664825439453125,
-0.026397705078125,
-0.01470184326171875,
0.00986480712890625,
0.060089111328125,
0.04193115234375,
0.0022430419921875,
-0.0261688232421875,
0.016265869140625,
0.00720977783203125,
0.048583984375,
-0.05316162109375,
-0.00937652587890625,
0.0467529296875,
-0.047332763671875,
-0.01094818115234375,
0.005573272705078125,
-0.031158447265625,
-0.031036376953125,
-0.0245819091796875,
0.01032257080078125,
-0.046783447265625,
-0.0279083251953125,
0.0133056640625,
-0.0101776123046875,
0.01480865478515625,
0.033660888671875,
-0.06658935546875,
0.03387451171875,
0.058868408203125,
0.057403564453125,
0.0034160614013671875,
-0.022979736328125,
0.00925445556640625,
0.005619049072265625,
-0.054931640625,
0.06768798828125,
-0.019317626953125,
-0.032684326171875,
-0.050323486328125,
-0.006099700927734375,
-0.00623321533203125,
-0.03131103515625,
0.0177001953125,
-0.0287628173828125,
0.0024242401123046875,
-0.032867431640625,
-0.054595947265625,
-0.0291595458984375,
-0.0062255859375,
-0.018890380859375,
0.04510498046875,
-0.02374267578125,
-0.045440673828125,
0.0246124267578125,
-0.077880859375,
-0.0308380126953125,
0.042633056640625,
-0.0260772705078125,
0.0004973411560058594,
0.00441741943359375,
-0.0262451171875,
0.017364501953125,
-0.031951904296875,
0.0343017578125,
-0.01134490966796875,
-0.02069091796875,
0.031280517578125,
-0.018218994140625,
0.09722900390625,
0.03399658203125,
-0.03887939453125,
-0.009765625,
-0.0679931640625,
0.0101318359375,
0.0167236328125,
-0.0174407958984375,
-0.018310546875,
0.0130767822265625,
-0.0189208984375,
0.01499176025390625,
0.0340576171875,
-0.059783935546875,
0.00537872314453125,
-0.0254058837890625,
0.0513916015625,
0.07421875,
0.0233001708984375,
0.042510986328125,
-0.0450439453125,
0.0272979736328125,
0.00543212890625,
0.029876708984375,
0.01154327392578125,
-0.047088623046875,
-0.04937744140625,
-0.01268768310546875,
0.0007481575012207031,
0.04620361328125,
0.007598876953125,
0.0679931640625,
0.0115203857421875,
-0.053314208984375,
-0.04388427734375,
-0.005352020263671875,
0.024658203125,
0.039459228515625,
0.0246429443359375,
-0.0199432373046875,
-0.048797607421875,
-0.07220458984375,
-0.0087890625,
-0.03668212890625,
-0.00226593017578125,
0.01837158203125,
0.037994384765625,
-0.005794525146484375,
0.073486328125,
-0.0269622802734375,
0.007022857666015625,
0.01354217529296875,
-0.01495361328125,
0.0088043212890625,
0.06298828125,
0.0230865478515625,
-0.06585693359375,
0.00949859619140625,
0.01036834716796875,
-0.09185791015625,
0.0240325927734375,
-0.0165252685546875,
-0.01849365234375,
-0.00223541259765625,
0.0280914306640625,
-0.05810546875,
0.0184326171875,
-0.0061187744140625,
-0.037353515625,
0.0364990234375,
-0.0211181640625,
0.00328826904296875,
-0.08636474609375,
0.017974853515625,
-0.01425933837890625,
0.007781982421875,
-0.046630859375,
0.00908660888671875,
-0.0013265609741210938,
0.0137939453125,
-0.043487548828125,
0.0389404296875,
-0.0202484130859375,
-0.0163116455078125,
-0.017791748046875,
-0.002742767333984375,
0.0025386810302734375,
0.037353515625,
-0.01197052001953125,
0.041717529296875,
0.037322998046875,
-0.07305908203125,
0.027740478515625,
0.01505279541015625,
-0.02825927734375,
0.041900634765625,
-0.053955078125,
0.004894256591796875,
-0.0231475830078125,
0.0302734375,
-0.052398681640625,
0.002197265625,
0.025146484375,
-0.053375244140625,
-0.00792694091796875,
0.0298919677734375,
-0.03533935546875,
-0.0270538330078125,
-0.0056915283203125,
0.0240478515625,
0.062164306640625,
-0.0084991455078125,
0.045867919921875,
0.0380859375,
-0.0009784698486328125,
-0.04620361328125,
-0.048187255859375,
0.0263671875,
-0.0120391845703125,
-0.04290771484375,
0.0290069580078125,
-0.03656005859375,
-0.06463623046875,
0.0037479400634765625,
-0.007526397705078125,
0.00699615478515625,
0.037994384765625,
0.06072998046875,
0.0187225341796875,
-0.00909423828125,
0.01480865478515625,
0.0032901763916015625,
-0.01202392578125,
0.0135650634765625,
-0.03118896484375,
0.035797119140625,
-0.0419921875,
-0.00533294677734375,
-0.03887939453125,
0.0220489501953125,
0.049591064453125,
0.015899658203125,
0.042755126953125,
0.032318115234375,
-0.04107666015625,
0.01361083984375,
-0.038421630859375,
-0.01259613037109375,
-0.035675048828125,
0.00824737548828125,
-0.0204620361328125,
-0.059539794921875,
0.042724609375,
0.0124664306640625,
0.01467132568359375,
0.05059814453125,
0.054046630859375,
-0.0350341796875,
0.0712890625,
0.0496826171875,
0.018707275390625,
0.034576416015625,
-0.029022216796875,
0.01947021484375,
-0.0814208984375,
-0.037109375,
-0.01885986328125,
-0.021728515625,
-0.039794921875,
-0.0002391338348388672,
0.0189666748046875,
0.0210723876953125,
-0.047454833984375,
0.02154541015625,
-0.0576171875,
0.034881591796875,
0.040374755859375,
0.0033817291259765625,
0.0210723876953125,
-0.033294677734375,
-0.004730224609375,
0.016387939453125,
-0.067138671875,
-0.059234619140625,
0.06427001953125,
0.02288818359375,
0.043792724609375,
0.0169219970703125,
0.04376220703125,
0.019805908203125,
-0.0014657974243164062,
-0.045623779296875,
0.0606689453125,
0.0196380615234375,
-0.0760498046875,
-0.02655029296875,
-0.018280029296875,
-0.0902099609375,
0.02484130859375,
-0.0236053466796875,
-0.07080078125,
0.02044677734375,
-0.007602691650390625,
-0.0738525390625,
0.0372314453125,
-0.053863525390625,
0.0440673828125,
-0.03741455078125,
-0.0328369140625,
-0.021820068359375,
-0.07745361328125,
0.038726806640625,
-0.0013685226440429688,
0.032806396484375,
-0.01885986328125,
-0.01552581787109375,
0.07904052734375,
-0.044281005859375,
0.07745361328125,
0.0024776458740234375,
-0.005970001220703125,
0.038360595703125,
0.0026988983154296875,
-0.012786865234375,
0.0230865478515625,
0.019622802734375,
0.0136566162109375,
0.01165771484375,
-0.032440185546875,
-0.03936767578125,
0.034515380859375,
-0.0911865234375,
-0.050933837890625,
-0.039794921875,
-0.0440673828125,
0.01849365234375,
0.033477783203125,
0.0297698974609375,
0.018890380859375,
-0.015777587890625,
0.00811004638671875,
0.035308837890625,
-0.0109100341796875,
0.03631591796875,
0.0364990234375,
-0.0251007080078125,
-0.0084991455078125,
0.0689697265625,
0.0160980224609375,
0.0010576248168945312,
0.0262451171875,
0.02215576171875,
-0.01383209228515625,
-0.03070068359375,
-0.008819580078125,
0.0171051025390625,
-0.05682373046875,
0.0006184577941894531,
-0.07537841796875,
-0.036224365234375,
-0.032501220703125,
-0.0233612060546875,
-0.0263671875,
-0.0286712646484375,
-0.0252685546875,
-0.0104827880859375,
0.026824951171875,
0.06646728515625,
-0.03125,
0.0005512237548828125,
-0.0474853515625,
0.0179443359375,
0.04052734375,
0.0020275115966796875,
0.0157470703125,
-0.03302001953125,
-0.007244110107421875,
0.004222869873046875,
-0.0237884521484375,
-0.031951904296875,
0.003398895263671875,
-0.001373291015625,
0.039520263671875,
0.04595947265625,
0.005863189697265625,
0.061676025390625,
-0.053375244140625,
0.0709228515625,
0.0235748291015625,
-0.06353759765625,
0.03704833984375,
-0.00949859619140625,
0.01195526123046875,
0.03515625,
0.03466796875,
-0.028594970703125,
-0.038421630859375,
-0.091796875,
-0.05108642578125,
0.01065826416015625,
0.0276641845703125,
0.01029205322265625,
-0.00957489013671875,
0.041351318359375,
0.0034942626953125,
0.0199432373046875,
-0.062408447265625,
-0.022552490234375,
-0.01355743408203125,
0.01519775390625,
0.007579803466796875,
0.008026123046875,
-0.0138092041015625,
-0.034149169921875,
0.032562255859375,
0.01419830322265625,
0.039794921875,
0.01422882080078125,
0.01258087158203125,
-0.039154052734375,
0.012054443359375,
0.05108642578125,
0.071533203125,
-0.018096923828125,
0.0107269287109375,
0.014373779296875,
-0.04010009765625,
0.025665283203125,
-0.01375579833984375,
-0.0156707763671875,
-0.0283966064453125,
0.03173828125,
0.05010986328125,
0.005947113037109375,
-0.0548095703125,
0.01531219482421875,
-0.01029205322265625,
-0.00801849365234375,
-0.058624267578125,
0.0169219970703125,
0.00948333740234375,
0.03631591796875,
0.036346435546875,
0.03961181640625,
0.0161895751953125,
-0.0623779296875,
0.007350921630859375,
0.033660888671875,
-0.0080413818359375,
-0.0380859375,
0.0712890625,
-0.0055084228515625,
-0.049468994140625,
0.025787353515625,
-0.0163116455078125,
-0.030242919921875,
0.0653076171875,
0.045440673828125,
0.04791259765625,
-0.01629638671875,
0.002796173095703125,
0.03631591796875,
0.040069580078125,
0.00675201416015625,
0.0202178955078125,
-0.0198822021484375,
-0.06817626953125,
-0.0194244384765625,
-0.06280517578125,
-0.038299560546875,
-0.00012046098709106445,
-0.04833984375,
-0.0009832382202148438,
-0.061553955078125,
-0.027801513671875,
0.019287109375,
0.017364501953125,
-0.033355712890625,
0.0216217041015625,
0.01165008544921875,
0.062225341796875,
-0.04302978515625,
0.034881591796875,
0.0267333984375,
-0.04595947265625,
-0.04144287109375,
-0.01849365234375,
-0.00830841064453125,
-0.0755615234375,
0.043487548828125,
0.01413726806640625,
-0.0026760101318359375,
0.01519012451171875,
-0.068115234375,
-0.0875244140625,
0.0924072265625,
0.0215911865234375,
-0.034942626953125,
-0.00848388671875,
-0.032379150390625,
0.0460205078125,
-0.0221710205078125,
0.0135955810546875,
0.028106689453125,
0.041229248046875,
0.002338409423828125,
-0.034942626953125,
0.006183624267578125,
-0.03564453125,
0.0089263916015625,
0.01390838623046875,
-0.080322265625,
0.0755615234375,
-0.04193115234375,
-0.0300140380859375,
0.00946044921875,
0.059173583984375,
0.01531219482421875,
0.01450347900390625,
0.0273895263671875,
0.041290283203125,
0.06658935546875,
0.0298004150390625,
0.083251953125,
-0.04522705078125,
0.0141143798828125,
0.05535888671875,
-0.0255279541015625,
0.044097900390625,
0.03558349609375,
-0.039276123046875,
0.03662109375,
0.06622314453125,
0.0030117034912109375,
0.0158843994140625,
0.00627899169921875,
-0.01232147216796875,
-0.0304718017578125,
-0.00998687744140625,
-0.0295257568359375,
0.02783203125,
0.029022216796875,
-0.0069122314453125,
0.0066986083984375,
0.01129150390625,
-0.0017671585083007812,
-0.00759124755859375,
-0.0171356201171875,
0.062255859375,
0.026092529296875,
-0.0389404296875,
0.032196044921875,
-0.0120086669921875,
0.043304443359375,
-0.032379150390625,
0.0086822509765625,
0.00550079345703125,
-0.00855255126953125,
-0.0167083740234375,
-0.03900146484375,
0.01479339599609375,
-0.01076507568359375,
0.004459381103515625,
-0.0099639892578125,
0.052886962890625,
0.002857208251953125,
-0.036590576171875,
0.0119171142578125,
0.045867919921875,
0.0390625,
0.023834228515625,
-0.063720703125,
0.0007023811340332031,
-0.0197906494140625,
-0.01268768310546875,
0.0259552001953125,
0.017425537109375,
0.0100555419921875,
0.06536865234375,
0.06207275390625,
0.0094146728515625,
0.00518035888671875,
-0.0297698974609375,
0.05517578125,
-0.07647705078125,
-0.05059814453125,
-0.07586669921875,
0.0277252197265625,
0.0003712177276611328,
-0.03143310546875,
0.052398681640625,
0.043914794921875,
0.039947509765625,
0.0253143310546875,
0.0618896484375,
-0.0215301513671875,
0.01422882080078125,
-0.0284881591796875,
0.06854248046875,
-0.04315185546875,
-0.0021648406982421875,
-0.00896453857421875,
-0.0745849609375,
-0.033203125,
0.058074951171875,
-0.0220489501953125,
0.01009368896484375,
0.0601806640625,
0.0263824462890625,
0.020721435546875,
0.0101776123046875,
0.043487548828125,
0.028472900390625,
0.01352691650390625,
0.0343017578125,
0.0728759765625,
-0.025909423828125,
0.0220489501953125,
-0.016448974609375,
-0.0243072509765625,
-0.0206451416015625,
-0.043792724609375,
-0.0457763671875,
-0.06451416015625,
-0.026092529296875,
-0.02337646484375,
0.0031642913818359375,
0.0762939453125,
0.04693603515625,
-0.037109375,
-0.031402587890625,
-0.030975341796875,
0.0101776123046875,
-0.03302001953125,
-0.01715087890625,
0.01271820068359375,
-0.025299072265625,
-0.031463623046875,
0.0272216796875,
0.043182373046875,
0.031707763671875,
-0.01261138916015625,
0.004894256591796875,
-0.0220489501953125,
0.0169677734375,
0.06298828125,
0.040863037109375,
-0.0601806640625,
-0.03564453125,
0.01222991943359375,
-0.01221466064453125,
0.01312255859375,
0.0413818359375,
-0.05010986328125,
0.0229339599609375,
-0.0004425048828125,
0.03631591796875,
0.01071929931640625,
-0.0052032470703125,
0.0214691162109375,
-0.001697540283203125,
0.01168060302734375,
0.0435791015625,
0.03472900390625,
0.0007829666137695312,
-0.041595458984375,
0.039764404296875,
0.0231475830078125,
-0.053955078125,
-0.06988525390625,
0.0216827392578125,
-0.086669921875,
-0.046600341796875,
0.06683349609375,
0.02655029296875,
-0.055755615234375,
-0.01983642578125,
-0.01776123046875,
0.032440185546875,
-0.027740478515625,
0.0687255859375,
0.03472900390625,
-0.04400634765625,
-0.00960540771484375,
-0.042755126953125,
0.0262451171875,
0.03814697265625,
-0.045867919921875,
-0.01068878173828125,
0.0242462158203125,
0.0174407958984375,
0.0443115234375,
0.0697021484375,
-0.00004082918167114258,
0.0256195068359375,
0.0195159912109375,
0.0178375244140625,
-0.0018625259399414062,
-0.01470184326171875,
-0.015411376953125,
0.0105133056640625,
-0.01033782958984375,
-0.0198211669921875
]
] |
CobraMamba/mamba-gpt-3b-v2 | 2023-08-02T10:06:03.000Z | [
"transformers",
"pytorch",
"safetensors",
"llama",
"text-generation",
"gpt",
"llm",
"large language model",
"en",
"license:apache-2.0",
"has_space",
"text-generation-inference",
"region:us"
] | text-generation | CobraMamba | null | null | CobraMamba/mamba-gpt-3b-v2 | 9 | 5,966 | transformers | 2023-07-27T06:06:44 | ---
language:
- en
library_name: transformers
tags:
- gpt
- llm
- large language model
inference: false
thumbnail: >-
https://h2o.ai/etc.clientlibs/h2o/clientlibs/clientlib-site/resources/images/favicon.ico
license: apache-2.0
---
# Model Card
**The Best 3B Model! Surpassing dolly-v2-12b**
The best 3B model on the [Open LLM Leaderboard](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard), with performance surpassing dolly-v2-12b
| Metric | Value |
|-----------------------|-------|
| MMLU (5-shot) | 27.1 |
| ARC (25-shot) | 42.2 |
| HellaSwag (10-shot) | 71.5 |
| TruthfulQA (0-shot) | 36.7 |
| Avg. | 44.4 |
We use state-of-the-art [Language Model Evaluation Harness](https://github.com/EleutherAI/lm-evaluation-harness) to run the benchmark tests above.
## Summary
We have fine-tuned the open-lama model and surpassed the original model in multiple evaluation subtasks, making it currently the best performing 3B model with comparable performance to llama-7b
- Base model: [openlm-research/open_llama_3b_v2](https://huggingface.co/openlm-research/open_llama_3b_v2)
## Usage
To use the model with the `transformers` library on a machine with GPUs, first make sure you have the `transformers`, `accelerate` and `torch` libraries installed.
```bash
pip install transformers==4.29.2
pip install accelerate==0.19.0
pip install torch==2.0.0
```
```python
import torch
from transformers import pipeline
generate_text = pipeline(
model="CobraMamba/mamba-gpt-3b-v2",
torch_dtype="auto",
trust_remote_code=True,
use_fast=False,
device_map={"": "cuda:0"},
)
res = generate_text(
"Why is drinking water so healthy?",
min_new_tokens=2,
max_new_tokens=1024,
do_sample=False,
num_beams=1,
temperature=float(0.3),
repetition_penalty=float(1.2),
renormalize_logits=True
)
print(res[0]["generated_text"])
```
You can print a sample prompt after the preprocessing step to see how it is feed to the tokenizer:
```python
print(generate_text.preprocess("Why is drinking water so healthy?")["prompt_text"])
```
```bash
<|prompt|>Why is drinking water so healthy?</s><|answer|>
```
Alternatively, you can download the mamba_gpt_pipeline.py, store it alongside your notebook, and construct the pipeline yourself from the loaded model and tokenizer. If the model and the tokenizer are fully supported in the `transformers` package, this will allow you to set `trust_remote_code=False`.
```python
import torch
from mamba_gpt_pipeline import MambaGPTTextGenerationPipeline
from transformers import AutoModelForCausalLM, AutoTokenizer
tokenizer = AutoTokenizer.from_pretrained(
"CobraMamba/mamba-gpt-3b-v2",
use_fast=False,
padding_side="left",
trust_remote_code=False,
)
model = AutoModelForCausalLM.from_pretrained(
"CobraMamba/mamba-gpt-3b-v2",
torch_dtype="auto",
device_map={"": "cuda:0"},
trust_remote_code=False,
)
generate_text = MambaGPTTextGenerationPipeline(model=model, tokenizer=tokenizer)
res = generate_text(
"Why is drinking water so healthy?",
min_new_tokens=2,
max_new_tokens=1024,
do_sample=False,
num_beams=1,
temperature=float(0.3),
repetition_penalty=float(1.2),
renormalize_logits=True
)
print(res[0]["generated_text"])
```
You may also construct the pipeline from the loaded model and tokenizer yourself and consider the preprocessing steps:
```python
from transformers import AutoModelForCausalLM, AutoTokenizer
model_name = "CobraMamba/mamba-gpt-3b-v2" # either local folder or huggingface model name
# Important: The prompt needs to be in the same format the model was trained with.
# You can find an example prompt in the experiment logs.
prompt = "<|prompt|>How are you?</s><|answer|>"
tokenizer = AutoTokenizer.from_pretrained(
model_name,
use_fast=False,
trust_remote_code=False,
)
model = AutoModelForCausalLM.from_pretrained(
model_name,
torch_dtype="auto",
device_map={"": "cuda:0"},
trust_remote_code=False,
)
model.cuda().eval()
inputs = tokenizer(prompt, return_tensors="pt", add_special_tokens=False).to("cuda")
# generate configuration can be modified to your needs
tokens = model.generate(
**inputs,
min_new_tokens=2,
max_new_tokens=1024,
do_sample=False,
num_beams=1,
temperature=float(0.3),
repetition_penalty=float(1.2),
renormalize_logits=True
)[0]
tokens = tokens[inputs["input_ids"].shape[1]:]
answer = tokenizer.decode(tokens, skip_special_tokens=True)
print(answer)
```
## Model Architecture
```
LlamaForCausalLM(
(model): LlamaModel(
(embed_tokens): Embedding(32000, 4096, padding_idx=0)
(layers): ModuleList(
(0-31): 32 x LlamaDecoderLayer(
(self_attn): LlamaAttention(
(q_proj): Linear(in_features=4096, out_features=4096, bias=False)
(k_proj): Linear(in_features=4096, out_features=4096, bias=False)
(v_proj): Linear(in_features=4096, out_features=4096, bias=False)
(o_proj): Linear(in_features=4096, out_features=4096, bias=False)
(rotary_emb): LlamaRotaryEmbedding()
)
(mlp): LlamaMLP(
(gate_proj): Linear(in_features=4096, out_features=11008, bias=False)
(down_proj): Linear(in_features=11008, out_features=4096, bias=False)
(up_proj): Linear(in_features=4096, out_features=11008, bias=False)
(act_fn): SiLUActivation()
)
(input_layernorm): LlamaRMSNorm()
(post_attention_layernorm): LlamaRMSNorm()
)
)
(norm): LlamaRMSNorm()
)
(lm_head): Linear(in_features=4096, out_features=32000, bias=False)
)
```
## Citation
If this work is helpful, please kindly cite as:
```bibtex
@Misc{mamba-gpt-3b-v2,
title = {Mamba-GPT-3b-v2},
author = {chiliu},
howpublished = {\url{https://huggingface.co/CobraMamba/mamba-gpt-3b-v2}},
year = {2023}
}
```
## Disclaimer
Please read this disclaimer carefully before using the large language model provided in this repository. Your use of the model signifies your agreement to the following terms and conditions.
- Biases and Offensiveness: The large language model is trained on a diverse range of internet text data, which may contain biased, racist, offensive, or otherwise inappropriate content. By using this model, you acknowledge and accept that the generated content may sometimes exhibit biases or produce content that is offensive or inappropriate. The developers of this repository do not endorse, support, or promote any such content or viewpoints.
- Limitations: The large language model is an AI-based tool and not a human. It may produce incorrect, nonsensical, or irrelevant responses. It is the user's responsibility to critically evaluate the generated content and use it at their discretion.
- Use at Your Own Risk: Users of this large language model must assume full responsibility for any consequences that may arise from their use of the tool. The developers and contributors of this repository shall not be held liable for any damages, losses, or harm resulting from the use or misuse of the provided model.
- Ethical Considerations: Users are encouraged to use the large language model responsibly and ethically. By using this model, you agree not to use it for purposes that promote hate speech, discrimination, harassment, or any form of illegal or harmful activities.
- Reporting Issues: If you encounter any biased, offensive, or otherwise inappropriate content generated by the large language model, please report it to the repository maintainers through the provided channels. Your feedback will help improve the model and mitigate potential issues.
- Changes to this Disclaimer: The developers of this repository reserve the right to modify or update this disclaimer at any time without prior notice. It is the user's responsibility to periodically review the disclaimer to stay informed about any changes.
By using the large language model provided in this repository, you agree to accept and comply with the terms and conditions outlined in this disclaimer. If you do not agree with any part of this disclaimer, you should refrain from using the model and any content generated by it.
| 8,247 | [
[
-0.030059814453125,
-0.071533203125,
0.0193328857421875,
0.017913818359375,
-0.0343017578125,
-0.00832366943359375,
-0.01459503173828125,
-0.02142333984375,
0.017730712890625,
0.021270751953125,
-0.025726318359375,
-0.048004150390625,
-0.0523681640625,
-0.001010894775390625,
-0.00763702392578125,
0.068603515625,
-0.01335906982421875,
-0.00493621826171875,
0.01192474365234375,
0.0036182403564453125,
-0.0218505859375,
-0.04705810546875,
-0.043731689453125,
-0.0274505615234375,
0.0271759033203125,
-0.0016345977783203125,
0.04620361328125,
0.047271728515625,
0.0226287841796875,
0.023773193359375,
-0.0216522216796875,
0.009857177734375,
-0.0275115966796875,
-0.015167236328125,
0.01140594482421875,
-0.0253753662109375,
-0.041107177734375,
0.0027618408203125,
0.042388916015625,
0.0264892578125,
-0.011627197265625,
0.025726318359375,
0.00901031494140625,
0.027069091796875,
-0.040130615234375,
0.04119873046875,
-0.0364990234375,
-0.006694793701171875,
-0.022308349609375,
0.0036163330078125,
-0.022918701171875,
-0.031158447265625,
-0.0184783935546875,
-0.038909912109375,
-0.00917816162109375,
-0.0008940696716308594,
0.09967041015625,
0.017852783203125,
-0.0306549072265625,
-0.01605224609375,
-0.0270843505859375,
0.052703857421875,
-0.08343505859375,
0.019073486328125,
0.030029296875,
0.0126953125,
-0.01297760009765625,
-0.060302734375,
-0.052032470703125,
-0.01885986328125,
-0.0019426345825195312,
0.01395416259765625,
-0.0256805419921875,
-0.01082611083984375,
0.01995849609375,
0.029327392578125,
-0.038909912109375,
0.00684356689453125,
-0.033935546875,
-0.0205230712890625,
0.044677734375,
0.014892578125,
0.018524169921875,
-0.03497314453125,
-0.021453857421875,
-0.006008148193359375,
-0.04168701171875,
0.0194549560546875,
0.03753662109375,
0.0251312255859375,
-0.0288543701171875,
0.059783935546875,
-0.0093231201171875,
0.052032470703125,
0.0035266876220703125,
-0.0180206298828125,
0.03961181640625,
-0.0263671875,
-0.03619384765625,
-0.00016057491302490234,
0.079345703125,
0.0200653076171875,
0.0035247802734375,
0.01274871826171875,
-0.0120391845703125,
-0.00934600830078125,
-0.01132965087890625,
-0.070556640625,
-0.0174102783203125,
0.01534271240234375,
-0.04052734375,
-0.0308380126953125,
0.00028395652770996094,
-0.048553466796875,
-0.0207366943359375,
-0.0042724609375,
0.03515625,
-0.0300750732421875,
-0.03369140625,
0.01204681396484375,
0.0172271728515625,
0.0263671875,
-0.006622314453125,
-0.0673828125,
0.01091766357421875,
0.0310821533203125,
0.07843017578125,
0.0128631591796875,
-0.0190277099609375,
-0.0325927734375,
0.016021728515625,
-0.0152587890625,
0.04022216796875,
-0.0268096923828125,
-0.0379638671875,
-0.0200042724609375,
0.01102447509765625,
-0.0152587890625,
-0.0290985107421875,
0.03887939453125,
-0.01568603515625,
0.0184478759765625,
-0.01218414306640625,
-0.0269927978515625,
-0.028167724609375,
0.00960540771484375,
-0.03717041015625,
0.09002685546875,
0.014739990234375,
-0.0635986328125,
0.01552581787109375,
-0.058563232421875,
-0.0230560302734375,
-0.0226593017578125,
0.00656890869140625,
-0.056121826171875,
-0.01258087158203125,
0.040435791015625,
0.043853759765625,
-0.02508544921875,
0.022247314453125,
-0.0298919677734375,
-0.0189361572265625,
0.0179443359375,
-0.01971435546875,
0.09039306640625,
0.0247955322265625,
-0.036285400390625,
0.021453857421875,
-0.055938720703125,
-0.00424957275390625,
0.038665771484375,
-0.0287628173828125,
0.0026702880859375,
-0.0206451416015625,
-0.005523681640625,
0.0191497802734375,
0.016204833984375,
-0.0274505615234375,
0.033935546875,
-0.03948974609375,
0.051544189453125,
0.06927490234375,
0.01293182373046875,
0.01543426513671875,
-0.0179443359375,
0.0369873046875,
0.0164947509765625,
0.0255126953125,
0.0010137557983398438,
-0.05780029296875,
-0.06683349609375,
-0.034576416015625,
0.02294921875,
0.03350830078125,
-0.047393798828125,
0.047821044921875,
-0.0124664306640625,
-0.056915283203125,
-0.041351318359375,
0.013031005859375,
0.02325439453125,
0.044891357421875,
0.03472900390625,
-0.01317596435546875,
-0.04248046875,
-0.06768798828125,
0.01568603515625,
-0.0098419189453125,
0.004070281982421875,
0.0259246826171875,
0.055938720703125,
-0.0257720947265625,
0.047698974609375,
-0.0537109375,
-0.0192108154296875,
-0.0172271728515625,
0.00530242919921875,
0.04052734375,
0.04412841796875,
0.049652099609375,
-0.03466796875,
-0.02398681640625,
0.0011882781982421875,
-0.058135986328125,
-0.0038852691650390625,
0.008026123046875,
-0.019866943359375,
0.0212554931640625,
0.025238037109375,
-0.060516357421875,
0.0308380126953125,
0.04400634765625,
-0.0241241455078125,
0.039031982421875,
-0.014068603515625,
-0.0034885406494140625,
-0.0955810546875,
0.0246124267578125,
-0.0070648193359375,
0.0012159347534179688,
-0.0361328125,
-0.0034961700439453125,
0.0030002593994140625,
-0.0011453628540039062,
-0.050140380859375,
0.056915283203125,
-0.0352783203125,
0.0003178119659423828,
-0.0013837814331054688,
0.007110595703125,
-0.00353240966796875,
0.04638671875,
-0.01953125,
0.05194091796875,
0.05523681640625,
-0.051849365234375,
0.03802490234375,
0.0185699462890625,
-0.026153564453125,
0.0116424560546875,
-0.06475830078125,
0.0119476318359375,
0.0072021484375,
0.0307464599609375,
-0.0701904296875,
-0.022125244140625,
0.042633056640625,
-0.041748046875,
0.0290069580078125,
-0.0111541748046875,
-0.04705810546875,
-0.05157470703125,
-0.01885986328125,
0.0141448974609375,
0.04449462890625,
-0.034576416015625,
0.0482177734375,
0.028289794921875,
0.0011472702026367188,
-0.054229736328125,
-0.055999755859375,
-0.0109100341796875,
-0.0259552001953125,
-0.049072265625,
0.009918212890625,
0.003475189208984375,
-0.0035839080810546875,
-0.0035228729248046875,
0.00484466552734375,
0.007297515869140625,
0.0130157470703125,
0.0270843505859375,
0.02978515625,
-0.011505126953125,
-0.01074981689453125,
-0.0049591064453125,
-0.01010894775390625,
0.004573822021484375,
-0.01580810546875,
0.0687255859375,
-0.021026611328125,
-0.0184783935546875,
-0.042205810546875,
-0.005672454833984375,
0.03076171875,
-0.0251312255859375,
0.06878662109375,
0.062744140625,
-0.0316162109375,
0.0112762451171875,
-0.033721923828125,
-0.0025539398193359375,
-0.035400390625,
0.020965576171875,
-0.0360107421875,
-0.040863037109375,
0.059906005859375,
0.0234527587890625,
0.0007376670837402344,
0.0482177734375,
0.0677490234375,
0.01087188720703125,
0.07647705078125,
0.03173828125,
-0.007457733154296875,
0.026519775390625,
-0.052947998046875,
0.0172576904296875,
-0.0745849609375,
-0.0335693359375,
-0.030364990234375,
-0.0191802978515625,
-0.039031982421875,
-0.03619384765625,
0.0131988525390625,
0.017852783203125,
-0.0400390625,
0.0290985107421875,
-0.038299560546875,
0.0120086669921875,
0.0408935546875,
0.01390838623046875,
-0.0019512176513671875,
-0.01727294921875,
-0.0343017578125,
0.00759124755859375,
-0.03582763671875,
-0.042449951171875,
0.08270263671875,
0.0380859375,
0.046875,
0.0067138671875,
0.053558349609375,
-0.01268768310546875,
0.02960205078125,
-0.0438232421875,
0.039764404296875,
0.0120697021484375,
-0.05194091796875,
-0.007076263427734375,
-0.0269622802734375,
-0.072509765625,
0.021728515625,
-0.007122039794921875,
-0.0562744140625,
0.004150390625,
0.0016202926635742188,
-0.0262298583984375,
0.04083251953125,
-0.053436279296875,
0.062744140625,
-0.01221466064453125,
-0.0303192138671875,
0.00962066650390625,
-0.0465087890625,
0.0501708984375,
-0.0013952255249023438,
0.016754150390625,
-0.01447296142578125,
-0.01213836669921875,
0.05499267578125,
-0.03973388671875,
0.0587158203125,
-0.0266876220703125,
-0.006954193115234375,
0.03704833984375,
-0.00656890869140625,
0.038909912109375,
0.005374908447265625,
-0.0170440673828125,
0.0257110595703125,
-0.023590087890625,
-0.038421630859375,
-0.0352783203125,
0.056488037109375,
-0.0718994140625,
-0.047882080078125,
-0.0389404296875,
-0.032379150390625,
0.003009796142578125,
0.0203094482421875,
0.0219879150390625,
0.01806640625,
0.0106353759765625,
0.01314544677734375,
0.021270751953125,
-0.03619384765625,
0.04339599609375,
0.0110321044921875,
-0.036773681640625,
-0.042083740234375,
0.0645751953125,
0.0114593505859375,
0.0185546875,
0.004833221435546875,
0.015655517578125,
-0.032196044921875,
-0.035888671875,
-0.04345703125,
0.03363037109375,
-0.0565185546875,
-0.0276947021484375,
-0.053924560546875,
-0.0276641845703125,
-0.0377197265625,
-0.0030651092529296875,
-0.02215576171875,
-0.022796630859375,
-0.04071044921875,
-0.016876220703125,
0.033172607421875,
0.043670654296875,
-0.0167388916015625,
0.0214385986328125,
-0.043212890625,
0.01885986328125,
0.0213165283203125,
0.01611328125,
0.00152587890625,
-0.05670166015625,
-0.0161590576171875,
0.0119476318359375,
-0.040618896484375,
-0.058807373046875,
0.0450439453125,
-0.00478363037109375,
0.0506591796875,
0.024444580078125,
-0.01061248779296875,
0.059539794921875,
-0.015777587890625,
0.07684326171875,
0.0192718505859375,
-0.08050537109375,
0.050201416015625,
-0.028350830078125,
0.022125244140625,
0.00659942626953125,
0.0274505615234375,
-0.01290130615234375,
-0.032745361328125,
-0.050933837890625,
-0.07208251953125,
0.04644775390625,
0.032562255859375,
0.009979248046875,
-0.003047943115234375,
0.0262908935546875,
-0.005100250244140625,
0.00567626953125,
-0.0830078125,
-0.03875732421875,
-0.040283203125,
-0.007717132568359375,
-0.002780914306640625,
-0.0032291412353515625,
-0.0041351318359375,
-0.03302001953125,
0.0689697265625,
-0.003437042236328125,
0.03460693359375,
0.006496429443359375,
-0.01093292236328125,
0.000591278076171875,
0.004352569580078125,
0.055328369140625,
0.04705810546875,
-0.0196380615234375,
-0.0010004043579101562,
0.028564453125,
-0.0440673828125,
0.01241302490234375,
0.01953125,
-0.0151214599609375,
-0.0032634735107421875,
0.01409149169921875,
0.07318115234375,
0.01047515869140625,
-0.0192718505859375,
0.0333251953125,
-0.0033931732177734375,
-0.01806640625,
-0.00861358642578125,
0.00865936279296875,
0.0278472900390625,
0.01264190673828125,
0.022003173828125,
-0.004367828369140625,
-0.012115478515625,
-0.03753662109375,
-0.0033855438232421875,
0.0310821533203125,
-0.00690460205078125,
-0.01543426513671875,
0.083251953125,
0.00768280029296875,
-0.0079193115234375,
0.03961181640625,
-0.0099029541015625,
-0.037628173828125,
0.05792236328125,
0.039764404296875,
0.062103271484375,
-0.0185546875,
0.00476837158203125,
0.055633544921875,
0.02606201171875,
0.0036563873291015625,
0.01611328125,
0.01378631591796875,
-0.034149169921875,
-0.0233001708984375,
-0.05523681640625,
-0.005340576171875,
0.0205230712890625,
-0.048004150390625,
0.043243408203125,
-0.0450439453125,
-0.018524169921875,
-0.002025604248046875,
0.01079559326171875,
-0.0460205078125,
0.01551055908203125,
0.0162811279296875,
0.05242919921875,
-0.05810546875,
0.08245849609375,
0.035400390625,
-0.067138671875,
-0.07781982421875,
-0.00926971435546875,
-0.00009185075759887695,
-0.0771484375,
0.051727294921875,
0.0206146240234375,
-0.0014400482177734375,
0.0084686279296875,
-0.044586181640625,
-0.07080078125,
0.10211181640625,
0.032867431640625,
-0.02532958984375,
-0.0019292831420898438,
-0.005855560302734375,
0.03167724609375,
-0.0196380615234375,
0.04052734375,
0.03985595703125,
0.038543701171875,
0.0042266845703125,
-0.076904296875,
0.021697998046875,
-0.0215911865234375,
0.009674072265625,
0.003704071044921875,
-0.07391357421875,
0.08636474609375,
-0.0296478271484375,
-0.0104217529296875,
0.00472259521484375,
0.057952880859375,
0.036712646484375,
0.001598358154296875,
0.024627685546875,
0.04205322265625,
0.062164306640625,
-0.0162200927734375,
0.08953857421875,
-0.0243072509765625,
0.046600341796875,
0.06243896484375,
-0.005279541015625,
0.0576171875,
0.0235748291015625,
-0.02215576171875,
0.041656494140625,
0.06536865234375,
-0.016632080078125,
0.034759521484375,
0.0248260498046875,
-0.007213592529296875,
-0.0007638931274414062,
-0.0024585723876953125,
-0.057281494140625,
0.033203125,
0.0175933837890625,
-0.0318603515625,
0.0007467269897460938,
-0.00034046173095703125,
0.0247650146484375,
-0.020965576171875,
-0.0213165283203125,
0.034637451171875,
0.0116119384765625,
-0.03057861328125,
0.08001708984375,
0.003173828125,
0.056976318359375,
-0.04510498046875,
0.01367950439453125,
-0.01776123046875,
0.0120697021484375,
-0.0199127197265625,
-0.058319091796875,
0.010498046875,
-0.0016698837280273438,
0.0033016204833984375,
-0.0009965896606445312,
0.03179931640625,
-0.0142822265625,
-0.046783447265625,
0.02630615234375,
0.02215576171875,
0.0221099853515625,
0.01291656494140625,
-0.0673828125,
0.0206298828125,
0.0060272216796875,
-0.04815673828125,
0.01534271240234375,
0.0172271728515625,
0.018768310546875,
0.062103271484375,
0.05859375,
0.0073089599609375,
0.021240234375,
0.0078582763671875,
0.0810546875,
-0.053741455078125,
-0.024505615234375,
-0.07769775390625,
0.048736572265625,
-0.003070831298828125,
-0.0301513671875,
0.061920166015625,
0.06231689453125,
0.057403564453125,
-0.002880096435546875,
0.053558349609375,
-0.015899658203125,
0.018768310546875,
-0.050933837890625,
0.060150146484375,
-0.04180908203125,
0.015625,
-0.0266265869140625,
-0.0770263671875,
-0.02178955078125,
0.0665283203125,
-0.025665283203125,
0.0163116455078125,
0.048736572265625,
0.066650390625,
-0.002277374267578125,
-0.00875091552734375,
0.01068115234375,
0.04608154296875,
0.0396728515625,
0.054351806640625,
0.043304443359375,
-0.0523681640625,
0.053924560546875,
-0.036376953125,
-0.00588226318359375,
-0.0156402587890625,
-0.059295654296875,
-0.0771484375,
-0.0433349609375,
-0.0228271484375,
-0.049652099609375,
-0.018890380859375,
0.06964111328125,
0.0499267578125,
-0.057708740234375,
-0.013427734375,
0.00954437255859375,
0.006603240966796875,
-0.01007080078125,
-0.01788330078125,
0.04168701171875,
-0.00421142578125,
-0.06854248046875,
0.0096282958984375,
0.006778717041015625,
0.0240325927734375,
-0.0259246826171875,
-0.0155792236328125,
-0.036956787109375,
-0.000988006591796875,
0.038848876953125,
0.015869140625,
-0.066162109375,
-0.01433563232421875,
-0.00371551513671875,
-0.018890380859375,
0.01212310791015625,
0.02825927734375,
-0.04241943359375,
0.029144287109375,
0.0216522216796875,
0.022613525390625,
0.057891845703125,
-0.004970550537109375,
0.02154541015625,
-0.032135009765625,
0.029266357421875,
0.00534820556640625,
0.042327880859375,
0.0289306640625,
-0.027374267578125,
0.04736328125,
0.0189208984375,
-0.046630859375,
-0.07391357421875,
-0.0059051513671875,
-0.0850830078125,
-0.0156402587890625,
0.0992431640625,
-0.0180816650390625,
-0.03570556640625,
0.004673004150390625,
-0.032958984375,
0.04205322265625,
-0.02813720703125,
0.050018310546875,
0.0537109375,
-0.01220703125,
-0.00577545166015625,
-0.052032470703125,
0.01535797119140625,
0.00958251953125,
-0.0667724609375,
-0.00594329833984375,
0.01605224609375,
0.038421630859375,
0.007556915283203125,
0.0418701171875,
-0.01384735107421875,
0.006183624267578125,
0.0046844482421875,
0.00875091552734375,
-0.027557373046875,
0.01641845703125,
-0.017822265625,
-0.0013666152954101562,
-0.0133056640625,
-0.0206146240234375
]
] |
TigerResearch/tigerbot-70b-base | 2023-09-20T06:55:42.000Z | [
"transformers",
"pytorch",
"llama",
"text-generation",
"zh",
"en",
"license:apache-2.0",
"endpoints_compatible",
"text-generation-inference",
"region:us"
] | text-generation | TigerResearch | null | null | TigerResearch/tigerbot-70b-base | 12 | 5,965 | transformers | 2023-09-05T09:12:36 | ---
license: apache-2.0
language:
- zh
- en
---
<div style="width: 100%;">
<p align="center" width="20%">
<img src="http://x-pai.algolet.com/bot/img/logo_core.png" alt="TigerBot" width="20%", style="display: block; margin: auto;"></img>
</p>
</div>
<p align="center">
<font face="黑体" size=5"> A cutting-edge foundation for your very own LLM. </font>
</p>
<p align="center">
💻<a href="https://github.com/TigerResearch/TigerBot" target="_blank">Github</a> • 🌐 <a href="https://tigerbot.com/" target="_blank">TigerBot</a> • 🤗 <a href="https://huggingface.co/TigerResearch" target="_blank">Hugging Face</a>
</p>
# 快速开始
- 方法1，通过transformers使用
- 下载 TigerBot Repo
```shell
git clone https://github.com/TigerResearch/TigerBot.git
```
- 启动infer代码
```shell
python infer.py --model_path TigerResearch/tigerbot-70b-chat
```
- 方法2:
- 下载 TigerBot Repo
```shell
git clone https://github.com/TigerResearch/TigerBot.git
```
- 安装git lfs： `git lfs install`
- 通过huggingface或modelscope平台下载权重
```shell
git clone https://huggingface.co/TigerResearch/tigerbot-70b-chat
git clone https://www.modelscope.cn/TigerResearch/tigerbot-70b-chat-v1.git
```
- 启动infer代码
```shell
python infer.py --model_path tigerbot-70b-chat(-v1)
```
------
# Quick Start
- Method 1, use through transformers
- Clone TigerBot Repo
```shell
git clone https://github.com/TigerResearch/TigerBot.git
```
- Run infer script
```shell
python infer.py --model_path TigerResearch/tigerbot-70b-chat
```
- Method 2:
- Clone TigerBot Repo
```shell
git clone https://github.com/TigerResearch/TigerBot.git
```
- install git lfs： `git lfs install`
- Download weights from huggingface or modelscope
```shell
git clone https://huggingface.co/TigerResearch/tigerbot-70b-chat
git clone https://www.modelscope.cn/TigerResearch/tigerbot-70b-chat-v1.git
```
- Run infer script
```shell
python infer.py --model_path tigerbot-70b-chat(-v1)
``` | 2,089 | [
[
-0.0360107421875,
-0.046112060546875,
0.0179443359375,
0.0220184326171875,
-0.0270538330078125,
0.007030487060546875,
-0.004425048828125,
-0.0261688232421875,
0.043121337890625,
0.0212249755859375,
-0.06317138671875,
-0.027618408203125,
-0.0196533203125,
-0.00083160400390625,
0.0043487548828125,
0.07080078125,
0.007579803466796875,
-0.0035762786865234375,
-0.0231475830078125,
-0.0148468017578125,
0.00565338134765625,
-0.025909423828125,
-0.06329345703125,
-0.019195556640625,
0.0078582763671875,
0.01114654541015625,
0.059234619140625,
0.05694580078125,
0.05047607421875,
0.0303497314453125,
-0.0030364990234375,
0.004146575927734375,
-0.0143280029296875,
0.01354217529296875,
-0.0020961761474609375,
-0.0323486328125,
-0.0462646484375,
-0.004039764404296875,
0.043212890625,
0.0112457275390625,
-0.00032806396484375,
0.0135955810546875,
0.00861358642578125,
0.0477294921875,
-0.0426025390625,
0.025360107421875,
-0.02740478515625,
-0.01204681396484375,
-0.017669677734375,
0.0010166168212890625,
0.0014400482177734375,
-0.044219970703125,
0.0162353515625,
-0.062408447265625,
-0.00769805908203125,
-0.0023212432861328125,
0.10986328125,
0.002674102783203125,
-0.0126495361328125,
-0.0181427001953125,
-0.0007948875427246094,
0.05059814453125,
-0.07623291015625,
0.007114410400390625,
0.023529052734375,
0.01800537109375,
-0.01763916015625,
-0.0682373046875,
-0.04278564453125,
-0.0195159912109375,
-0.04241943359375,
0.006343841552734375,
-0.0303497314453125,
-0.005748748779296875,
0.0272674560546875,
0.0274810791015625,
-0.048248291015625,
-0.019561767578125,
-0.0239410400390625,
-0.0179290771484375,
0.04925537109375,
0.0003070831298828125,
0.032379150390625,
-0.035125732421875,
-0.0242919921875,
-0.035125732421875,
-0.018402099609375,
0.0260772705078125,
0.0255126953125,
0.021453857421875,
-0.0343017578125,
0.033905029296875,
-0.030914306640625,
0.018096923828125,
0.025360107421875,
-0.035736083984375,
0.031463623046875,
-0.0263519287109375,
-0.0172119140625,
-0.01123809814453125,
0.0858154296875,
0.040252685546875,
0.0059356689453125,
0.037261962890625,
-0.0321044921875,
-0.006984710693359375,
-0.01036834716796875,
-0.062103271484375,
-0.0108642578125,
0.026580810546875,
-0.034332275390625,
-0.0308380126953125,
0.016448974609375,
-0.068115234375,
-0.01090240478515625,
0.0171661376953125,
0.045166015625,
-0.0345458984375,
-0.037261962890625,
0.00457763671875,
-0.0193939208984375,
0.0416259765625,
0.015960693359375,
-0.054046630859375,
0.00252532958984375,
0.039520263671875,
0.06756591796875,
0.01983642578125,
-0.006237030029296875,
-0.018280029296875,
0.0123138427734375,
-0.004119873046875,
0.0262603759765625,
-0.00843048095703125,
-0.0355224609375,
0.00878143310546875,
0.020751953125,
-0.00140380859375,
-0.03125,
0.032470703125,
-0.01047515869140625,
0.0266571044921875,
-0.002651214599609375,
-0.0203857421875,
-0.03839111328125,
0.0229949951171875,
-0.038726806640625,
0.08123779296875,
0.033660888671875,
-0.0682373046875,
-0.006755828857421875,
-0.047943115234375,
-0.0340576171875,
0.002288818359375,
0.009368896484375,
-0.07025146484375,
-0.02447509765625,
0.0295257568359375,
0.057708740234375,
0.011444091796875,
-0.003932952880859375,
-0.061248779296875,
0.0013990402221679688,
0.01387786865234375,
0.015960693359375,
0.090576171875,
0.00370025634765625,
-0.0364990234375,
0.0104217529296875,
-0.042633056640625,
-0.0037384033203125,
0.0526123046875,
-0.0132904052734375,
-0.0220489501953125,
-0.0174102783203125,
-0.005474090576171875,
0.004627227783203125,
0.041168212890625,
-0.0309295654296875,
0.049224853515625,
-0.0286407470703125,
0.05010986328125,
0.04913330078125,
0.00429534912109375,
0.04522705078125,
-0.060821533203125,
0.0275726318359375,
0.0012273788452148438,
0.0233612060546875,
-0.01678466796875,
-0.019500732421875,
-0.0751953125,
-0.03228759765625,
-0.016510009765625,
0.03656005859375,
-0.057220458984375,
0.03289794921875,
-0.0222930908203125,
-0.056884765625,
-0.03497314453125,
0.0020294189453125,
0.0240020751953125,
0.01444244384765625,
0.0094451904296875,
-0.031097412109375,
-0.057861328125,
-0.072021484375,
-0.0001499652862548828,
-0.032196044921875,
0.0028438568115234375,
0.034912109375,
0.06793212890625,
-0.029144287109375,
0.0499267578125,
-0.04412841796875,
-0.01364898681640625,
-0.01406097412109375,
-0.0189361572265625,
0.036376953125,
0.056060791015625,
0.06317138671875,
-0.0355224609375,
-0.039520263671875,
-0.0073699951171875,
-0.0574951171875,
0.01091766357421875,
-0.01806640625,
-0.0333251953125,
0.0153961181640625,
0.01119232177734375,
-0.07025146484375,
0.02685546875,
0.02740478515625,
-0.0169677734375,
0.048004150390625,
0.004871368408203125,
0.00981903076171875,
-0.08563232421875,
0.006679534912109375,
0.0244903564453125,
-0.0278167724609375,
-0.0218048095703125,
0.030548095703125,
-0.004486083984375,
0.01261138916015625,
-0.037628173828125,
0.057464599609375,
-0.031494140625,
0.0064849853515625,
0.00899505615234375,
0.01352691650390625,
0.01149749755859375,
0.040252685546875,
-0.0156707763671875,
0.0511474609375,
0.054656982421875,
-0.0362548828125,
0.04278564453125,
0.0224456787109375,
-0.01169586181640625,
0.0296173095703125,
-0.044189453125,
0.0030384063720703125,
0.024566650390625,
0.026611328125,
-0.07403564453125,
-0.022552490234375,
0.06005859375,
-0.049530029296875,
0.0257720947265625,
-0.01430511474609375,
-0.0260009765625,
-0.03228759765625,
-0.063232421875,
0.0018215179443359375,
0.05072021484375,
-0.034912109375,
0.045501708984375,
0.0262451171875,
-0.0005931854248046875,
-0.03277587890625,
-0.043243408203125,
-0.0263519287109375,
-0.025848388671875,
-0.05426025390625,
0.0171051025390625,
-0.00641632080078125,
-0.02569580078125,
-0.0058746337890625,
-0.0020694732666015625,
-0.0221405029296875,
-0.0072479248046875,
0.0316162109375,
0.0280303955078125,
-0.039794921875,
-0.0247650146484375,
-0.01262664794921875,
0.0028781890869140625,
0.013702392578125,
-0.022705078125,
0.027008056640625,
-0.0203094482421875,
-0.0275421142578125,
-0.05401611328125,
-0.018524169921875,
0.057586669921875,
-0.0107269287109375,
0.037628173828125,
0.056060791015625,
-0.015960693359375,
-0.00022542476654052734,
-0.04815673828125,
-0.0308380126953125,
-0.039520263671875,
0.015625,
-0.0184326171875,
-0.041900634765625,
0.034393310546875,
0.00991058349609375,
0.0157623291015625,
0.046661376953125,
0.022552490234375,
-0.0203704833984375,
0.0654296875,
0.0474853515625,
-0.024627685546875,
0.054168701171875,
-0.039306640625,
-0.015716552734375,
-0.05389404296875,
-0.01312255859375,
-0.03009033203125,
-0.006053924560546875,
-0.059234619140625,
-0.025909423828125,
0.0232696533203125,
0.01763916015625,
-0.0316162109375,
0.051971435546875,
-0.06817626953125,
-0.00659942626953125,
0.060089111328125,
0.025115966796875,
0.0131072998046875,
0.00388336181640625,
-0.017425537109375,
0.0019016265869140625,
-0.0283355712890625,
-0.037872314453125,
0.0643310546875,
0.02288818359375,
0.04071044921875,
0.0043792724609375,
0.045806884765625,
-0.016998291015625,
0.015899658203125,
-0.046722412109375,
0.04937744140625,
0.01983642578125,
-0.058807373046875,
-0.028778076171875,
-0.0081787109375,
-0.06292724609375,
0.0233612060546875,
-0.014190673828125,
-0.054931640625,
-0.0222930908203125,
-0.00704193115234375,
-0.022216796875,
0.04339599609375,
-0.025299072265625,
0.033782958984375,
-0.02490234375,
-0.03277587890625,
-0.00201416015625,
-0.048004150390625,
0.044708251953125,
0.005176544189453125,
0.006053924560546875,
-0.0215301513671875,
-0.00826263427734375,
0.06439208984375,
-0.052490234375,
0.05023193359375,
-0.001163482666015625,
0.007572174072265625,
0.036895751953125,
0.00505828857421875,
0.051605224609375,
0.02880859375,
-0.0132598876953125,
0.0026760101318359375,
0.0216064453125,
-0.0153961181640625,
-0.043243408203125,
0.03729248046875,
-0.05596923828125,
-0.055633544921875,
-0.04962158203125,
-0.00534820556640625,
0.0249176025390625,
0.02166748046875,
0.00909423828125,
0.0046844482421875,
0.01073455810546875,
0.02093505859375,
0.03497314453125,
-0.019622802734375,
0.049896240234375,
0.02435302734375,
-0.034759521484375,
-0.05145263671875,
0.046661376953125,
-0.01593017578125,
0.0111236572265625,
0.041839599609375,
0.0148773193359375,
-0.0272674560546875,
-0.004241943359375,
-0.035675048828125,
0.041290283203125,
-0.039215087890625,
-0.0259857177734375,
-0.03668212890625,
-0.0440673828125,
-0.03826904296875,
-0.0281829833984375,
-0.029693603515625,
-0.0173492431640625,
-0.0386962890625,
0.0201416015625,
0.066650390625,
0.03497314453125,
-0.0172271728515625,
0.032684326171875,
-0.04705810546875,
0.00955963134765625,
0.030426025390625,
0.01154327392578125,
-0.0055389404296875,
-0.042724609375,
-0.0037593841552734375,
0.01288604736328125,
-0.047821044921875,
-0.055419921875,
0.049652099609375,
-0.0009908676147460938,
0.049774169921875,
0.0361328125,
0.00804901123046875,
0.06390380859375,
0.007091522216796875,
0.04449462890625,
0.02264404296875,
-0.07415771484375,
0.059661865234375,
-0.04351806640625,
0.01110076904296875,
0.0230865478515625,
0.033203125,
-0.0323486328125,
-0.0268096923828125,
-0.057037353515625,
-0.046905517578125,
0.05316162109375,
0.034210205078125,
-0.00030350685119628906,
0.035308837890625,
0.0308380126953125,
-0.0203857421875,
0.004543304443359375,
-0.0650634765625,
-0.0440673828125,
-0.0184478759765625,
0.006679534912109375,
-0.00025272369384765625,
-0.0088958740234375,
0.000041425228118896484,
-0.036956787109375,
0.07373046875,
-0.00849151611328125,
0.0178985595703125,
0.0190277099609375,
0.00548553466796875,
-0.01476287841796875,
-0.00650787353515625,
0.03582763671875,
0.033782958984375,
-0.0305023193359375,
-0.017059326171875,
0.018798828125,
-0.0275421142578125,
0.01039886474609375,
0.0019273757934570312,
-0.00922393798828125,
0.02056884765625,
0.03448486328125,
0.046783447265625,
0.0184478759765625,
-0.024688720703125,
0.049407958984375,
0.00968170166015625,
-0.0285491943359375,
-0.03289794921875,
0.0020275115966796875,
0.0312042236328125,
0.03912353515625,
0.024261474609375,
-0.00919342041015625,
-0.0085296630859375,
-0.0276031494140625,
0.025360107421875,
0.054595947265625,
-0.028106689453125,
-0.01070404052734375,
0.0458984375,
0.00421142578125,
-0.0188446044921875,
0.04229736328125,
-0.013427734375,
-0.052276611328125,
0.06494140625,
0.016265869140625,
0.06683349609375,
0.0027008056640625,
0.0043487548828125,
0.0604248046875,
0.020599365234375,
-0.01036834716796875,
0.00971221923828125,
0.00421142578125,
-0.034912109375,
0.0026645660400390625,
-0.050079345703125,
-0.02996826171875,
0.0318603515625,
-0.040985107421875,
0.04278564453125,
-0.055419921875,
-0.01491546630859375,
-0.0005903244018554688,
0.02838134765625,
-0.0360107421875,
0.00762939453125,
-0.0027370452880859375,
0.05194091796875,
-0.033599853515625,
0.055145263671875,
0.0753173828125,
-0.053863525390625,
-0.06878662109375,
-0.00717926025390625,
0.026123046875,
-0.07550048828125,
0.035919189453125,
0.0172576904296875,
0.00846099853515625,
-0.0010080337524414062,
-0.059906005859375,
-0.072509765625,
0.09759521484375,
0.01035308837890625,
-0.037261962890625,
0.004119873046875,
-0.0252532958984375,
0.00970458984375,
-0.0159759521484375,
0.047698974609375,
0.017303466796875,
0.04412841796875,
0.005542755126953125,
-0.08465576171875,
0.0309600830078125,
-0.0233154296875,
-0.00426483154296875,
0.004154205322265625,
-0.0733642578125,
0.07098388671875,
-0.005214691162109375,
-0.0195465087890625,
0.0233154296875,
0.044647216796875,
0.046142578125,
0.01605224609375,
0.03277587890625,
0.0399169921875,
0.0192108154296875,
-0.01641845703125,
0.043487548828125,
-0.0246124267578125,
0.06591796875,
0.040252685546875,
0.0101470947265625,
0.047821044921875,
0.0207061767578125,
-0.03851318359375,
0.049957275390625,
0.06976318359375,
-0.0238037109375,
0.025177001953125,
0.022064208984375,
-0.01898193359375,
-0.00914764404296875,
0.001384735107421875,
-0.045074462890625,
0.0239715576171875,
0.0222015380859375,
-0.022979736328125,
-0.00859832763671875,
-0.022735595703125,
0.01953125,
-0.01605224609375,
-0.00417327880859375,
0.047271728515625,
0.00951385498046875,
-0.034881591796875,
0.05010986328125,
-0.01238250732421875,
0.0870361328125,
-0.0604248046875,
-0.002010345458984375,
-0.02880859375,
0.034210205078125,
-0.0178375244140625,
-0.068359375,
0.004802703857421875,
-0.03765869140625,
0.001800537109375,
-0.01276397705078125,
0.0653076171875,
-0.005401611328125,
-0.0284423828125,
0.02947998046875,
0.042083740234375,
0.0189208984375,
-0.0064849853515625,
-0.0875244140625,
0.00995635986328125,
0.023040771484375,
-0.05596923828125,
0.01678466796875,
0.032012939453125,
0.0027008056640625,
0.07806396484375,
0.059478759765625,
0.0034809112548828125,
-0.0189361572265625,
-0.041900634765625,
0.07281494140625,
-0.049346923828125,
-0.0238800048828125,
-0.08624267578125,
0.041473388671875,
-0.00739288330078125,
-0.0257720947265625,
0.061309814453125,
0.0394287109375,
0.07537841796875,
-0.0101776123046875,
0.0682373046875,
-0.023193359375,
0.0202484130859375,
-0.00771331787109375,
0.07843017578125,
-0.0767822265625,
-0.01444244384765625,
-0.036895751953125,
-0.044189453125,
-0.01401519775390625,
0.070068359375,
0.005252838134765625,
0.015533447265625,
0.042755126953125,
0.09033203125,
-0.00542449951171875,
-0.0311126708984375,
0.00893402099609375,
-0.0018978118896484375,
0.01495361328125,
0.052947998046875,
0.0570068359375,
-0.04461669921875,
0.057159423828125,
-0.05426025390625,
-0.0029087066650390625,
-0.041259765625,
-0.06292724609375,
-0.0699462890625,
-0.04229736328125,
-0.033416748046875,
-0.052764892578125,
-0.004444122314453125,
0.07080078125,
0.07745361328125,
-0.05194091796875,
-0.03192138671875,
0.0177764892578125,
0.0270843505859375,
-0.0120391845703125,
-0.0237274169921875,
0.021942138671875,
0.0098724365234375,
-0.057586669921875,
-0.0091705322265625,
0.027984619140625,
0.0302276611328125,
-0.021759033203125,
-0.01248931884765625,
-0.0311279296875,
-0.0165557861328125,
0.0182342529296875,
0.044219970703125,
-0.07147216796875,
-0.002826690673828125,
-0.01171875,
-0.052703857421875,
0.021484375,
0.01474761962890625,
-0.04296875,
0.01432037353515625,
0.0487060546875,
0.0179595947265625,
0.02911376953125,
-0.0108642578125,
0.021392822265625,
-0.0283050537109375,
0.01401519775390625,
-0.00481414794921875,
0.036376953125,
0.0176239013671875,
-0.03131103515625,
0.051971435546875,
0.043060302734375,
-0.029541015625,
-0.0489501953125,
-0.004230499267578125,
-0.07489013671875,
-0.01546478271484375,
0.0826416015625,
-0.018646240234375,
-0.0266571044921875,
0.0298004150390625,
-0.0268096923828125,
0.05059814453125,
-0.058868408203125,
0.017364501953125,
0.038116455078125,
-0.006610870361328125,
-0.0160064697265625,
-0.05059814453125,
0.0379638671875,
0.0180511474609375,
-0.05279541015625,
-0.0352783203125,
-0.000049233436584472656,
0.03125,
0.02838134765625,
0.050872802734375,
0.01032257080078125,
0.00391387939453125,
0.004375457763671875,
0.024810791015625,
-0.0009675025939941406,
0.03668212890625,
-0.0189056396484375,
-0.009613037109375,
-0.0188446044921875,
-0.03399658203125
]
] |
CHIH-HUNG/llama-2-13b-Open_Platypus_and_ccp_2.6w-3_epoch | 2023-09-11T17:57:44.000Z | [
"transformers",
"pytorch",
"llama",
"text-generation",
"dataset:garage-bAInd/Open-Platypus",
"license:llama2",
"endpoints_compatible",
"text-generation-inference",
"region:us"
] | text-generation | CHIH-HUNG | null | null | CHIH-HUNG/llama-2-13b-Open_Platypus_and_ccp_2.6w-3_epoch | 0 | 5,965 | transformers | 2023-09-05T15:35:40 | ---
license: llama2
datasets:
- garage-bAInd/Open-Platypus
---
# Model Card for Model ID
<!-- Provide a quick summary of what the model is/does. -->
在llama-2-13b上使用garage-bAInd/Open-Platypus資料集進行訓練，總資料筆數約2.5w + ccp
# Fine-Tuning Information
- **GPU:** RTX4090 (single core / 24564MiB)
- **model:** meta-llama/Llama-2-13b-hf
- **dataset:** garage-bAInd/Open-Platypus (共約2.5w筆訓練集) + ccp (約1200筆)
- **peft_type:** LoRA
- **lora_rank:** 8
- **lora_target:** gate_proj, up_proj, down_proj
- **per_device_train_batch_size:** 8
- **gradient_accumulation_steps:** 8
- **learning_rate :** 5e-5
- **epoch:** 3
- **precision:** bf16
- **quantization:** load_in_4bit
# Fine-Tuning Detail
- **train_loss:** 0.6
- **train_runtime:** 12:24:34 (use deepspeed)
# Evaluation
- 評估結果來自**HuggingFaceH4/open_llm_leaderboard**
- 與Llama-2-13b比較4種Benchmark，包含**ARC**、**HellaSwag**、**MMLU**、**TruthfulQA**
| Model |Average| ARC |HellaSwag| MMLU |TruthfulQA|
|---------------------------------------------------------|-------|-------|---------|-------|----------|
|meta-llama/Llama-2-13b-hf | 56.9 | 58.11 | 80.97 | 54.34 | 34.17 |
|meta-llama/Llama-2-13b-chat-hf | 59.93 | 59.04 | 81.94 | 54.64 | 44.12 |
|Open-Orca/OpenOrca-Platypus2-13B | 63.19 | 61.52 | 82.27 | 58.85 | 50.11 |
|CHIH-HUNG/llama-2-13b-Open_Platypus_and_ccp_2.6w | 59.41 | 58.96 | 82.51 | 56.12 | 40.07 |
|CHIH-HUNG/llama-2-13b-Open_Platypus_and_ccp_2.6w-3_epoch | 59.78 | 58.62 | 82.56 | 55.84 | 42.09 |
# How to convert dataset to json
- 在**load_dataset**中輸入資料集名稱，並且在**take**中輸入要取前幾筆資料
- 觀察該資料集的欄位名稱，填入**example**欄位中(例如instruction、input、output)
- 最後指定json檔儲存位置 (**json_filename**)
```py
import json
from datasets import load_dataset
# 讀取數據集，take可以取得該數據集前n筆資料
dataset = load_dataset("garage-bAInd/Open-Platypus", split="train", streaming=True)
# 提取所需欄位並建立新的字典列表
extracted_data = []
for example in dataset:
extracted_example = {
"instruction": example["instruction"],
"input": example["input"],
"output": example["output"]
}
extracted_data.append(extracted_example)
# 指定 JSON 文件名稱
json_filename = "Open-Platypus.json"
# 寫入 JSON 文件
with open(json_filename, "w") as json_file:
json.dump(extracted_data, json_file, indent=4)
print(f"數據已提取並保存為 {json_filename}")
``` | 2,425 | [
[
-0.039825439453125,
-0.041046142578125,
0.0157318115234375,
0.01248931884765625,
-0.05047607421875,
0.00860595703125,
-0.01312255859375,
-0.017852783203125,
0.0138702392578125,
0.035369873046875,
-0.04144287109375,
-0.04693603515625,
-0.044708251953125,
0.00942230224609375,
-0.01544189453125,
0.08026123046875,
-0.0146331787109375,
-0.015716552734375,
0.01380157470703125,
0.00388336181640625,
-0.0399169921875,
-0.033599853515625,
-0.04693603515625,
-0.0235595703125,
0.01540374755859375,
0.0102996826171875,
0.052154541015625,
0.0771484375,
0.051910400390625,
0.0212249755859375,
-0.00982666015625,
0.0232391357421875,
-0.03814697265625,
-0.0192108154296875,
0.0175018310546875,
-0.04449462890625,
-0.04437255859375,
-0.00493621826171875,
0.04742431640625,
0.025390625,
0.00415802001953125,
0.04132080078125,
0.009796142578125,
0.043426513671875,
-0.02911376953125,
0.027862548828125,
-0.02740478515625,
0.00521087646484375,
-0.0275726318359375,
-0.026824951171875,
-0.004436492919921875,
-0.01971435546875,
-0.0145721435546875,
-0.068603515625,
0.0109100341796875,
0.00884246826171875,
0.10302734375,
0.0295867919921875,
-0.0219879150390625,
0.0009055137634277344,
-0.035125732421875,
0.06573486328125,
-0.07611083984375,
-0.00383758544921875,
0.023040771484375,
0.0270233154296875,
-0.0156097412109375,
-0.048919677734375,
-0.053497314453125,
0.0078277587890625,
-0.015655517578125,
0.014801025390625,
-0.0028514862060546875,
-0.0229949951171875,
0.023712158203125,
0.037628173828125,
-0.03173828125,
0.0024585723876953125,
-0.036834716796875,
0.0079803466796875,
0.065673828125,
0.0352783203125,
0.000057578086853027344,
-0.019927978515625,
-0.02117919921875,
-0.030853271484375,
-0.03515625,
0.0255126953125,
0.036834716796875,
0.033477783203125,
-0.03497314453125,
0.0355224609375,
-0.0394287109375,
0.031585693359375,
-0.002925872802734375,
-0.040069580078125,
0.052490234375,
-0.01715087890625,
-0.037506103515625,
-0.0029430389404296875,
0.078369140625,
0.047271728515625,
-0.00586700439453125,
0.0143280029296875,
-0.009979248046875,
-0.01244354248046875,
-0.0030193328857421875,
-0.05877685546875,
-0.02459716796875,
0.042236328125,
-0.050628662109375,
-0.032012939453125,
0.007183074951171875,
-0.06610107421875,
-0.009246826171875,
-0.0002944469451904297,
0.0193328857421875,
-0.0279083251953125,
-0.041778564453125,
0.0079345703125,
-0.0168914794921875,
0.025421142578125,
0.02569580078125,
-0.0538330078125,
0.015106201171875,
0.04461669921875,
0.058380126953125,
0.00955963134765625,
-0.0259246826171875,
-0.0031585693359375,
0.0168914794921875,
-0.0250701904296875,
0.0504150390625,
-0.00702667236328125,
-0.0308990478515625,
-0.021270751953125,
0.021820068359375,
-0.007183074951171875,
-0.038665771484375,
0.05450439453125,
-0.0293121337890625,
-0.0087127685546875,
-0.035186767578125,
-0.01546478271484375,
-0.041900634765625,
0.0301513671875,
-0.05120849609375,
0.08123779296875,
0.00506591796875,
-0.06756591796875,
0.02679443359375,
-0.055572509765625,
-0.020050048828125,
0.007198333740234375,
0.00228118896484375,
-0.039215087890625,
-0.0204925537109375,
0.028411865234375,
0.04071044921875,
-0.034576416015625,
0.0128326416015625,
-0.016082763671875,
-0.032684326171875,
0.02630615234375,
-0.0193634033203125,
0.07171630859375,
0.02728271484375,
-0.015045166015625,
0.001007080078125,
-0.06787109375,
-0.00007575750350952148,
0.05291748046875,
-0.03497314453125,
-0.005382537841796875,
-0.0016498565673828125,
-0.00682830810546875,
-0.004253387451171875,
0.0347900390625,
-0.0191650390625,
0.02630615234375,
-0.01299285888671875,
0.0355224609375,
0.06317138671875,
-0.0008859634399414062,
0.00826263427734375,
-0.04150390625,
0.02044677734375,
0.0082550048828125,
0.0227508544921875,
-0.008148193359375,
-0.0413818359375,
-0.07098388671875,
-0.0269927978515625,
0.007320404052734375,
0.04022216796875,
-0.0298004150390625,
0.05914306640625,
-0.0250701904296875,
-0.0537109375,
-0.0537109375,
0.00782012939453125,
0.0131683349609375,
0.042999267578125,
0.038238525390625,
0.00030112266540527344,
-0.054290771484375,
-0.067626953125,
0.007480621337890625,
-0.006092071533203125,
0.0087432861328125,
0.02581787109375,
0.055816650390625,
-0.01617431640625,
0.04266357421875,
-0.037628173828125,
-0.020477294921875,
-0.02264404296875,
0.0023593902587890625,
0.066650390625,
0.043121337890625,
0.0479736328125,
-0.03253173828125,
-0.0298309326171875,
0.0024566650390625,
-0.08477783203125,
0.0118408203125,
-0.00725555419921875,
-0.0177459716796875,
-0.006488800048828125,
-0.0023632049560546875,
-0.049560546875,
0.0372314453125,
0.030487060546875,
-0.00894927978515625,
0.037933349609375,
0.0037364959716796875,
0.0226898193359375,
-0.07586669921875,
0.0158233642578125,
-0.02099609375,
0.006526947021484375,
-0.024993896484375,
0.0169830322265625,
-0.010711669921875,
0.0196075439453125,
-0.0294647216796875,
0.0186767578125,
-0.02490234375,
0.005916595458984375,
-0.0115814208984375,
-0.00232696533203125,
0.0036869049072265625,
0.0406494140625,
-0.003704071044921875,
0.048828125,
0.043060302734375,
-0.048736572265625,
0.040252685546875,
0.0357666015625,
-0.026214599609375,
0.005764007568359375,
-0.03521728515625,
0.00405120849609375,
0.005054473876953125,
0.026214599609375,
-0.0740966796875,
-0.0225677490234375,
0.0419921875,
-0.029937744140625,
0.0155029296875,
-0.031585693359375,
-0.035003662109375,
-0.047698974609375,
-0.038909912109375,
0.016265869140625,
0.03265380859375,
-0.047271728515625,
0.02081298828125,
0.00809478759765625,
0.01371002197265625,
-0.051910400390625,
-0.05645751953125,
-0.005146026611328125,
-0.017425537109375,
-0.04144287109375,
0.01560211181640625,
-0.01160430908203125,
-0.0035228729248046875,
0.003894805908203125,
0.0027446746826171875,
0.004119873046875,
0.00508880615234375,
0.0202789306640625,
0.039306640625,
-0.0215606689453125,
-0.0302886962890625,
-0.000013768672943115234,
-0.0067138671875,
0.00439453125,
0.0123748779296875,
0.06475830078125,
-0.0202789306640625,
-0.00887298583984375,
-0.05426025390625,
0.0033702850341796875,
0.031097412109375,
0.0012359619140625,
0.05377197265625,
0.05316162109375,
-0.013092041015625,
0.0030460357666015625,
-0.01971435546875,
0.00264739990234375,
-0.036773681640625,
0.0204925537109375,
-0.044708251953125,
-0.046722412109375,
0.05462646484375,
-0.0067138671875,
0.0192718505859375,
0.06561279296875,
0.0286407470703125,
-0.0158843994140625,
0.07794189453125,
0.013092041015625,
-0.01184844970703125,
0.01337432861328125,
-0.07574462890625,
0.007472991943359375,
-0.0789794921875,
-0.034698486328125,
-0.0335693359375,
-0.041748046875,
-0.04296875,
-0.0171661376953125,
0.01268768310546875,
0.023681640625,
-0.04327392578125,
0.028717041015625,
-0.06048583984375,
0.02001953125,
0.043121337890625,
0.01493072509765625,
0.0131988525390625,
-0.006069183349609375,
0.007450103759765625,
0.001499176025390625,
-0.036407470703125,
-0.0394287109375,
0.1026611328125,
0.02716064453125,
0.048492431640625,
-0.0018224716186523438,
0.05364990234375,
0.0063934326171875,
0.0157470703125,
-0.040924072265625,
0.046966552734375,
-0.00478363037109375,
-0.049407958984375,
-0.0079803466796875,
-0.0203857421875,
-0.05108642578125,
0.0264434814453125,
-0.013702392578125,
-0.060882568359375,
0.0037670135498046875,
0.0024814605712890625,
-0.031646728515625,
0.049407958984375,
-0.03729248046875,
0.050689697265625,
-0.035491943359375,
-0.0228118896484375,
0.003170013427734375,
-0.044647216796875,
0.05487060546875,
0.003093719482421875,
0.01561737060546875,
-0.0281524658203125,
-0.005435943603515625,
0.0828857421875,
-0.04998779296875,
0.0478515625,
-0.024810791015625,
0.007801055908203125,
0.0390625,
0.00287628173828125,
0.056365966796875,
0.0208892822265625,
-0.001220703125,
0.045989990234375,
0.0016527175903320312,
-0.018035888671875,
-0.02117919921875,
0.0531005859375,
-0.09393310546875,
-0.044830322265625,
-0.045867919921875,
-0.03094482421875,
0.0157928466796875,
0.02264404296875,
0.03118896484375,
-0.0117340087890625,
0.021240234375,
0.0186309814453125,
0.02801513671875,
-0.01104736328125,
0.0457763671875,
0.03021240234375,
-0.01226806640625,
-0.05645751953125,
0.062744140625,
0.0029010772705078125,
-0.005634307861328125,
0.0297393798828125,
0.0088958740234375,
-0.023223876953125,
-0.05145263671875,
-0.042083740234375,
0.0262603759765625,
-0.04266357421875,
-0.049957275390625,
-0.042572021484375,
-0.0382080078125,
-0.0382080078125,
-0.007419586181640625,
-0.039703369140625,
-0.0167999267578125,
-0.0596923828125,
-0.00875091552734375,
0.051910400390625,
0.032684326171875,
-0.004596710205078125,
0.059417724609375,
-0.054534912109375,
0.0231781005859375,
0.012359619140625,
0.00958251953125,
0.005985260009765625,
-0.05804443359375,
-0.0143585205078125,
0.0032596588134765625,
-0.031768798828125,
-0.0460205078125,
0.057220458984375,
0.00008285045623779297,
0.041351318359375,
0.04803466796875,
0.0020313262939453125,
0.08367919921875,
-0.0101776123046875,
0.0697021484375,
0.0166015625,
-0.049835205078125,
0.03985595703125,
-0.0248565673828125,
-0.013641357421875,
0.03253173828125,
0.0274658203125,
-0.017486572265625,
-0.0013408660888671875,
-0.043121337890625,
-0.0667724609375,
0.0772705078125,
0.0129241943359375,
-0.013946533203125,
0.0208892822265625,
0.018646240234375,
0.0160369873046875,
0.024383544921875,
-0.0623779296875,
-0.03826904296875,
-0.039947509765625,
-0.0021915435791015625,
0.0021228790283203125,
-0.0031871795654296875,
-0.0255889892578125,
-0.034698486328125,
0.0567626953125,
-0.001560211181640625,
0.034637451171875,
0.0101776123046875,
0.00998687744140625,
-0.0168304443359375,
0.0008749961853027344,
0.0286102294921875,
0.03253173828125,
-0.050079345703125,
-0.004383087158203125,
0.0115509033203125,
-0.043212890625,
-0.0006160736083984375,
0.0081634521484375,
-0.0111541748046875,
-0.0157470703125,
0.034881591796875,
0.06097412109375,
-0.01068115234375,
-0.03375244140625,
0.025115966796875,
0.002140045166015625,
-0.0219879150390625,
-0.02972412109375,
0.021636962890625,
-0.0031890869140625,
0.03570556640625,
0.04364013671875,
0.002178192138671875,
0.0005469322204589844,
-0.0205230712890625,
-0.01476287841796875,
0.023162841796875,
0.0277862548828125,
-0.0203399658203125,
0.0665283203125,
0.002838134765625,
-0.0120697021484375,
0.04736328125,
-0.015411376953125,
-0.034820556640625,
0.058563232421875,
0.039703369140625,
0.051910400390625,
-0.004337310791015625,
-0.003383636474609375,
0.057952880859375,
0.03741455078125,
-0.0100555419921875,
0.040283203125,
-0.0036163330078125,
-0.049774169921875,
-0.018402099609375,
-0.058197021484375,
-0.0091094970703125,
0.051971435546875,
-0.047332763671875,
0.0177154541015625,
-0.059661865234375,
-0.021209716796875,
-0.001708984375,
0.0277252197265625,
-0.055389404296875,
0.0198211669921875,
0.00890350341796875,
0.06719970703125,
-0.0640869140625,
0.06353759765625,
0.0284576416015625,
-0.038482666015625,
-0.06365966796875,
-0.0278778076171875,
-0.0146331787109375,
-0.08074951171875,
0.041046142578125,
0.00959014892578125,
0.0225982666015625,
-0.008209228515625,
-0.0697021484375,
-0.08123779296875,
0.111572265625,
0.01739501953125,
-0.04559326171875,
0.0150146484375,
0.0164337158203125,
0.024139404296875,
-0.0222015380859375,
0.0299072265625,
0.0538330078125,
0.046600341796875,
0.0067291259765625,
-0.057159423828125,
0.0213623046875,
-0.034149169921875,
-0.0066680908203125,
0.0035343170166015625,
-0.087158203125,
0.1007080078125,
-0.01409912109375,
0.0006270408630371094,
0.00969696044921875,
0.0447998046875,
0.040008544921875,
0.0204925537109375,
0.0292510986328125,
0.059112548828125,
0.049224853515625,
-0.0218048095703125,
0.060638427734375,
-0.0082550048828125,
0.037109375,
0.060272216796875,
-0.011749267578125,
0.059783935546875,
0.0280914306640625,
-0.037689208984375,
0.03955078125,
0.0711669921875,
-0.032745361328125,
0.0552978515625,
-0.01483917236328125,
-0.00751495361328125,
-0.010986328125,
0.0021648406982421875,
-0.05078125,
0.022308349609375,
0.03179931640625,
-0.0299072265625,
0.000225067138671875,
-0.021392822265625,
0.016510009765625,
-0.0313720703125,
-0.022491455078125,
0.039398193359375,
-0.0204925537109375,
-0.0232696533203125,
0.07232666015625,
-0.0003414154052734375,
0.054595947265625,
-0.0469970703125,
-0.00865936279296875,
-0.014923095703125,
0.009765625,
-0.035491943359375,
-0.0653076171875,
-0.0026378631591796875,
0.001476287841796875,
-0.01129913330078125,
0.01380157470703125,
0.035369873046875,
-0.002513885498046875,
-0.036102294921875,
0.0225372314453125,
0.00830078125,
0.029937744140625,
0.0115509033203125,
-0.0684814453125,
0.031707763671875,
0.0183258056640625,
-0.03887939453125,
0.020294189453125,
0.016204833984375,
0.0176849365234375,
0.05340576171875,
0.078369140625,
0.011322021484375,
0.018707275390625,
-0.0172576904296875,
0.07928466796875,
-0.060638427734375,
-0.0291748046875,
-0.053070068359375,
0.039703369140625,
-0.02099609375,
-0.038848876953125,
0.052734375,
0.06231689453125,
0.06719970703125,
-0.0035839080810546875,
0.07257080078125,
-0.0244598388671875,
0.03704833984375,
-0.027435302734375,
0.045501708984375,
-0.045318603515625,
0.0134124755859375,
-0.010345458984375,
-0.040008544921875,
-0.00864410400390625,
0.06072998046875,
-0.0128631591796875,
-0.004802703857421875,
0.048004150390625,
0.0469970703125,
-0.00027751922607421875,
0.0082550048828125,
-0.00002968311309814453,
0.024139404296875,
0.027801513671875,
0.06219482421875,
0.04925537109375,
-0.0714111328125,
0.05413818359375,
-0.058929443359375,
-0.01100921630859375,
-0.0263519287109375,
-0.049957275390625,
-0.06512451171875,
-0.0215911865234375,
-0.02056884765625,
-0.0254669189453125,
-0.01446533203125,
0.05523681640625,
0.04278564453125,
-0.05706787109375,
-0.0298919677734375,
0.0023021697998046875,
0.005767822265625,
-0.03228759765625,
-0.020843505859375,
0.04974365234375,
0.0017032623291015625,
-0.050872802734375,
0.03045654296875,
-0.0002256631851196289,
0.01039886474609375,
-0.0016508102416992188,
-0.0201568603515625,
-0.01776123046875,
-0.0199737548828125,
0.0224456787109375,
0.0244140625,
-0.049774169921875,
-0.009246826171875,
-0.0188140869140625,
0.005107879638671875,
0.024749755859375,
0.0204925537109375,
-0.044219970703125,
0.0161895751953125,
0.036529541015625,
0.0192108154296875,
0.0546875,
-0.005462646484375,
-0.004863739013671875,
-0.035125732421875,
0.0247344970703125,
-0.00060272216796875,
0.02923583984375,
0.014984130859375,
-0.035614013671875,
0.053985595703125,
0.0391845703125,
-0.04034423828125,
-0.07879638671875,
-0.033782958984375,
-0.0938720703125,
-0.0145111083984375,
0.0771484375,
-0.00853729248046875,
-0.0545654296875,
0.01611328125,
-0.0171051025390625,
0.040557861328125,
-0.046112060546875,
0.047210693359375,
0.02398681640625,
-0.01458740234375,
-0.001953125,
-0.05279541015625,
0.021209716796875,
-0.01171112060546875,
-0.05364990234375,
-0.0038433074951171875,
-0.00029969215393066406,
0.026763916015625,
0.0222320556640625,
0.0278472900390625,
-0.00004297494888305664,
0.0012025833129882812,
0.0185699462890625,
0.01186370849609375,
-0.0242919921875,
-0.00829315185546875,
-0.0007524490356445312,
-0.0041656494140625,
-0.0193023681640625,
-0.045379638671875
]
] |
Yntec/LehinaModel | 2023-09-17T21:18:19.000Z | [
"diffusers",
"Photorealistic",
"Realism",
"Photo",
"Jehovah",
"stable-diffusion",
"stable-diffusion-diffusers",
"text-to-image",
"license:creativeml-openrail-m",
"endpoints_compatible",
"has_space",
"diffusers:StableDiffusionPipeline",
"region:us"
] | text-to-image | Yntec | null | null | Yntec/LehinaModel | 1 | 5,965 | diffusers | 2023-09-17T20:43:01 | ---
license: creativeml-openrail-m
library_name: diffusers
pipeline_tag: text-to-image
tags:
- Photorealistic
- Realism
- Photo
- Jehovah
- stable-diffusion
- stable-diffusion-diffusers
- diffusers
- text-to-image
---
Original model page: https://civitai.com/models/66043/lehinamodel-v11
Sample and prompt:
Anime fine details portrait of joyful cute little girl play school class room, bokeh. anime masterpiece by studio ghibli. 8k, sharp high quality classic anime from 1990 in style of hayao miyazaki. Wikipedia. hugging. OIL PAINTING. DOCTOR with short hair in coat BEAUTIFUL girl eyes. she has pigtails | 727 | [
[
-0.031402587890625,
-0.057647705078125,
0.0179290771484375,
-0.00547027587890625,
-0.0108642578125,
-0.0258331298828125,
0.0289306640625,
-0.05133056640625,
0.056488037109375,
0.0300445556640625,
-0.057861328125,
-0.0194244384765625,
-0.041961669921875,
-0.01441192626953125,
-0.0183258056640625,
0.038055419921875,
-0.0020389556884765625,
0.0090484619140625,
0.0023784637451171875,
0.017059326171875,
-0.03363037109375,
0.002124786376953125,
-0.061065673828125,
-0.0312347412109375,
0.04302978515625,
0.03936767578125,
0.044036865234375,
0.0227813720703125,
0.01336669921875,
0.031585693359375,
-0.0025463104248046875,
-0.0142669677734375,
-0.041961669921875,
0.0098419189453125,
-0.01367950439453125,
-0.03875732421875,
-0.06439208984375,
0.01337432861328125,
0.0333251953125,
0.050628662109375,
0.0017271041870117188,
0.00264739990234375,
0.01418304443359375,
0.0484619140625,
-0.01812744140625,
0.00357818603515625,
0.0152435302734375,
0.005527496337890625,
-0.0157470703125,
0.0230255126953125,
-0.020782470703125,
-0.0182037353515625,
0.006603240966796875,
-0.07159423828125,
0.016204833984375,
-0.031341552734375,
0.10516357421875,
-0.0043792724609375,
-0.038848876953125,
-0.00513458251953125,
-0.01194000244140625,
0.0433349609375,
-0.034637451171875,
0.01175689697265625,
0.038330078125,
0.04937744140625,
-0.0234375,
-0.046356201171875,
-0.00865936279296875,
0.0186004638671875,
-0.01451873779296875,
0.01776123046875,
-0.0230712890625,
-0.015716552734375,
0.034698486328125,
0.0494384765625,
-0.045867919921875,
0.004802703857421875,
-0.039154052734375,
-0.0006279945373535156,
0.06353759765625,
0.00765228271484375,
0.046966552734375,
-0.0004029273986816406,
-0.005191802978515625,
-0.01491546630859375,
-0.03582763671875,
0.010833740234375,
0.048370361328125,
-0.019287109375,
-0.03472900390625,
0.047393798828125,
-0.031951904296875,
0.048553466796875,
0.0124053955078125,
-0.002376556396484375,
0.009124755859375,
0.00899505615234375,
-0.0249786376953125,
-0.003345489501953125,
0.052001953125,
0.0648193359375,
0.0194244384765625,
0.028533935546875,
0.02459716796875,
0.014434814453125,
0.014434814453125,
-0.11962890625,
-0.00675201416015625,
0.0006465911865234375,
-0.05609130859375,
-0.042572021484375,
-0.0030536651611328125,
-0.0794677734375,
-0.005733489990234375,
-0.00530242919921875,
0.0169677734375,
-0.03460693359375,
-0.0241241455078125,
-0.01178741455078125,
0.0015773773193359375,
0.00537109375,
0.043426513671875,
-0.07781982421875,
-0.0014514923095703125,
0.0300445556640625,
0.055206298828125,
0.01161956787109375,
0.004238128662109375,
-0.006641387939453125,
0.008636474609375,
-0.049560546875,
0.0292816162109375,
-0.02886962890625,
-0.038543701171875,
-0.005924224853515625,
0.044647216796875,
0.0104827880859375,
-0.047454833984375,
0.03204345703125,
-0.0258331298828125,
0.007232666015625,
-0.0009984970092773438,
-0.041015625,
-0.03399658203125,
-0.00826263427734375,
-0.05419921875,
0.0247955322265625,
0.0312347412109375,
-0.052032470703125,
0.0107879638671875,
-0.02056884765625,
-0.005146026611328125,
0.0027294158935546875,
0.002716064453125,
-0.0160064697265625,
0.03985595703125,
-0.01523590087890625,
0.0223388671875,
0.014434814453125,
-0.00489044189453125,
-0.049957275390625,
-0.03240966796875,
0.002941131591796875,
-0.01537322998046875,
0.08026123046875,
0.03228759765625,
-0.002574920654296875,
0.001537322998046875,
-0.055206298828125,
0.00740814208984375,
0.034698486328125,
0.0130615234375,
-0.03851318359375,
-0.03704833984375,
-0.006374359130859375,
0.0192108154296875,
0.042236328125,
-0.032684326171875,
0.03314208984375,
-0.0203399658203125,
0.0154876708984375,
0.0200653076171875,
0.0011758804321289062,
-0.000028967857360839844,
-0.057952880859375,
0.046295166015625,
-0.00812530517578125,
0.028076171875,
-0.01406097412109375,
-0.04827880859375,
-0.0673828125,
-0.0297698974609375,
0.01800537109375,
0.01317596435546875,
-0.01071929931640625,
0.0311431884765625,
0.0166015625,
-0.07623291015625,
-0.07080078125,
-0.005096435546875,
0.01166534423828125,
0.03753662109375,
-0.0008635520935058594,
-0.047607421875,
-0.044189453125,
-0.09173583984375,
0.007099151611328125,
0.00565338134765625,
-0.0079345703125,
0.02197265625,
0.028839111328125,
-0.03656005859375,
0.022430419921875,
-0.02972412109375,
-0.017303466796875,
-0.002227783203125,
-0.00789642333984375,
0.06365966796875,
0.035797119140625,
0.0675048828125,
-0.041717529296875,
-0.051727294921875,
0.0084228515625,
-0.0716552734375,
-0.00948333740234375,
0.0557861328125,
-0.045562744140625,
0.00897979736328125,
0.0241241455078125,
-0.05780029296875,
0.0751953125,
0.030609130859375,
-0.043426513671875,
0.038604736328125,
-0.0249176025390625,
0.05133056640625,
-0.07843017578125,
0.0193634033203125,
-0.0028171539306640625,
-0.02264404296875,
-0.0292816162109375,
0.059661865234375,
-0.0277252197265625,
-0.005893707275390625,
-0.042022705078125,
0.050567626953125,
-0.0284576416015625,
0.0160369873046875,
-0.044158935546875,
-0.0071868896484375,
0.02081298828125,
0.00208282470703125,
0.019927978515625,
0.010589599609375,
0.05975341796875,
-0.01056671142578125,
0.0303802490234375,
0.0435791015625,
-0.0211334228515625,
0.07794189453125,
-0.06939697265625,
0.018096923828125,
0.0058441162109375,
-0.00859832763671875,
-0.06488037109375,
-0.07305908203125,
0.029937744140625,
-0.0179443359375,
0.0258331298828125,
-0.022735595703125,
-0.035003662109375,
-0.0298309326171875,
-0.02374267578125,
0.0377197265625,
0.05572509765625,
-0.033355712890625,
0.01172637939453125,
0.0080718994140625,
-0.0290985107421875,
-0.0162200927734375,
-0.03173828125,
-0.0272979736328125,
-0.01456451416015625,
-0.038116455078125,
0.03564453125,
-0.0033779144287109375,
-0.035614013671875,
-0.00554656982421875,
-0.000274658203125,
-0.0295257568359375,
-0.0178985595703125,
0.046356201171875,
0.045440673828125,
-0.02996826171875,
-0.0447998046875,
0.0037975311279296875,
-0.0039043426513671875,
-0.01386260986328125,
0.0577392578125,
0.057952880859375,
-0.0262603759765625,
-0.0206146240234375,
-0.09735107421875,
0.03717041015625,
0.0595703125,
0.0472412109375,
0.0242767333984375,
0.029541015625,
-0.0440673828125,
0.01239776611328125,
-0.019683837890625,
-0.03143310546875,
-0.03704833984375,
0.00009387731552124023,
-0.02447509765625,
-0.02972412109375,
0.027191162109375,
0.0338134765625,
-0.033294677734375,
0.053070068359375,
0.0298309326171875,
-0.031829833984375,
0.08062744140625,
0.047576904296875,
-0.0181121826171875,
0.0208892822265625,
-0.052734375,
-0.0159149169921875,
-0.031402587890625,
-0.0022125244140625,
-0.0092926025390625,
-0.0653076171875,
-0.0484619140625,
-0.0108489990234375,
-0.0092010498046875,
0.03515625,
-0.0270538330078125,
0.028594970703125,
-0.0254364013671875,
0.02960205078125,
0.029571533203125,
0.0166473388671875,
0.04107666015625,
0.003940582275390625,
0.0379638671875,
-0.0245513916015625,
-0.0112457275390625,
-0.03564453125,
0.0299835205078125,
0.023223876953125,
0.049041748046875,
0.0302734375,
0.057647705078125,
-0.02166748046875,
0.0177001953125,
-0.0572509765625,
0.056549072265625,
-0.0238800048828125,
-0.0775146484375,
-0.0194244384765625,
-0.0149383544921875,
-0.0716552734375,
0.017974853515625,
0.004779815673828125,
-0.054534912109375,
0.024139404296875,
0.0099029541015625,
-0.018402099609375,
-0.0010137557983398438,
-0.044464111328125,
0.07513427734375,
0.006984710693359375,
-0.048797607421875,
-0.034423828125,
-0.0265655517578125,
0.043975830078125,
0.01482391357421875,
0.01380157470703125,
-0.00939178466796875,
0.01519012451171875,
0.0255279541015625,
-0.0682373046875,
0.07611083984375,
-0.0150909423828125,
-0.007335662841796875,
0.0248260498046875,
0.0230865478515625,
-0.0007824897766113281,
0.0177459716796875,
0.005191802978515625,
-0.0007619857788085938,
0.0204315185546875,
-0.0234527587890625,
-0.02001953125,
0.07684326171875,
-0.060089111328125,
-0.025054931640625,
-0.0311737060546875,
-0.0165557861328125,
0.0018110275268554688,
0.02734375,
0.07562255859375,
0.03985595703125,
-0.01398468017578125,
0.0181121826171875,
0.020233154296875,
-0.00968170166015625,
0.05145263671875,
0.0033588409423828125,
-0.06121826171875,
-0.0197296142578125,
0.04791259765625,
0.023040771484375,
0.0167694091796875,
-0.01019287109375,
0.02545166015625,
0.00010597705841064453,
0.0112152099609375,
-0.0232086181640625,
0.045867919921875,
-0.01629638671875,
0.000023245811462402344,
-0.008087158203125,
-0.0279998779296875,
-0.00923919677734375,
-0.0221710205078125,
-0.03173828125,
-0.0151214599609375,
-0.049285888671875,
0.01280975341796875,
0.0435791015625,
0.06268310546875,
0.00475311279296875,
0.029144287109375,
-0.036102294921875,
0.02606201171875,
0.057342529296875,
0.03424072265625,
0.018218994140625,
-0.048828125,
0.01064300537109375,
-0.00342559814453125,
-0.059356689453125,
-0.055389404296875,
0.049652099609375,
0.0238800048828125,
0.0282440185546875,
0.049591064453125,
-0.03167724609375,
0.0516357421875,
-0.03240966796875,
0.03619384765625,
0.0274200439453125,
-0.032806396484375,
0.05572509765625,
-0.05499267578125,
0.012664794921875,
0.046966552734375,
0.03131103515625,
-0.0267181396484375,
0.001285552978515625,
-0.0804443359375,
-0.055084228515625,
0.034515380859375,
-0.006626129150390625,
0.01061248779296875,
0.03656005859375,
0.031036376953125,
0.031951904296875,
0.03997802734375,
-0.02850341796875,
-0.0264892578125,
-0.034454345703125,
-0.027313232421875,
-0.00592041015625,
-0.01482391357421875,
0.002223968505859375,
-0.050079345703125,
0.053497314453125,
-0.0091552734375,
0.00968170166015625,
0.00591278076171875,
-0.0015249252319335938,
-0.028167724609375,
-0.0031414031982421875,
0.0330810546875,
0.0163726806640625,
-0.036834716796875,
-0.010955810546875,
-0.018280029296875,
-0.04736328125,
0.03302001953125,
-0.01629638671875,
-0.041839599609375,
0.0297393798828125,
0.041534423828125,
0.0645751953125,
0.0157470703125,
-0.0311737060546875,
0.050323486328125,
0.010467529296875,
-0.000591278076171875,
-0.01274871826171875,
0.038543701171875,
0.0244293212890625,
0.03814697265625,
-0.000278472900390625,
-0.01447296142578125,
0.039825439453125,
-0.06005859375,
0.01309967041015625,
0.01153564453125,
-0.040130615234375,
-0.051177978515625,
0.046844482421875,
-0.033294677734375,
-0.018829345703125,
0.038330078125,
-0.005191802978515625,
0.0012264251708984375,
0.04791259765625,
0.040435791015625,
0.050262451171875,
-0.037139892578125,
0.047576904296875,
0.052398681640625,
-0.031494140625,
-0.00261688232421875,
0.04718017578125,
0.0279541015625,
-0.02069091796875,
-0.0096435546875,
-0.04638671875,
-0.042388916015625,
0.03411865234375,
-0.0472412109375,
0.03961181640625,
-0.0518798828125,
-0.0092620849609375,
-0.006816864013671875,
0.005443572998046875,
-0.038543701171875,
0.031280517578125,
0.01114654541015625,
0.0794677734375,
-0.049591064453125,
0.0526123046875,
0.050445556640625,
-0.034088134765625,
-0.055206298828125,
-0.0189971923828125,
0.01042938232421875,
-0.045257568359375,
0.0322265625,
0.049835205078125,
0.00783538818359375,
-0.01171875,
-0.05560302734375,
-0.0237884521484375,
0.06402587890625,
0.034637451171875,
-0.0794677734375,
-0.007160186767578125,
-0.036102294921875,
0.04437255859375,
-0.03521728515625,
0.0252532958984375,
0.044769287109375,
0.024017333984375,
0.0179443359375,
-0.046630859375,
-0.03302001953125,
-0.05694580078125,
0.01456451416015625,
-0.01464080810546875,
-0.0804443359375,
0.061981201171875,
-0.0132598876953125,
0.00815582275390625,
0.06109619140625,
0.065185546875,
0.04400634765625,
0.0195770263671875,
0.047637939453125,
0.0653076171875,
0.0045928955078125,
-0.0162200927734375,
0.06536865234375,
0.00646209716796875,
0.0159454345703125,
0.06829833984375,
0.002376556396484375,
0.06671142578125,
0.0264434814453125,
-0.019866943359375,
0.045074462890625,
0.0809326171875,
-0.0211334228515625,
0.06439208984375,
0.01111602783203125,
-0.0102081298828125,
-0.006702423095703125,
-0.02874755859375,
-0.03363037109375,
0.0193939208984375,
0.00308990478515625,
-0.01454925537109375,
-0.0118255615234375,
0.00518035888671875,
-0.0026264190673828125,
0.0025386810302734375,
-0.0460205078125,
0.041839599609375,
0.03204345703125,
-0.0215911865234375,
0.034637451171875,
-0.0399169921875,
0.043975830078125,
-0.034576416015625,
-0.01580810546875,
-0.01605224609375,
0.002651214599609375,
-0.0238189697265625,
-0.051727294921875,
0.0222320556640625,
-0.0240325927734375,
-0.01294708251953125,
-0.011138916015625,
0.05755615234375,
-0.023193359375,
-0.047271728515625,
0.0296478271484375,
-0.0037212371826171875,
0.03582763671875,
0.0194854736328125,
-0.0345458984375,
0.0127105712890625,
-0.01226043701171875,
-0.026092529296875,
0.019378662109375,
0.028839111328125,
0.004730224609375,
0.01043701171875,
-0.004764556884765625,
0.028656005859375,
-0.0251312255859375,
0.0196990966796875,
0.007808685302734375,
-0.0236358642578125,
-0.03387451171875,
-0.05023193359375,
0.04815673828125,
-0.02874755859375,
-0.048065185546875,
0.042266845703125,
0.0297698974609375,
0.0518798828125,
-0.055328369140625,
0.035247802734375,
-0.0092620849609375,
0.038421630859375,
-0.0261993408203125,
0.0736083984375,
-0.07958984375,
-0.054290771484375,
-0.0238189697265625,
-0.056610107421875,
0.004505157470703125,
0.08074951171875,
0.012237548828125,
0.00994110107421875,
0.0069580078125,
0.04815673828125,
-0.010986328125,
0.00787353515625,
0.0235748291015625,
0.0279998779296875,
-0.0013418197631835938,
0.0237274169921875,
0.0806884765625,
-0.050323486328125,
-0.0352783203125,
-0.04534912109375,
-0.02886962890625,
-0.07568359375,
-0.054534912109375,
-0.072998046875,
-0.036346435546875,
-0.039459228515625,
-0.0216827392578125,
-0.02777099609375,
0.0435791015625,
0.06646728515625,
-0.069091796875,
-0.0244293212890625,
0.0151214599609375,
-0.01181793212890625,
-0.006229400634765625,
-0.0194244384765625,
0.014984130859375,
0.040008544921875,
-0.0850830078125,
0.0216827392578125,
0.0074310302734375,
0.055267333984375,
-0.0183563232421875,
0.004467010498046875,
0.0234375,
-0.0099639892578125,
0.0159454345703125,
0.01300811767578125,
-0.03863525390625,
-0.033416748046875,
-0.006450653076171875,
-0.01800537109375,
0.001567840576171875,
0.02069091796875,
0.0021820068359375,
0.01523590087890625,
0.044891357421875,
0.004730224609375,
0.036956787109375,
-0.0173492431640625,
0.0237274169921875,
-0.016876220703125,
0.0296478271484375,
-0.010467529296875,
0.0513916015625,
0.02337646484375,
-0.030670166015625,
0.04931640625,
0.050262451171875,
-0.04315185546875,
-0.06842041015625,
0.0167999267578125,
-0.09381103515625,
-0.01427459716796875,
0.033782958984375,
0.01056671142578125,
-0.04449462890625,
0.04534912109375,
-0.058197021484375,
0.015899658203125,
-0.012664794921875,
0.03192138671875,
0.0628662109375,
-0.00543212890625,
-0.026519775390625,
-0.0684814453125,
0.0100555419921875,
0.00482177734375,
-0.045562744140625,
-0.022613525390625,
0.033599853515625,
0.0303802490234375,
0.031585693359375,
0.025665283203125,
-0.03314208984375,
0.0163421630859375,
-0.003314971923828125,
0.0245208740234375,
0.0311737060546875,
-0.034393310546875,
0.002590179443359375,
0.018310546875,
-0.004390716552734375,
-0.021331787109375
]
] |
Gryphe/MythoMix-L2-13b | 2023-08-11T05:44:46.000Z | [
"transformers",
"pytorch",
"llama",
"text-generation",
"en",
"license:other",
"endpoints_compatible",
"has_space",
"text-generation-inference",
"region:us"
] | text-generation | Gryphe | null | null | Gryphe/MythoMix-L2-13b | 16 | 5,964 | transformers | 2023-08-08T15:01:01 | ---
license: other
language:
- en
---
**UPDATE:** There's an improved version now! [Check it MythoMax!](https://huggingface.co/Gryphe/MythoMax-L2-13b)
A requested variant of [MythoLogic-L2](https://huggingface.co/Gryphe/MythoLogic-L2-13b) and [Huginn](https://huggingface.co/The-Face-Of-Goonery/Huginn-13b-FP16) using a highly experimental tensor type merge technique.
This model is proficient at both roleplaying and storywriting due to its unique nature.
Quantized models are available from TheBloke: [GGML](https://huggingface.co/TheBloke/MythoMix-L2-13B-GGML) - [GPTQ](https://huggingface.co/TheBloke/MythoMix-L2-13B-GPTQ) (You're the best!)
## Model details
The idea behind this merge is that each layer is composed of several tensors, which are in turn responsible for specific functions. Using MythoLogic-L2's robust understanding as its input and Huginn's extensive writing capability as its output seems to have resulted in a model that exceeds at both, confirming my theory. (More details to be released at a later time)
This type of merge is incapable of being illustrated, as each of its 360 tensors has an unique ratio applied to it. As with my prior merges, gradients were part of these ratios to further finetune its behaviour.
## Prompt Format
This model primarily uses Alpaca formatting, so for optimal model performance, use:
```
<System prompt/Character Card>
### Instruction:
Your instruction or question here.
For roleplay purposes, I suggest the following - Write <CHAR NAME>'s next reply in a chat between <YOUR NAME> and <CHAR NAME>. Write a single reply only.
### Response:
```
---
license: other
--- | 1,638 | [
[
-0.0218505859375,
-0.052947998046875,
0.021026611328125,
0.0222015380859375,
-0.0221405029296875,
0.005565643310546875,
-0.003246307373046875,
-0.058685302734375,
0.03759765625,
0.050384521484375,
-0.047607421875,
-0.0214080810546875,
-0.03936767578125,
-0.01236724853515625,
-0.007556915283203125,
0.09222412109375,
-0.01806640625,
0.0172119140625,
-0.023193359375,
-0.0236968994140625,
-0.0200042724609375,
-0.036376953125,
-0.08441162109375,
-0.04010009765625,
0.051422119140625,
-0.009857177734375,
0.07122802734375,
0.048919677734375,
0.051422119140625,
0.0241546630859375,
-0.0175933837890625,
0.027679443359375,
-0.05145263671875,
0.007678985595703125,
0.0142364501953125,
-0.043548583984375,
-0.07281494140625,
0.01261138916015625,
0.056732177734375,
0.03729248046875,
-0.0264434814453125,
0.0190277099609375,
0.004009246826171875,
0.03216552734375,
-0.0233154296875,
0.0078277587890625,
0.00890350341796875,
0.0201416015625,
-0.005886077880859375,
0.00707244873046875,
-0.01183319091796875,
-0.05047607421875,
0.007659912109375,
-0.0712890625,
0.0118560791015625,
0.0144805908203125,
0.054779052734375,
0.0181732177734375,
-0.0257415771484375,
-0.00652313232421875,
-0.04132080078125,
0.04571533203125,
-0.05340576171875,
0.02032470703125,
0.033233642578125,
0.0302276611328125,
0.0028285980224609375,
-0.072998046875,
-0.0270843505859375,
-0.03076171875,
-0.02294921875,
0.0173187255859375,
-0.037261962890625,
-0.005298614501953125,
0.034759521484375,
0.042236328125,
-0.05181884765625,
0.0103607177734375,
-0.051605224609375,
-0.0269317626953125,
0.042816162109375,
0.0220947265625,
0.018829345703125,
-0.0138092041015625,
-0.049072265625,
-0.0249786376953125,
-0.03656005859375,
0.0035343170166015625,
0.0450439453125,
-0.0145263671875,
-0.02960205078125,
0.06292724609375,
0.00595855712890625,
0.03814697265625,
0.019683837890625,
-0.00865936279296875,
0.0093841552734375,
-0.010284423828125,
-0.022369384765625,
-0.0195770263671875,
0.05712890625,
0.046112060546875,
-0.00904083251953125,
0.0043792724609375,
-0.01416015625,
0.005615234375,
0.0133209228515625,
-0.06463623046875,
-0.009033203125,
0.0235443115234375,
-0.043731689453125,
-0.05316162109375,
-0.034088134765625,
-0.0416259765625,
-0.04302978515625,
-0.0123291015625,
0.04248046875,
-0.058746337890625,
-0.0428466796875,
0.0084381103515625,
-0.0136260986328125,
0.01129150390625,
0.030029296875,
-0.04833984375,
0.028289794921875,
0.042572021484375,
0.06292724609375,
-0.0041961669921875,
-0.022491455078125,
-0.0303802490234375,
-0.0095367431640625,
-0.0447998046875,
0.039703369140625,
-0.02032470703125,
-0.0274505615234375,
-0.00971221923828125,
-0.029510498046875,
0.007747650146484375,
-0.034942626953125,
0.05810546875,
-0.04486083984375,
0.052734375,
-0.0205535888671875,
-0.05126953125,
-0.01184844970703125,
0.00962066650390625,
-0.0784912109375,
0.0701904296875,
0.0416259765625,
-0.06005859375,
0.004421234130859375,
-0.04705810546875,
0.00243377685546875,
0.00936126708984375,
0.0134124755859375,
-0.0302276611328125,
0.007556915283203125,
-0.005126953125,
0.0164642333984375,
-0.03619384765625,
0.019989013671875,
-0.0404052734375,
-0.0206298828125,
0.0288543701171875,
-0.01751708984375,
0.058807373046875,
0.0287628173828125,
-0.00406646728515625,
-0.00052642822265625,
-0.049468994140625,
0.00595855712890625,
0.0114593505859375,
-0.00955963134765625,
-0.005924224853515625,
-0.0411376953125,
0.0248260498046875,
0.0073089599609375,
0.03216552734375,
-0.00726318359375,
0.039337158203125,
0.0109710693359375,
0.031402587890625,
0.0338134765625,
-0.0118408203125,
0.04400634765625,
-0.054779052734375,
0.052337646484375,
0.004291534423828125,
0.0262451171875,
0.005279541015625,
-0.03173828125,
-0.05926513671875,
-0.048370361328125,
0.035858154296875,
0.039581298828125,
-0.06683349609375,
0.003025054931640625,
0.01654052734375,
-0.059234619140625,
-0.0452880859375,
-0.0270233154296875,
0.052001953125,
0.0347900390625,
0.030517578125,
-0.030487060546875,
-0.037994384765625,
-0.07476806640625,
-0.016082763671875,
-0.038238525390625,
-0.003131866455078125,
0.0306549072265625,
-0.0052490234375,
-0.0257720947265625,
0.052520751953125,
-0.052734375,
-0.01287078857421875,
-0.013336181640625,
0.033172607421875,
0.015716552734375,
0.044525146484375,
0.060333251953125,
-0.041168212890625,
-0.0103759765625,
-0.0031070709228515625,
-0.07537841796875,
-0.0020084381103515625,
0.004825592041015625,
-0.0220489501953125,
0.01055908203125,
0.0169525146484375,
-0.0748291015625,
0.0200042724609375,
0.026123046875,
-0.040130615234375,
0.041900634765625,
-0.02587890625,
0.0253143310546875,
-0.0966796875,
0.01317596435546875,
0.0184783935546875,
-0.024658203125,
-0.057464599609375,
0.035980224609375,
0.008270263671875,
-0.003997802734375,
-0.02166748046875,
0.053314208984375,
-0.0287322998046875,
0.005184173583984375,
-0.003082275390625,
-0.02392578125,
-0.0160675048828125,
0.036376953125,
0.004306793212890625,
0.0292205810546875,
0.058685302734375,
-0.02294921875,
0.054290771484375,
0.034576416015625,
-0.0115203857421875,
0.039947509765625,
-0.061981201171875,
0.01061248779296875,
-0.0247344970703125,
0.04473876953125,
-0.060333251953125,
-0.030303955078125,
0.03753662109375,
-0.0217742919921875,
0.033111572265625,
-0.020233154296875,
-0.0288543701171875,
-0.037017822265625,
-0.0309600830078125,
0.05126953125,
0.06561279296875,
-0.034912109375,
0.054229736328125,
0.0013217926025390625,
-0.005970001220703125,
-0.051361083984375,
-0.047637939453125,
-0.005451202392578125,
-0.04400634765625,
-0.04962158203125,
0.0240631103515625,
-0.02691650390625,
-0.005130767822265625,
-0.0017309188842773438,
0.0042266845703125,
-0.01885986328125,
-0.0275421142578125,
0.004955291748046875,
0.055877685546875,
-0.032379150390625,
-0.0281982421875,
0.0185089111328125,
-0.00348663330078125,
-0.0070648193359375,
-0.004917144775390625,
0.0390625,
-0.01513671875,
0.0217742919921875,
-0.03826904296875,
0.027618408203125,
0.060150146484375,
-0.0112152099609375,
0.038482666015625,
0.07452392578125,
-0.044342041015625,
0.007572174072265625,
-0.05792236328125,
-0.0162811279296875,
-0.034423828125,
0.016876220703125,
-0.02386474609375,
-0.08245849609375,
0.0552978515625,
0.01393890380859375,
0.0001386404037475586,
0.04205322265625,
0.0229949951171875,
-0.0137939453125,
0.061676025390625,
0.044677734375,
0.00968170166015625,
0.0153045654296875,
-0.033905029296875,
0.004283905029296875,
-0.062744140625,
-0.025909423828125,
0.0017547607421875,
-0.02276611328125,
-0.06378173828125,
-0.04833984375,
0.03656005859375,
0.0283660888671875,
-0.0024051666259765625,
0.044677734375,
-0.01218414306640625,
0.028076171875,
0.03424072265625,
0.0215911865234375,
0.0214385986328125,
0.0218505859375,
0.01519775390625,
-0.0156097412109375,
-0.047027587890625,
-0.0226287841796875,
0.0706787109375,
0.053070068359375,
0.059417724609375,
0.0305938720703125,
0.05694580078125,
0.0130462646484375,
0.018768310546875,
-0.039215087890625,
0.03857421875,
-0.013275146484375,
-0.0435791015625,
-0.0163116455078125,
-0.04150390625,
-0.031280517578125,
0.0222625732421875,
-0.0361328125,
-0.0372314453125,
0.0223388671875,
0.033233642578125,
-0.045989990234375,
0.0101776123046875,
-0.054840087890625,
0.045989990234375,
0.00638580322265625,
-0.04132080078125,
-0.0251617431640625,
-0.04132080078125,
0.05450439453125,
-0.0064697265625,
-0.01433563232421875,
0.020965576171875,
0.001220703125,
0.046966552734375,
-0.0323486328125,
0.05010986328125,
0.0003190040588378906,
-0.017730712890625,
0.026763916015625,
0.0276947021484375,
0.034454345703125,
0.0239715576171875,
0.01059722900390625,
0.01346588134765625,
0.00913238525390625,
-0.0190887451171875,
-0.059814453125,
0.05364990234375,
-0.072509765625,
-0.034637451171875,
-0.039520263671875,
-0.041351318359375,
-0.0115203857421875,
-0.0059661865234375,
0.044921875,
0.066650390625,
-0.024749755859375,
-0.004150390625,
0.055816650390625,
-0.0009179115295410156,
0.0276031494140625,
0.03125,
-0.03662109375,
-0.044342041015625,
0.053955078125,
-0.002288818359375,
0.01036834716796875,
0.022430419921875,
0.0131378173828125,
-0.0350341796875,
-0.00806427001953125,
-0.040771484375,
0.048675537109375,
-0.029449462890625,
-0.0229644775390625,
-0.050323486328125,
-0.033782958984375,
-0.0272216796875,
0.001468658447265625,
-0.026031494140625,
-0.06597900390625,
-0.023284912109375,
-0.00307464599609375,
0.042236328125,
0.054840087890625,
-0.0308837890625,
0.0164642333984375,
-0.03936767578125,
0.02764892578125,
0.026275634765625,
0.03265380859375,
-0.0145263671875,
-0.051666259765625,
0.004711151123046875,
-0.0041046142578125,
-0.01216888427734375,
-0.054168701171875,
0.0330810546875,
-0.00604248046875,
0.0125732421875,
0.0199127197265625,
-0.01506805419921875,
0.046630859375,
-0.03924560546875,
0.054046630859375,
0.047943115234375,
-0.04827880859375,
0.01324462890625,
-0.0360107421875,
-0.00479888916015625,
0.0159454345703125,
0.044342041015625,
-0.038787841796875,
-0.028167724609375,
-0.0631103515625,
-0.047088623046875,
0.06756591796875,
0.04180908203125,
-0.00201416015625,
0.0281982421875,
0.03863525390625,
0.0186309814453125,
0.01428985595703125,
-0.05230712890625,
-0.039886474609375,
-0.031280517578125,
-0.0005583763122558594,
-0.00867462158203125,
-0.042816162109375,
-0.02740478515625,
-0.0169677734375,
0.04827880859375,
0.0022735595703125,
0.040191650390625,
0.00493621826171875,
0.01479339599609375,
-0.016021728515625,
-0.00955963134765625,
0.055938720703125,
0.0350341796875,
-0.032501220703125,
0.0052032470703125,
0.006015777587890625,
-0.04754638671875,
-0.013671875,
0.03192138671875,
-0.0014886856079101562,
0.0013875961303710938,
0.03662109375,
0.0927734375,
0.0011081695556640625,
-0.0205535888671875,
0.0196685791015625,
-0.0243377685546875,
-0.024566650390625,
-0.01129150390625,
0.0155792236328125,
0.0026607513427734375,
0.047088623046875,
0.0107574462890625,
0.0262451171875,
0.02099609375,
-0.041015625,
0.0004169940948486328,
0.016387939453125,
-0.0216827392578125,
-0.01226043701171875,
0.0592041015625,
0.0037593841552734375,
-0.0225372314453125,
0.035003662109375,
-0.018341064453125,
-0.02093505859375,
0.04638671875,
0.04925537109375,
0.058349609375,
-0.0278778076171875,
0.04302978515625,
0.045989990234375,
0.0096435546875,
-0.0196075439453125,
0.0128173828125,
0.00797271728515625,
-0.033843994140625,
-0.0192413330078125,
-0.036956787109375,
-0.026123046875,
0.0206298828125,
-0.058563232421875,
0.02325439453125,
-0.03826904296875,
-0.0250244140625,
0.00907135009765625,
-0.000014007091522216797,
-0.041656494140625,
0.027679443359375,
0.0039520263671875,
0.059478759765625,
-0.0523681640625,
0.048797607421875,
0.05230712890625,
-0.02978515625,
-0.06866455078125,
-0.0182342529296875,
-0.01433563232421875,
-0.05908203125,
0.019561767578125,
-0.0172576904296875,
0.0178985595703125,
-0.005786895751953125,
-0.04571533203125,
-0.0811767578125,
0.11090087890625,
0.040191650390625,
-0.037200927734375,
-0.01114654541015625,
0.00638580322265625,
0.06005859375,
-0.03369140625,
0.05279541015625,
0.038482666015625,
0.0182647705078125,
0.025634765625,
-0.072021484375,
0.0025844573974609375,
-0.0304107666015625,
0.021026611328125,
0.00281524658203125,
-0.0765380859375,
0.06890869140625,
-0.01629638671875,
0.00765228271484375,
0.036102294921875,
0.0660400390625,
0.0364990234375,
0.0238800048828125,
0.04205322265625,
0.05908203125,
0.060150146484375,
-0.0059814453125,
0.0877685546875,
-0.008331298828125,
0.04205322265625,
0.054779052734375,
-0.01079559326171875,
0.037445068359375,
0.0175018310546875,
-0.004512786865234375,
0.0203857421875,
0.06072998046875,
-0.00506591796875,
0.03271484375,
0.0067138671875,
-0.022857666015625,
-0.0170745849609375,
-0.007579803466796875,
-0.05255126953125,
0.023223876953125,
-0.00627899169921875,
-0.0103607177734375,
0.0030460357666015625,
-0.0115966796875,
0.022705078125,
-0.01509857177734375,
-0.01447296142578125,
0.028839111328125,
0.00392913818359375,
-0.0577392578125,
0.0458984375,
0.02178955078125,
0.0556640625,
-0.055023193359375,
0.0024871826171875,
-0.054107666015625,
-0.0010662078857421875,
-0.00974273681640625,
-0.05218505859375,
-0.0019273757934570312,
-0.004482269287109375,
-0.02252197265625,
-0.0109710693359375,
0.04815673828125,
-0.070068359375,
-0.044097900390625,
0.006877899169921875,
0.0302886962890625,
0.005535125732421875,
0.0311279296875,
-0.046234130859375,
0.02099609375,
-0.008819580078125,
-0.007129669189453125,
0.01503753662109375,
0.0102386474609375,
-0.01152801513671875,
0.043243408203125,
0.0226287841796875,
-0.004955291748046875,
-0.00910186767578125,
0.00785064697265625,
0.06524658203125,
-0.01523590087890625,
-0.041107177734375,
-0.032470703125,
0.0401611328125,
-0.01505279541015625,
-0.051177978515625,
0.02569580078125,
0.052978515625,
0.034881591796875,
-0.031707763671875,
0.04150390625,
0.000637054443359375,
0.02044677734375,
-0.046417236328125,
0.0675048828125,
-0.0526123046875,
-0.01013946533203125,
-0.01139068603515625,
-0.08978271484375,
0.0020236968994140625,
0.042236328125,
0.0221099853515625,
0.0087127685546875,
0.06884765625,
0.04974365234375,
-0.014984130859375,
0.0093536376953125,
0.0266571044921875,
0.0209197998046875,
0.00231170654296875,
0.050689697265625,
0.0718994140625,
-0.05767822265625,
0.01107025146484375,
-0.0222930908203125,
-0.04248046875,
-0.005157470703125,
-0.06072998046875,
-0.07366943359375,
-0.054046630859375,
-0.047943115234375,
-0.049774169921875,
0.0079345703125,
0.059814453125,
0.047943115234375,
-0.0251922607421875,
-0.01788330078125,
0.019317626953125,
-0.0203399658203125,
-0.0121002197265625,
-0.0134124755859375,
0.01678466796875,
0.01322174072265625,
-0.05908203125,
0.01468658447265625,
-0.0005578994750976562,
0.0173797607421875,
-0.00811767578125,
-0.0105438232421875,
0.0299072265625,
0.018280029296875,
0.03741455078125,
0.0216064453125,
-0.048797607421875,
-0.0122833251953125,
0.007007598876953125,
-0.01324462890625,
-0.0223388671875,
0.07049560546875,
-0.037261962890625,
0.007007598876953125,
0.016632080078125,
0.02581787109375,
0.059478759765625,
-0.0181884765625,
0.041168212890625,
-0.03607177734375,
0.002899169921875,
0.00907135009765625,
0.0228271484375,
0.0273895263671875,
-0.05108642578125,
0.051605224609375,
0.0245361328125,
-0.0249481201171875,
-0.0625,
0.0248870849609375,
-0.11505126953125,
0.005859375,
0.0965576171875,
0.0103302001953125,
-0.0029659271240234375,
0.058502197265625,
-0.037200927734375,
0.030975341796875,
-0.019378662109375,
0.043243408203125,
0.054443359375,
-0.01107025146484375,
-0.0185546875,
-0.016571044921875,
0.0552978515625,
0.03961181640625,
-0.0482177734375,
-0.00786590576171875,
0.047454833984375,
0.01541900634765625,
0.00821685791015625,
0.0443115234375,
-0.0216827392578125,
0.036865234375,
-0.0158538818359375,
0.0011186599731445312,
-0.0157928466796875,
-0.01029205322265625,
-0.032318115234375,
0.0076904296875,
-0.01275634765625,
0.0015592575073242188
]
] |
CHIH-HUNG/llama-2-13b-FINETUNE1_17w-r16 | 2023-09-12T12:30:26.000Z | [
"transformers",
"pytorch",
"llama",
"text-generation",
"dataset:huangyt/FINETUNE1",
"license:llama2",
"endpoints_compatible",
"text-generation-inference",
"region:us"
] | text-generation | CHIH-HUNG | null | null | CHIH-HUNG/llama-2-13b-FINETUNE1_17w-r16 | 0 | 5,964 | transformers | 2023-09-06T09:57:47 | ---
license: llama2
datasets:
- huangyt/FINETUNE1
---
# Model Card for Model ID
<!-- Provide a quick summary of what the model is/does. -->
在llama-2-13b上使用huangyt/FINETUNE1資料集進行訓練，總資料筆數約17w
# Fine-Tuning Information
- **GPU:** RTX4090 (single core / 24564MiB)
- **model:** meta-llama/Llama-2-13b-hf
- **dataset:** huangyt/FINETUNE1 (共約17w筆訓練集)
- **peft_type:** LoRA
- **lora_rank:** 16
- **lora_target:** gate_proj, up_proj, down_proj
- **per_device_train_batch_size:** 8
- **gradient_accumulation_steps:** 8
- **learning_rate :** 5e-5
- **epoch:** 1
- **precision:** bf16
- **quantization:** load_in_4bit
# Fine-Tuning Detail
- **train_loss:** 0.66
- **train_runtime:** 16:26:58 (use deepspeed)
# Evaluation
- 評估結果來自**HuggingFaceH4/open_llm_leaderboard**
- 與Llama-2-13b比較4種Benchmark，包含**ARC**、**HellaSwag**、**MMLU**、**TruthfulQA**
| Model |Average| ARC |HellaSwag| MMLU |TruthfulQA|
|--------------------------------------------------------|-------|-------|---------|-------|----------|
|meta-llama/Llama-2-13b-hf | 56.9 | 58.11 | 80.97 | 54.34 | 34.17 |
|meta-llama/Llama-2-13b-chat-hf | 59.93 | 59.04 | 81.94 | 54.64 | 44.12 |
|CHIH-HUNG/llama-2-13b-Fintune_1_17w | 58.24 | 59.47 | 81 | 54.31 | 38.17 |
|CHIH-HUNG/llama-2-13b-huangyt_Fintune_1_17w-q_k_v_o_proj| 58.49 | 59.73 | 81.06 | 54.53 | 38.64 |
|CHIH-HUNG/llama-2-13b-Fintune_1_17w-gate_up_down_proj | 58.81 | 57.17 | 82.26 | 55.89 | 39.93 |
|CHIH-HUNG/llama-2-13b-FINETUNE1_17w-r16 | 58.86 | 57.25 | 82.27 | 56.16 | 39.75 |
|CHIH-HUNG/llama-2-13b-FINETUNE1_17w-r4 | 58.71 | 56.74 | 82.27 | 56.18 | 39.65 |
# How to convert dataset to json
- 在**load_dataset**中輸入資料集名稱，並且在**take**中輸入要取前幾筆資料
- 觀察該資料集的欄位名稱，填入**example**欄位中(例如system_prompt、question、response)
- 最後指定json檔儲存位置 (**json_filename**)
```py
import json
from datasets import load_dataset
# 讀取數據集，take可以取得該數據集前n筆資料
dataset = load_dataset("huangyt/FINETUNE1", split="train", streaming=True)
# 提取所需欄位並建立新的字典列表
extracted_data = []
for example in dataset:
extracted_example = {
"instruction": example["instruction"],
"input": example["input"],
"output": example["output"]
}
extracted_data.append(extracted_example)
# 指定 JSON 文件名稱
json_filename = "huangyt_FINETUNE_1.json"
# 寫入 JSON 文件
with open(json_filename, "w") as json_file:
json.dump(extracted_data, json_file, indent=4)
print(f"數據已提取並保存為 {json_filename}")
``` | 2,581 | [
[
-0.0455322265625,
-0.048614501953125,
0.013427734375,
0.01233673095703125,
-0.0479736328125,
0.004352569580078125,
-0.01123046875,
-0.019500732421875,
0.0186004638671875,
0.03265380859375,
-0.04498291015625,
-0.040863037109375,
-0.0428466796875,
0.01345062255859375,
-0.02008056640625,
0.082763671875,
-0.00897216796875,
-0.0126953125,
0.01947021484375,
0.0060882568359375,
-0.038055419921875,
-0.024444580078125,
-0.05242919921875,
-0.02978515625,
0.024322509765625,
0.0169677734375,
0.04949951171875,
0.07012939453125,
0.05487060546875,
0.0205078125,
-0.0131072998046875,
0.0185699462890625,
-0.046173095703125,
-0.0220184326171875,
0.01873779296875,
-0.0430908203125,
-0.047454833984375,
-0.005954742431640625,
0.0494384765625,
0.0240936279296875,
0.004032135009765625,
0.0445556640625,
0.0167236328125,
0.04443359375,
-0.0236968994140625,
0.0204315185546875,
-0.0242767333984375,
0.009033203125,
-0.025299072265625,
-0.0273895263671875,
-0.001918792724609375,
-0.0248870849609375,
-0.0106353759765625,
-0.06781005859375,
0.0057525634765625,
0.01142120361328125,
0.10467529296875,
0.03143310546875,
-0.020782470703125,
0.00726318359375,
-0.036651611328125,
0.06298828125,
-0.076904296875,
0.0002923011779785156,
0.0267181396484375,
0.02874755859375,
-0.00934600830078125,
-0.050811767578125,
-0.054473876953125,
0.00806427001953125,
-0.01378631591796875,
0.01519012451171875,
-0.00902557373046875,
-0.0196533203125,
0.026031494140625,
0.037506103515625,
-0.032257080078125,
0.003082275390625,
-0.037353515625,
0.006847381591796875,
0.0635986328125,
0.03192138671875,
0.005611419677734375,
-0.0247802734375,
-0.0222015380859375,
-0.019744873046875,
-0.039459228515625,
0.0202178955078125,
0.032135009765625,
0.031463623046875,
-0.038604736328125,
0.0347900390625,
-0.03839111328125,
0.03240966796875,
0.01096343994140625,
-0.030670166015625,
0.047760009765625,
-0.0193023681640625,
-0.04144287109375,
0.0026397705078125,
0.0775146484375,
0.044647216796875,
-0.00547027587890625,
0.0172882080078125,
-0.0088043212890625,
-0.01385498046875,
-0.004680633544921875,
-0.06719970703125,
-0.0232696533203125,
0.04010009765625,
-0.053436279296875,
-0.033905029296875,
0.00847625732421875,
-0.0645751953125,
-0.005992889404296875,
-0.008758544921875,
0.0220794677734375,
-0.0238189697265625,
-0.044647216796875,
0.0006670951843261719,
-0.0126495361328125,
0.025634765625,
0.025177001953125,
-0.058868408203125,
0.01202392578125,
0.047637939453125,
0.0538330078125,
0.00855255126953125,
-0.02435302734375,
-0.00650787353515625,
0.01396942138671875,
-0.0266876220703125,
0.04974365234375,
-0.006000518798828125,
-0.0288543701171875,
-0.01715087890625,
0.0208892822265625,
-0.003246307373046875,
-0.0384521484375,
0.058685302734375,
-0.03173828125,
-0.0067138671875,
-0.038848876953125,
-0.0197601318359375,
-0.035736083984375,
0.035614013671875,
-0.05291748046875,
0.078369140625,
0.006534576416015625,
-0.06658935546875,
0.0242462158203125,
-0.051361083984375,
-0.0142364501953125,
0.004608154296875,
0.0021209716796875,
-0.035919189453125,
-0.02130126953125,
0.0213470458984375,
0.041473388671875,
-0.0347900390625,
0.015167236328125,
-0.0173797607421875,
-0.0438232421875,
0.022186279296875,
-0.029571533203125,
0.07366943359375,
0.03125,
-0.016845703125,
0.003631591796875,
-0.07208251953125,
0.004550933837890625,
0.0460205078125,
-0.039031982421875,
-0.0046844482421875,
-0.0092926025390625,
0.002887725830078125,
-0.0029506683349609375,
0.0309906005859375,
-0.0172271728515625,
0.0267333984375,
-0.0145721435546875,
0.031829833984375,
0.0684814453125,
0.0016155242919921875,
0.0096282958984375,
-0.039276123046875,
0.0249176025390625,
0.00890350341796875,
0.01995849609375,
-0.003208160400390625,
-0.03448486328125,
-0.074462890625,
-0.02032470703125,
0.01092529296875,
0.040283203125,
-0.033416748046875,
0.052886962890625,
-0.02447509765625,
-0.053802490234375,
-0.055145263671875,
0.005176544189453125,
0.0188446044921875,
0.04083251953125,
0.03900146484375,
0.008087158203125,
-0.05377197265625,
-0.06634521484375,
0.0026874542236328125,
-0.0052337646484375,
0.007274627685546875,
0.02587890625,
0.050140380859375,
-0.0248260498046875,
0.0404052734375,
-0.038330078125,
-0.0233154296875,
-0.0250091552734375,
0.0000032782554626464844,
0.0692138671875,
0.043487548828125,
0.05059814453125,
-0.037017822265625,
-0.03369140625,
0.006214141845703125,
-0.08465576171875,
0.0120849609375,
-0.006801605224609375,
-0.0206298828125,
-0.00771331787109375,
0.0024204254150390625,
-0.047393798828125,
0.033447265625,
0.034698486328125,
-0.0172271728515625,
0.04315185546875,
0.00751495361328125,
0.025482177734375,
-0.07830810546875,
0.012969970703125,
-0.0170745849609375,
0.006153106689453125,
-0.033111572265625,
0.0151519775390625,
-0.013031005859375,
0.02227783203125,
-0.02899169921875,
0.0229949951171875,
-0.024810791015625,
0.010345458984375,
-0.01381683349609375,
-0.002643585205078125,
0.0013589859008789062,
0.048187255859375,
-0.01233673095703125,
0.047882080078125,
0.039581298828125,
-0.055877685546875,
0.04248046875,
0.03509521484375,
-0.0299072265625,
0.01476287841796875,
-0.03955078125,
0.002155303955078125,
0.005764007568359375,
0.0224456787109375,
-0.0732421875,
-0.02593994140625,
0.044708251953125,
-0.0313720703125,
0.016326904296875,
-0.0283660888671875,
-0.0274810791015625,
-0.04925537109375,
-0.030181884765625,
0.022247314453125,
0.02386474609375,
-0.0445556640625,
0.0162200927734375,
0.0103912353515625,
0.0150146484375,
-0.051849365234375,
-0.06396484375,
-0.00559234619140625,
-0.019744873046875,
-0.0360107421875,
0.01751708984375,
-0.01071929931640625,
-0.00814056396484375,
0.0052337646484375,
-0.0014476776123046875,
-0.0015363693237304688,
0.010101318359375,
0.01336669921875,
0.035797119140625,
-0.0247344970703125,
-0.029022216796875,
0.00612640380859375,
-0.008270263671875,
0.0036945343017578125,
0.01186370849609375,
0.060791015625,
-0.01708984375,
-0.016632080078125,
-0.059295654296875,
0.004695892333984375,
0.0271453857421875,
0.0041961669921875,
0.04388427734375,
0.05810546875,
-0.0179901123046875,
0.004978179931640625,
-0.019073486328125,
-0.0020275115966796875,
-0.0380859375,
0.02447509765625,
-0.043975830078125,
-0.052581787109375,
0.05255126953125,
-0.001735687255859375,
0.0191192626953125,
0.06396484375,
0.02655029296875,
-0.0161285400390625,
0.074951171875,
0.01367950439453125,
-0.019683837890625,
0.0181121826171875,
-0.07147216796875,
0.005405426025390625,
-0.0755615234375,
-0.025482177734375,
-0.03662109375,
-0.044647216796875,
-0.048248291015625,
-0.013458251953125,
0.0169219970703125,
0.0215606689453125,
-0.0487060546875,
0.0313720703125,
-0.0625,
0.02215576171875,
0.045318603515625,
0.016693115234375,
0.0167999267578125,
-0.007183074951171875,
0.0102386474609375,
0.0033359527587890625,
-0.038299560546875,
-0.034149169921875,
0.0977783203125,
0.025299072265625,
0.0517578125,
0.00405120849609375,
0.054473876953125,
0.0101776123046875,
0.0101776123046875,
-0.04833984375,
0.0467529296875,
-0.0004127025604248047,
-0.052490234375,
-0.01378631591796875,
-0.022674560546875,
-0.05072021484375,
0.0274810791015625,
-0.016632080078125,
-0.0567626953125,
0.007701873779296875,
0.0026798248291015625,
-0.034515380859375,
0.042877197265625,
-0.03204345703125,
0.05255126953125,
-0.0284881591796875,
-0.025238037109375,
0.0017614364624023438,
-0.0411376953125,
0.053741455078125,
0.007110595703125,
0.0118560791015625,
-0.024871826171875,
0.00823974609375,
0.081298828125,
-0.04376220703125,
0.04547119140625,
-0.0224761962890625,
-0.0029468536376953125,
0.040771484375,
0.004062652587890625,
0.0523681640625,
0.0232086181640625,
-0.001678466796875,
0.042694091796875,
0.0036373138427734375,
-0.0166473388671875,
-0.0230865478515625,
0.056549072265625,
-0.0887451171875,
-0.048004150390625,
-0.04327392578125,
-0.0252685546875,
0.0167999267578125,
0.0275421142578125,
0.03839111328125,
-0.0055694580078125,
0.01403045654296875,
0.0196685791015625,
0.034759521484375,
-0.004543304443359375,
0.0418701171875,
0.0210418701171875,
-0.01525115966796875,
-0.05499267578125,
0.060302734375,
0.003570556640625,
-0.0008635520935058594,
0.0284423828125,
0.01007080078125,
-0.0185394287109375,
-0.045257568359375,
-0.0430908203125,
0.0187835693359375,
-0.039459228515625,
-0.04656982421875,
-0.036895751953125,
-0.036834716796875,
-0.03814697265625,
-0.00182342529296875,
-0.040771484375,
-0.0174713134765625,
-0.05804443359375,
-0.012298583984375,
0.051544189453125,
0.0308074951171875,
-0.00469207763671875,
0.054656982421875,
-0.059417724609375,
0.0285491943359375,
0.01316070556640625,
0.01275634765625,
0.00814056396484375,
-0.0623779296875,
-0.02288818359375,
0.00772857666015625,
-0.03326416015625,
-0.04656982421875,
0.045196533203125,
-0.00119781494140625,
0.0394287109375,
0.05859375,
0.00017309188842773438,
0.08648681640625,
-0.01483917236328125,
0.06756591796875,
0.0159759521484375,
-0.052490234375,
0.0408935546875,
-0.032989501953125,
-0.0081024169921875,
0.037750244140625,
0.024169921875,
-0.029632568359375,
-0.003231048583984375,
-0.03863525390625,
-0.0604248046875,
0.0762939453125,
0.01348114013671875,
-0.006732940673828125,
0.019927978515625,
0.01654052734375,
0.00743865966796875,
0.0187225341796875,
-0.065673828125,
-0.046722412109375,
-0.03631591796875,
-0.00264739990234375,
0.00505828857421875,
-0.0113677978515625,
-0.02886962890625,
-0.03753662109375,
0.05682373046875,
-0.0023651123046875,
0.03948974609375,
0.012542724609375,
0.01439666748046875,
-0.0178070068359375,
0.00791168212890625,
0.029754638671875,
0.032562255859375,
-0.04205322265625,
-0.00829315185546875,
0.0114288330078125,
-0.0413818359375,
0.002105712890625,
0.00943756103515625,
-0.019378662109375,
-0.01087188720703125,
0.03546142578125,
0.065673828125,
0.00029397010803222656,
-0.026519775390625,
0.0218658447265625,
0.0036449432373046875,
-0.0240020751953125,
-0.032440185546875,
0.0209808349609375,
-0.0035533905029296875,
0.03717041015625,
0.042327880859375,
0.0016326904296875,
0.007343292236328125,
-0.02349853515625,
-0.00942230224609375,
0.0213165283203125,
0.01235198974609375,
-0.018829345703125,
0.0687255859375,
0.003704071044921875,
-0.01111602783203125,
0.041717529296875,
-0.0139007568359375,
-0.033843994140625,
0.05810546875,
0.039459228515625,
0.05682373046875,
-0.01047515869140625,
-0.002750396728515625,
0.061248779296875,
0.0311279296875,
-0.01055908203125,
0.0401611328125,
-0.0024127960205078125,
-0.0487060546875,
-0.01372528076171875,
-0.054656982421875,
-0.00875091552734375,
0.042755126953125,
-0.0523681640625,
0.0223541259765625,
-0.054901123046875,
-0.022308349609375,
-0.00592041015625,
0.026123046875,
-0.05322265625,
0.0208740234375,
0.0101470947265625,
0.0650634765625,
-0.055084228515625,
0.06756591796875,
0.025787353515625,
-0.0418701171875,
-0.07220458984375,
-0.02008056640625,
-0.01190948486328125,
-0.07342529296875,
0.04083251953125,
0.0122833251953125,
0.0193328857421875,
-0.000843048095703125,
-0.06781005859375,
-0.0797119140625,
0.10845947265625,
0.01367950439453125,
-0.047149658203125,
0.00852203369140625,
0.01477813720703125,
0.02490234375,
-0.0129852294921875,
0.03057861328125,
0.0545654296875,
0.04833984375,
0.0028476715087890625,
-0.060028076171875,
0.02386474609375,
-0.034912109375,
-0.009979248046875,
0.0010805130004882812,
-0.08978271484375,
0.099853515625,
-0.013031005859375,
0.002162933349609375,
0.00951385498046875,
0.05194091796875,
0.041107177734375,
0.027130126953125,
0.02789306640625,
0.05487060546875,
0.0511474609375,
-0.0239715576171875,
0.05401611328125,
-0.0072479248046875,
0.04156494140625,
0.0623779296875,
-0.0061187744140625,
0.056121826171875,
0.0305023193359375,
-0.03839111328125,
0.0377197265625,
0.069580078125,
-0.0335693359375,
0.052520751953125,
-0.00920867919921875,
-0.007049560546875,
-0.01181793212890625,
0.0021190643310546875,
-0.054962158203125,
0.025848388671875,
0.029083251953125,
-0.027130126953125,
0.005741119384765625,
-0.020477294921875,
0.0165863037109375,
-0.027130126953125,
-0.024993896484375,
0.04132080078125,
-0.01209259033203125,
-0.026214599609375,
0.07647705078125,
-0.00684356689453125,
0.057647705078125,
-0.045806884765625,
-0.01082611083984375,
-0.0167388916015625,
0.01335906982421875,
-0.037017822265625,
-0.0616455078125,
-0.0015201568603515625,
0.0028533935546875,
-0.01154327392578125,
0.01485443115234375,
0.034271240234375,
-0.00926971435546875,
-0.036956787109375,
0.027069091796875,
0.005367279052734375,
0.024200439453125,
0.0086212158203125,
-0.0660400390625,
0.026275634765625,
0.0192718505859375,
-0.042999267578125,
0.018768310546875,
0.0232391357421875,
0.022247314453125,
0.053863525390625,
0.0706787109375,
0.005359649658203125,
0.0153656005859375,
-0.010406494140625,
0.07794189453125,
-0.06219482421875,
-0.029327392578125,
-0.057403564453125,
0.036529541015625,
-0.0173797607421875,
-0.038360595703125,
0.055419921875,
0.05670166015625,
0.0650634765625,
-0.0027065277099609375,
0.0714111328125,
-0.0226287841796875,
0.037322998046875,
-0.03155517578125,
0.058258056640625,
-0.055511474609375,
0.01177215576171875,
-0.0224761962890625,
-0.04144287109375,
-0.007183074951171875,
0.060272216796875,
-0.004741668701171875,
-0.0030975341796875,
0.0428466796875,
0.043731689453125,
-0.0008392333984375,
0.011199951171875,
0.0018520355224609375,
0.025421142578125,
0.0279083251953125,
0.06427001953125,
0.047943115234375,
-0.07733154296875,
0.054656982421875,
-0.0528564453125,
-0.006610870361328125,
-0.028594970703125,
-0.047332763671875,
-0.064208984375,
-0.019622802734375,
-0.0186614990234375,
-0.0292816162109375,
-0.0207061767578125,
0.0638427734375,
0.038543701171875,
-0.058502197265625,
-0.027862548828125,
0.001262664794921875,
0.008758544921875,
-0.033843994140625,
-0.0219268798828125,
0.05078125,
0.006137847900390625,
-0.059906005859375,
0.0270538330078125,
-0.00922393798828125,
0.0084991455078125,
-0.00368499755859375,
-0.0215606689453125,
-0.0199737548828125,
-0.022796630859375,
0.026824951171875,
0.0233612060546875,
-0.052581787109375,
-0.01364898681640625,
-0.013153076171875,
-0.0011396408081054688,
0.02099609375,
0.0159149169921875,
-0.03717041015625,
0.00913238525390625,
0.037322998046875,
0.025390625,
0.046783447265625,
-0.002201080322265625,
-0.00399017333984375,
-0.0304107666015625,
0.0215911865234375,
-0.00047779083251953125,
0.0269317626953125,
0.006561279296875,
-0.03887939453125,
0.055511474609375,
0.035247802734375,
-0.046722412109375,
-0.0765380859375,
-0.03131103515625,
-0.0970458984375,
-0.01214599609375,
0.0831298828125,
-0.00438690185546875,
-0.045867919921875,
0.01934814453125,
-0.0232086181640625,
0.04339599609375,
-0.044769287109375,
0.04840087890625,
0.03070068359375,
-0.00911712646484375,
-0.004055023193359375,
-0.05279541015625,
0.0282745361328125,
-0.004711151123046875,
-0.05230712890625,
-0.0023479461669921875,
0.00714874267578125,
0.0227813720703125,
0.0210418701171875,
0.035430908203125,
0.004665374755859375,
0.00867462158203125,
0.0131988525390625,
0.0090484619140625,
-0.0198211669921875,
-0.0088043212890625,
-0.004184722900390625,
-0.007305145263671875,
-0.0205841064453125,
-0.043701171875
]
] |
Aeala/GPT4-x-AlpacaDente2-30b | 2023-05-06T19:01:10.000Z | [
"transformers",
"pytorch",
"llama",
"text-generation",
"endpoints_compatible",
"has_space",
"text-generation-inference",
"region:us"
] | text-generation | Aeala | null | null | Aeala/GPT4-x-AlpacaDente2-30b | 30 | 5,963 | transformers | 2023-05-04T05:28:39 | ## Fresh Alpasta, done Al Dente!
It's da *logical* choice! Now with a similar personality emulation quality to [GPT4-X-Alpasta-30b!](https://huggingface.co/MetaIX/GPT4-X-Alpasta-30b)
## Model Info:
ChanSung's [Alpaca-LoRA-30B-elina](https://huggingface.co/LLMs/Alpaca-LoRA-30B-elina) merged with [Open Assistant's second Finetune](https://huggingface.co/OpenAssistant/oasst-sft-7-llama-30b-xor)
## Benchmarks:
**Wikitext2:** 4.662261962890625
**PTB:** 24.547462463378906
**C4:** 7.05504846572876
[4bit](https://huggingface.co/Aeala/GPT4-x-AlpacaDente2-30b/blob/main/4bit.safetensors):
**Wikitext2:** 5.016242980957031
**PTB:** 25.576189041137695
**C4:** 7.332120418548584
~ Thanks to [askmyteapot](https://huggingface.co/askmyteapot) for performing these benchmarks! | 776 | [
[
-0.062042236328125,
-0.0506591796875,
0.033050537109375,
0.032928466796875,
-0.03216552734375,
-0.0004146099090576172,
-0.0033206939697265625,
-0.053558349609375,
0.06683349609375,
0.022125244140625,
-0.042999267578125,
-0.01445770263671875,
-0.047637939453125,
-0.0007700920104980469,
-0.03448486328125,
0.0677490234375,
-0.020233154296875,
-0.005649566650390625,
0.03826904296875,
-0.0253753662109375,
-0.0259246826171875,
-0.004917144775390625,
-0.0548095703125,
-0.0270233154296875,
0.04302978515625,
0.01361083984375,
0.0733642578125,
0.04193115234375,
0.0282135009765625,
0.022705078125,
0.00525665283203125,
-0.003208160400390625,
-0.00896453857421875,
-0.03106689453125,
0.00179290771484375,
-0.006595611572265625,
-0.06451416015625,
0.017730712890625,
0.0218505859375,
0.0171966552734375,
-0.0145111083984375,
0.024261474609375,
-0.0037174224853515625,
0.045501708984375,
-0.0364990234375,
0.0057525634765625,
-0.0271148681640625,
0.0163421630859375,
-0.006011962890625,
0.005489349365234375,
-0.01271820068359375,
-0.04864501953125,
-0.00678253173828125,
-0.06524658203125,
0.00812530517578125,
-0.00614166259765625,
0.09735107421875,
0.0218048095703125,
-0.026885986328125,
-0.0216522216796875,
-0.0270538330078125,
0.0239105224609375,
-0.0467529296875,
0.05023193359375,
0.0283966064453125,
0.0340576171875,
-0.0118560791015625,
-0.0704345703125,
-0.043792724609375,
-0.0062408447265625,
0.0201568603515625,
0.01421356201171875,
-0.03369140625,
-0.0253143310546875,
-0.000274658203125,
0.0374755859375,
-0.017852783203125,
0.0045166015625,
-0.0308380126953125,
-0.01593017578125,
0.0283355712890625,
-0.0249786376953125,
0.0265960693359375,
0.0013742446899414062,
-0.0285186767578125,
-0.0210418701171875,
-0.05328369140625,
0.00823211669921875,
0.0274505615234375,
-0.0020427703857421875,
-0.0258941650390625,
0.03973388671875,
-0.032928466796875,
0.027984619140625,
0.036895751953125,
0.004726409912109375,
0.06500244140625,
-0.0256805419921875,
-0.0252532958984375,
0.01126861572265625,
0.07525634765625,
0.0367431640625,
0.0016889572143554688,
0.00975799560546875,
0.00035309791564941406,
-0.004119873046875,
0.0179595947265625,
-0.06982421875,
-0.044525146484375,
-0.0005078315734863281,
-0.032806396484375,
-0.018707275390625,
0.017181396484375,
-0.0643310546875,
0.0078887939453125,
-0.00873565673828125,
0.036529541015625,
-0.049713134765625,
-0.0250701904296875,
0.0095672607421875,
0.01153564453125,
0.03948974609375,
0.0487060546875,
-0.06109619140625,
0.036407470703125,
0.0394287109375,
0.0633544921875,
-0.007537841796875,
-0.0162200927734375,
-0.022216796875,
0.013275146484375,
-0.035247802734375,
0.054351806640625,
-0.01465606689453125,
-0.0139923095703125,
-0.0267181396484375,
0.009063720703125,
0.0201568603515625,
-0.034912109375,
0.06414794921875,
-0.0280609130859375,
0.002346038818359375,
-0.05877685546875,
-0.0113525390625,
-0.0204620361328125,
0.00986480712890625,
-0.08929443359375,
0.07012939453125,
0.020477294921875,
-0.037261962890625,
0.04693603515625,
-0.05218505859375,
-0.00882720947265625,
-0.004184722900390625,
-0.0079803466796875,
-0.028564453125,
-0.00647735595703125,
0.0206146240234375,
0.02435302734375,
-0.018402099609375,
-0.0161895751953125,
-0.032501220703125,
-0.0267181396484375,
0.0264129638671875,
-0.032745361328125,
0.0447998046875,
0.0285491943359375,
-0.037750244140625,
-0.0030384063720703125,
-0.0555419921875,
0.00946807861328125,
0.0198211669921875,
-0.030242919921875,
0.007476806640625,
-0.034210205078125,
-0.002384185791015625,
0.0240020751953125,
0.03533935546875,
-0.043212890625,
0.00446319580078125,
0.00482177734375,
0.03778076171875,
0.07391357421875,
-0.025787353515625,
0.009185791015625,
-0.03375244140625,
0.03546142578125,
-0.01248931884765625,
0.021759033203125,
0.00799560546875,
-0.0565185546875,
-0.048919677734375,
-0.047393798828125,
0.0167694091796875,
0.031524658203125,
-0.035888671875,
0.044342041015625,
-0.001186370849609375,
-0.05242919921875,
-0.051239013671875,
0.01129913330078125,
0.03076171875,
0.0245208740234375,
0.034912109375,
-0.04498291015625,
-0.01776123046875,
-0.0645751953125,
0.0127716064453125,
-0.00667572021484375,
-0.00583648681640625,
0.0121917724609375,
0.020233154296875,
-0.0157470703125,
0.05621337890625,
-0.031219482421875,
-0.0166778564453125,
-0.00550079345703125,
-0.0196685791015625,
0.0220184326171875,
0.033599853515625,
0.0709228515625,
-0.04315185546875,
-0.043365478515625,
0.0092315673828125,
-0.03173828125,
-0.0250244140625,
0.02313232421875,
-0.01654052734375,
0.0082244873046875,
0.022186279296875,
-0.0789794921875,
0.0264434814453125,
0.04541015625,
-0.049102783203125,
0.052032470703125,
-0.0009074211120605469,
0.040130615234375,
-0.073974609375,
-0.0048980712890625,
-0.0095977783203125,
-0.0152740478515625,
-0.007228851318359375,
0.00983428955078125,
0.0164642333984375,
0.0164642333984375,
-0.0537109375,
0.03619384765625,
-0.0280609130859375,
-0.01519012451171875,
-0.004779815673828125,
-0.0069580078125,
0.0152435302734375,
0.046539306640625,
-0.039520263671875,
0.06341552734375,
0.0299072265625,
-0.022979736328125,
0.041168212890625,
0.0302581787109375,
-0.03326416015625,
0.0218353271484375,
-0.07354736328125,
0.0015039443969726562,
0.02081298828125,
0.0384521484375,
-0.057373046875,
-0.0222930908203125,
0.041015625,
-0.029205322265625,
0.00820159912109375,
0.018707275390625,
-0.044525146484375,
-0.0291595458984375,
-0.055877685546875,
0.045440673828125,
0.045196533203125,
-0.059478759765625,
0.036041259765625,
0.01393890380859375,
-0.002765655517578125,
-0.033477783203125,
-0.050750732421875,
-0.0072784423828125,
-0.0207672119140625,
-0.027923583984375,
0.041717529296875,
-0.0177459716796875,
0.005344390869140625,
0.01442718505859375,
-0.0241546630859375,
0.0020961761474609375,
-0.0178985595703125,
0.02081298828125,
0.0191650390625,
-0.0181732177734375,
-0.04345703125,
0.024627685546875,
-0.0056915283203125,
-0.0007305145263671875,
0.0196685791015625,
0.07269287109375,
-0.03314208984375,
-0.032440185546875,
-0.048187255859375,
0.00983428955078125,
0.035247802734375,
-0.0187835693359375,
0.0208740234375,
0.0616455078125,
-0.029388427734375,
0.00646209716796875,
-0.0548095703125,
0.00927734375,
-0.0430908203125,
0.01509857177734375,
-0.05517578125,
-0.054718017578125,
0.0528564453125,
0.021453857421875,
-0.00550079345703125,
0.037445068359375,
0.046356201171875,
0.0144805908203125,
0.081787109375,
0.0129241943359375,
-0.0166168212890625,
0.03424072265625,
-0.0193023681640625,
0.0038909912109375,
-0.056427001953125,
-0.029266357421875,
-0.027923583984375,
-0.019744873046875,
-0.06378173828125,
-0.03228759765625,
0.0080718994140625,
0.005062103271484375,
-0.039764404296875,
0.07464599609375,
-0.04351806640625,
0.034912109375,
0.04400634765625,
0.03277587890625,
0.0194244384765625,
-0.0114288330078125,
-0.0033397674560546875,
0.007537841796875,
-0.0164642333984375,
-0.031219482421875,
0.075927734375,
0.03216552734375,
0.0531005859375,
0.032806396484375,
0.051025390625,
0.0131072998046875,
0.0129852294921875,
-0.0264434814453125,
0.062255859375,
-0.01800537109375,
-0.021026611328125,
-0.003818511962890625,
-0.0269622802734375,
-0.072509765625,
0.02288818359375,
0.005023956298828125,
-0.059356689453125,
0.004695892333984375,
0.01432037353515625,
-0.0286712646484375,
0.020660400390625,
-0.057403564453125,
0.0677490234375,
-0.006038665771484375,
-0.01605224609375,
-0.00908660888671875,
-0.0194091796875,
0.034027099609375,
-0.00943756103515625,
0.0029888153076171875,
-0.036346435546875,
-0.01227569580078125,
0.060394287109375,
-0.042327880859375,
0.03338623046875,
0.0017194747924804688,
-0.0216827392578125,
0.0280303955078125,
0.0035800933837890625,
0.0194549560546875,
0.0190277099609375,
-0.00472259521484375,
0.022186279296875,
0.0079193115234375,
-0.034271240234375,
-0.038726806640625,
0.08258056640625,
-0.07489013671875,
-0.0233001708984375,
-0.036102294921875,
-0.01415252685546875,
-0.0149993896484375,
0.0198211669921875,
0.03594970703125,
0.0216064453125,
-0.0265350341796875,
0.00433349609375,
0.037628173828125,
-0.01441192626953125,
0.037078857421875,
0.035003662109375,
-0.0199127197265625,
-0.05096435546875,
0.063720703125,
-0.0010976791381835938,
0.026214599609375,
0.0224456787109375,
0.0129241943359375,
-0.0167388916015625,
-0.010772705078125,
-0.0408935546875,
0.0166778564453125,
-0.03656005859375,
-0.0258941650390625,
-0.04302978515625,
-0.01522064208984375,
-0.02984619140625,
-0.01263427734375,
-0.055999755859375,
-0.04364013671875,
-0.027984619140625,
-0.0250396728515625,
0.0408935546875,
0.03594970703125,
-0.016845703125,
0.0301971435546875,
-0.036346435546875,
0.03692626953125,
0.03369140625,
0.035888671875,
-0.013275146484375,
-0.03839111328125,
0.006504058837890625,
0.0007009506225585938,
-0.05670166015625,
-0.07843017578125,
0.0293731689453125,
0.002346038818359375,
0.03802490234375,
0.040130615234375,
-0.026092529296875,
0.05194091796875,
-0.0458984375,
0.047393798828125,
0.0364990234375,
-0.07122802734375,
0.0264434814453125,
-0.0297393798828125,
0.0169219970703125,
0.05712890625,
0.030303955078125,
-0.003704071044921875,
-0.021575927734375,
-0.04522705078125,
-0.068115234375,
0.039886474609375,
0.0176544189453125,
0.01194000244140625,
-0.0126953125,
0.00852203369140625,
0.035491943359375,
0.0192413330078125,
-0.06536865234375,
-0.00537872314453125,
-0.051849365234375,
-0.013641357421875,
0.01105499267578125,
0.0027523040771484375,
0.006755828857421875,
-0.031524658203125,
0.053131103515625,
-0.005733489990234375,
0.045440673828125,
0.00206756591796875,
0.0201873779296875,
-0.01052093505859375,
-0.006160736083984375,
0.038177490234375,
0.036773681640625,
-0.037872314453125,
0.0012826919555664062,
0.0032138824462890625,
-0.052703857421875,
0.0037937164306640625,
0.0160675048828125,
-0.005756378173828125,
0.0118255615234375,
0.0116119384765625,
0.06414794921875,
0.0254364013671875,
-0.048492431640625,
0.04400634765625,
-0.0034542083740234375,
0.005283355712890625,
-0.035797119140625,
0.02020263671875,
-0.00479888916015625,
0.00933074951171875,
0.0183868408203125,
0.0042572021484375,
0.0218048095703125,
-0.054107666015625,
0.0253753662109375,
0.02630615234375,
-0.01288604736328125,
-0.045654296875,
0.060302734375,
0.003971099853515625,
-0.0118865966796875,
0.0231170654296875,
-0.036163330078125,
-0.029388427734375,
0.06378173828125,
0.0496826171875,
0.059356689453125,
-0.04608154296875,
0.0134735107421875,
0.0300140380859375,
0.0255126953125,
-0.01338958740234375,
0.03326416015625,
-0.0060577392578125,
-0.061859130859375,
-0.0231781005859375,
-0.040252685546875,
-0.04046630859375,
0.007732391357421875,
-0.057464599609375,
0.0374755859375,
-0.0445556640625,
-0.01300811767578125,
0.007549285888671875,
-0.0006575584411621094,
-0.05908203125,
0.0185394287109375,
-0.0234527587890625,
0.09661865234375,
-0.058441162109375,
0.07879638671875,
0.022125244140625,
-0.03387451171875,
-0.0738525390625,
-0.011962890625,
-0.0046844482421875,
-0.0743408203125,
0.01137542724609375,
0.031585693359375,
0.009033203125,
-0.01509857177734375,
-0.020538330078125,
-0.08038330078125,
0.109619140625,
0.026702880859375,
-0.0297088623046875,
0.005939483642578125,
-0.007183074951171875,
0.01541900634765625,
-0.01934814453125,
0.0267791748046875,
0.04718017578125,
0.03955078125,
0.00965118408203125,
-0.0919189453125,
0.0256805419921875,
-0.041046142578125,
-0.0213165283203125,
0.0283966064453125,
-0.08575439453125,
0.07525634765625,
-0.01824951171875,
0.0176544189453125,
0.0228118896484375,
0.057708740234375,
0.04071044921875,
0.021728515625,
0.041534423828125,
0.056671142578125,
0.0304412841796875,
-0.01099395751953125,
0.0706787109375,
-0.0024204254150390625,
0.01122283935546875,
0.08416748046875,
-0.034515380859375,
0.056854248046875,
0.0306396484375,
-0.022125244140625,
0.045074462890625,
0.053924560546875,
-0.01435089111328125,
0.062164306640625,
-0.0185394287109375,
-0.0107269287109375,
0.0235443115234375,
0.01007080078125,
-0.050048828125,
0.03302001953125,
0.00951385498046875,
-0.01544189453125,
-0.002643585205078125,
-0.0005946159362792969,
0.0173187255859375,
-0.01556396484375,
-0.019073486328125,
0.03411865234375,
-0.0030364990234375,
-0.041717529296875,
0.0221405029296875,
0.0169219970703125,
0.06378173828125,
-0.0545654296875,
-0.0036373138427734375,
-0.0148773193359375,
-0.0008716583251953125,
-0.0178985595703125,
-0.06378173828125,
0.0091094970703125,
-0.019195556640625,
0.0161285400390625,
-0.004550933837890625,
0.03973388671875,
-0.00295257568359375,
-0.034423828125,
0.035064697265625,
0.007205963134765625,
0.01308441162109375,
0.00799560546875,
-0.048187255859375,
0.049713134765625,
0.0013151168823242188,
-0.038604736328125,
0.0235748291015625,
0.0193634033203125,
-0.00286865234375,
0.050537109375,
0.04449462890625,
0.004886627197265625,
0.02264404296875,
0.0025997161865234375,
0.07489013671875,
-0.0390625,
-0.044647216796875,
-0.056793212890625,
0.004558563232421875,
-0.00506591796875,
-0.04901123046875,
0.049285888671875,
0.060272216796875,
0.07049560546875,
-0.0130615234375,
0.0283050537109375,
-0.0303497314453125,
0.0171661376953125,
-0.04595947265625,
0.040618896484375,
-0.040374755859375,
0.004528045654296875,
-0.045654296875,
-0.0811767578125,
0.015960693359375,
0.0628662109375,
0.00730133056640625,
0.035797119140625,
0.03692626953125,
0.04974365234375,
-0.03436279296875,
0.01084136962890625,
0.006893157958984375,
0.005687713623046875,
0.0303192138671875,
0.052764892578125,
0.029632568359375,
-0.0477294921875,
0.034423828125,
-0.04608154296875,
-0.042633056640625,
-0.00914764404296875,
-0.0689697265625,
-0.034576416015625,
-0.0340576171875,
-0.028106689453125,
-0.020050048828125,
-0.01119232177734375,
0.07550048828125,
0.0643310546875,
-0.053466796875,
0.0006375312805175781,
0.0018911361694335938,
-0.0201873779296875,
-0.03216552734375,
-0.021026611328125,
0.03472900390625,
0.028106689453125,
-0.06524658203125,
0.0286407470703125,
0.00496673583984375,
0.01378631591796875,
0.003940582275390625,
-0.022216796875,
0.00693511962890625,
0.0209808349609375,
0.047332763671875,
0.0221710205078125,
-0.053924560546875,
-0.040924072265625,
-0.00527191162109375,
-0.0008234977722167969,
0.01190948486328125,
0.04400634765625,
-0.040130615234375,
-0.0180816650390625,
0.0254364013671875,
0.0274658203125,
0.049102783203125,
-0.004734039306640625,
0.017242431640625,
-0.0249786376953125,
0.0178985595703125,
-0.0063629150390625,
0.048187255859375,
0.0217742919921875,
-0.007205963134765625,
0.050140380859375,
-0.0046844482421875,
-0.036773681640625,
-0.060028076171875,
0.017425537109375,
-0.1297607421875,
-0.0077972412109375,
0.07366943359375,
0.004390716552734375,
-0.01898193359375,
0.028350830078125,
-0.0230560302734375,
0.01708984375,
-0.056610107421875,
0.06427001953125,
0.03985595703125,
-0.0138092041015625,
-0.0193023681640625,
-0.057098388671875,
0.01229095458984375,
0.0304412841796875,
-0.07061767578125,
-0.042205810546875,
0.02545166015625,
0.044525146484375,
0.0240478515625,
0.06585693359375,
-0.035369873046875,
0.060516357421875,
-0.022430419921875,
0.014251708984375,
-0.0138702392578125,
-0.01519012451171875,
0.014801025390625,
0.00995635986328125,
0.010009765625,
-0.047821044921875
]
] |
beomi/KoAlpaca-KoRWKV-6B | 2023-09-15T01:27:53.000Z | [
"transformers",
"pytorch",
"safetensors",
"rwkv",
"text-generation",
"generated_from_trainer",
"KoRWKV",
"KoAlpaca",
"ko",
"dataset:beomi/KoAlpaca-v1.1a",
"license:apache-2.0",
"endpoints_compatible",
"region:us"
] | text-generation | beomi | null | null | beomi/KoAlpaca-KoRWKV-6B | 7 | 5,963 | transformers | 2023-06-02T12:00:13 | ---
language:
- ko
license: apache-2.0
library_name: transformers
tags:
- generated_from_trainer
- KoRWKV
- KoAlpaca
datasets:
- beomi/KoAlpaca-v1.1a
pipeline_tag: text-generation
base_model: beomi/KoRWKV-6B
model-index:
- name: KoRWKV-6B-koalpaca-v1.1a
results: []
---
> 🚧 Note: this repo is under construction, current uploaded version is based on KoRWKV-6B, ~28Billion tokens trained ckpt 🚧
# 🐧 KoAlpaca-KoRWKV-6B (v1.1a)
This model is a fine-tuned version of [beomi/KoRWKV-6B](https://huggingface.co/beomi/KoRWKV-6B) on an [KoAlpaca v1.1a Dataset](https://huggingface.co/datasets/beomi/KoAlpaca-v1.1a).
Detail Codes are available at [KoAlpaca Github Repository](https://github.com/Beomi/KoAlpaca)
## Generation Code
### Install Requirements
```bash
pip install -U transformers tokenizers accelerate safetensors
```
### Generation Code with HF pipeline
```python
from transformers import pipeline
pipe = pipeline('text-generation', model='beomi/KoAlpaca-KoRWKV-6B', device=0)
def gen(x):
a = pipe(
f'### 질문: {x}\n\n### 답변:',
max_new_tokens=1024,
return_full_text=False,
do_sample=True,
top_p=0.9,
temperature=0.7,
early_stopping=True,
repetition_penalty=1.2,
)
print(a[0]['generated_text'])
```
### Generated Samples
```
gen("오늘 기분은 좀 어때?")
오늘은 기분이 좋은 날입니다. 왜냐하면, 오늘은 기쁜 일이 생길 수 있는 날이니까요!
1. 아침에 일찍 일어나서 기분 좋았습니다. 2. 맛있는 음식을 먹어서 기분이 좋습니다. 3. 수업시간에 열심히 들어서 기분이 좋습니다. 4. 친구가 생일 선물을 줘서 기분이 좋습니다. 5. 내일이 토요일이라서 기분이 좋습니다. 6. 숙제를 해서 기분이 좋습니다. 7. 부모님께서 생일 축하 노래를 불러줘서 기분이 좋습니다. 8. 방 청소를 해서 기분이 좋습니다. 9. 맛있는 과자를 먹어서 기분이 좋습니다. 10. 오늘 날씨를 보니 비가 안와서 기분이 좋습니다. 11. 부모님이 나를 사랑해주셔서 기분이 좋습니다. 12. 부모님과 함께 외식을 해서 기분이 좋습니다. 13. 친구랑 같이 놀아서 기분이 좋습니다. 14. 공부를 해서 기분이 좋습니다. 15. 숙제를 잘 해서 기분이 좋습니다. 16. 맛있는 음식을 먹어서 기분이 좋습니다. 17. 친구들과 어울려 놀아서 기분이 좋습니다. 18. 내가 좋아하는 책을 읽어서 기분이 좋습니다. 19. 오늘 하루 종일 기분이 좋습니다. 20. 엄마와 아빠가 나를 사랑해 주셔서 기분이 좋습니다.
```
```
gen("칸트의 정언명령에 따른다는 점이 일반적인 선함과 다른 이유가 뭐야?")
칸트는 도덕의 법칙을 두 가지로 구분했습니다. 첫 번째 법칙은 '정언명령'으로, 이 법칙은 어떤 상황에서든 절대적으로 지켜야 하는 보편적인 법칙입니다. 이에 따라 칸트는 "이성이 명령한 대로 행동하라"는 원칙을 제시했습니다.
두 번째 법칙은 "모든 인간은 평등하게 태어났고, 창조주로부터 양도할 수 없는 권리를 부여받았다"는 것입니다. 이 원칙에는 모든 인간이 평등하고, 모든 인간은 동등하게 존중받아야 한다는 의미가 담겨 있습니다. 이러한 칸트의 도덕 법칙을 따르면, 선함과 악함은 상황에 따라 다르게 적용될 수 있습니다. 예를 들어, 범죄자나 장애인 등 특별한 상황에서는 선이 될 수 있는 상황이 존재합니다.
```
```
gen("다음 글을 영어로 번역해주세요. 영어 문장: 'She looks handsome.' 번역 한글 문장:")
"She looks handsome."은 직역하면 "그녀는 잘생겨 보입니다"라는 뜻이지만, 한국어로는 "그녀는 아름다워 보입니다"로 해석됩니다. 이 문장은 원래 'The saddest thing'이라는 영화의 OST 가사 중 일부분입니다. 이 영화는 제인 폰다가 주연을 맡은 작품으로, 그녀가 매력적인 미모와 몸매를 가지고 있지만 불행한 삶을 사는 여성의 모습을 그리고 있습니다. 따라서 그녀는 아름답지만 불행해보이는 모습이라는 의미에서 사용되었습니다.
```
```
gen("서울이 인기있는 도시인 이유는?")
서울은 많은 사람들이 거주하고 있는 대도시이기 때문입니다. 또한, 다른 나라에서는 서울보다 더 큰 대도시들이 많이 있지만, 수도가 서울에 위치한 나라는 없습니다. 이는 서울의 도시 규모가 크고 교통과 통신 등 인프라가 잘 구축되어 있기 때문입니다.
추가 설명:
서울의 인구나 유동인구 등을 살펴보면, 서울을 방문하는 외국인 관광객도 매우 많습니다. 이러한 이유로 서울은 한국 내에서 가장 인기 있는 도시 중 하나로 자리잡고 있습니다.
```
```
gen("칸트 철학의 예시를 들고, 해당 철학이 가지는 의의를 3가지 알려줘.")
칸트는 인간의 지식을 경험론적 입장에서 분석하고 종합하여 도덕철학을 정립하였습니다. 이 과정에서 그는 모든 인간은 이성을 가지고 있으며, 도덕법칙을 인식할 수 있다고 보았습니다. 이러한 인간 이성의 기능을 통해, 칸트는 '실천이성'과 '판단력'을 중시합니다. 즉, 이성적으로 판단하여 올바른 판단을 내릴 수 있는 능력이 인간에게는 있다는 것입니다. 또한, 칸트는 인간이 가진 선천적인 관념론적 사고방식을 거부하고, 경험을 통한 실천적인 사고방식으로서의 이성을 강조합니다.
추가 설명:
칸트의 철학은 형이상학, 인식론, 윤리학, 정치학 등 다양한 분야에서 영향을 미치고 있습니다. 특히, 그는 경험론과 합리론의 대립으로 인해 분열된 유럽의 사상계를 통합하여 새로운 철학을 제시하고자 노력하였습니다. 이를 통해 근대 철학에 큰 영향을 주었으며, 현재는 그 영향력이 여전히 유지되고 있습니다.
```
```
gen("리튬이온전지에서 발생하는 화학 현상을 화학식을 기반으로 자세히 설명해주세요.")
리튬이온전지는 양극과 음극 사이에 있는 분리막을 통해 이온이 이동하면서 충전과 방전을 하는 원리로 작동합니다. 따라서, 전극의 반응에 의해 화학 반응이 일어납니다.
1. 리튬이온전지 내부에는 리튬이온들이 이동하기 위한 통로가 있는데, 이를 전해질이라고 합니다. 즉, 액체 전해질을 채우고 있는 이온들의 농도 차이 때문에 전극에서 전자가 이동하는 것입니다. 이때 전해질은 물질이 아닌 화학반응을 일으키는 역할을 합니다. 2. 음극은 전자를 받아 이온으로 생성하는 부분입니다. 3. 음극에서는 산화 환원 반응을 통해 전자와 수소 이온이 이동하며, 이를 통해 전기가 발생합니다. 4. 전자는 음극에서 양극으로 이동하며, 이는 전자의 흐름이며, 이를 통해 전류가 흐르게 됩니다. 5. 전자는 양극에서 음극으로 이동하는데, 이는 전자가 전자이동을 하면서 발생한 운동에너지로 인해 발생합니다. 6. 전기적으로 가장 안정된 상태인 리튬이온은 가장 높은 온도(약 700도)에서도 잘 견딜 수 있습니다.
```
```
gen("투자 초심자가 하기 좋은 선물 투자 종류를 알려주세요.")
주식, 채권, 금, 원유 등 다양한 상품이 있습니다.
1. 주식 (주식투자) - 장기적인 관점에서 투자할 때 적합한 상품입니다.
2. 금 (금투자) - 안전한 투자를 원하는 사람에게 좋습니다.
3. 원유 (원자재투자) - 안정적인 수익과 가격 변동성이 적습니다.
4. 부동산 (부동산투자) - 인플레이션이 발생하는 시기에 가장 적합한 투자 방법 중 하나입니다.
5. 보험 (보험투자) - 보장자산에 대한 니즈가 있는 사람에게 적합한 상품입니다.
```
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 2e-05
- train_batch_size: 1
- seed: 42
- gradient_accumulation_steps: 8
- total_train_batch_size: 8
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 1.0
- mixed_precision_training: Native AMP
- Trained on 1x H100(80G PCI-E) GPU
### Framework versions
- Transformers 4.29.2
- Pytorch 1.13.1
- Datasets 2.12.0
- Tokenizers 0.13.3 | 4,797 | [
[
-0.051116943359375,
-0.046600341796875,
0.01497650146484375,
0.026153564453125,
-0.03173828125,
0.0038623809814453125,
0.015411376953125,
-0.0293121337890625,
0.045379638671875,
0.01396942138671875,
-0.026123046875,
-0.03271484375,
-0.055419921875,
0.0160064697265625,
-0.0188446044921875,
0.036224365234375,
-0.00457763671875,
-0.01010894775390625,
-0.0015211105346679688,
0.00807952880859375,
-0.035247802734375,
-0.00795745849609375,
-0.0648193359375,
-0.01171112060546875,
-0.002899169921875,
0.0163421630859375,
0.053955078125,
0.03631591796875,
0.044219970703125,
0.038421630859375,
-0.0009188652038574219,
-0.005611419677734375,
-0.01910400390625,
0.002414703369140625,
0.00719451904296875,
-0.025909423828125,
-0.005092620849609375,
-0.0086822509765625,
0.026885986328125,
0.0418701171875,
-0.01488494873046875,
0.0288848876953125,
0.01010894775390625,
0.06976318359375,
-0.03729248046875,
0.005939483642578125,
0.00576019287109375,
0.0120086669921875,
-0.008270263671875,
-0.01512908935546875,
0.0175323486328125,
-0.08282470703125,
-0.005397796630859375,
-0.046539306640625,
0.0209503173828125,
0.011993408203125,
0.09002685546875,
-0.01910400390625,
-0.0113677978515625,
-0.011932373046875,
-0.05291748046875,
0.058624267578125,
-0.052001953125,
0.0175018310546875,
0.038330078125,
0.00626373291015625,
-0.0318603515625,
-0.04583740234375,
-0.046905517578125,
0.0054779052734375,
-0.01313018798828125,
0.0258636474609375,
-0.01360321044921875,
-0.040130615234375,
0.032440185546875,
0.036285400390625,
-0.0288238525390625,
-0.0100250244140625,
-0.0311126708984375,
-0.00972747802734375,
0.060760498046875,
0.006473541259765625,
0.043243408203125,
-0.040802001953125,
-0.048553466796875,
0.0007648468017578125,
-0.00353240966796875,
0.030670166015625,
0.0367431640625,
-0.0196990966796875,
-0.03363037109375,
0.0474853515625,
-0.027252197265625,
0.030517578125,
0.0202789306640625,
-0.022247314453125,
0.047454833984375,
-0.04443359375,
-0.037384033203125,
-0.0008635520935058594,
0.0780029296875,
0.043792724609375,
0.005214691162109375,
0.00902557373046875,
0.002559661865234375,
-0.0008034706115722656,
-0.00728607177734375,
-0.05450439453125,
-0.0027446746826171875,
0.0283050537109375,
-0.0545654296875,
-0.0261993408203125,
0.024688720703125,
-0.09649658203125,
-0.0016422271728515625,
-0.0239105224609375,
0.0252838134765625,
-0.043121337890625,
-0.0305938720703125,
0.00510406494140625,
-0.0166778564453125,
0.0155029296875,
0.0231170654296875,
-0.059661865234375,
0.0250091552734375,
0.0037784576416015625,
0.05230712890625,
0.0163116455078125,
-0.006622314453125,
0.0278778076171875,
0.0146331787109375,
-0.02923583984375,
0.049835205078125,
0.010009765625,
-0.03448486328125,
-0.00560760498046875,
0.0156707763671875,
-0.032470703125,
-0.004718780517578125,
0.0236358642578125,
-0.004405975341796875,
0.0116729736328125,
-0.004413604736328125,
-0.03466796875,
0.00298309326171875,
0.0190582275390625,
-0.040802001953125,
0.09393310546875,
0.01117706298828125,
-0.0809326171875,
0.006221771240234375,
-0.0272979736328125,
-0.0198211669921875,
0.00853729248046875,
0.00055694580078125,
-0.04486083984375,
-0.01212310791015625,
0.02008056640625,
0.04156494140625,
-0.019256591796875,
-0.01390838623046875,
-0.009002685546875,
-0.025787353515625,
0.0185089111328125,
-0.0125732421875,
0.0830078125,
0.03271484375,
-0.036224365234375,
-0.007045745849609375,
-0.066650390625,
0.0245361328125,
0.051239013671875,
-0.028411865234375,
0.0006690025329589844,
-0.034271240234375,
-0.0184173583984375,
0.030670166015625,
0.0272216796875,
-0.041961669921875,
0.00881195068359375,
-0.034820556640625,
0.03826904296875,
0.07269287109375,
0.0177001953125,
0.0290679931640625,
-0.05181884765625,
0.043701171875,
-0.00638580322265625,
0.0311126708984375,
0.007106781005859375,
-0.032562255859375,
-0.049072265625,
-0.00991058349609375,
0.004383087158203125,
0.04925537109375,
-0.053009033203125,
0.050384521484375,
-0.003894805908203125,
-0.054443359375,
-0.02996826171875,
-0.00689697265625,
0.018218994140625,
0.029022216796875,
0.017730712890625,
-0.0005092620849609375,
-0.056365966796875,
-0.045196533203125,
-0.0009241104125976562,
-0.0083465576171875,
-0.003635406494140625,
0.036834716796875,
0.0633544921875,
-0.019439697265625,
0.0618896484375,
-0.05474853515625,
-0.0266265869140625,
-0.025604248046875,
-0.01418304443359375,
0.03741455078125,
0.055267333984375,
0.06494140625,
-0.06988525390625,
-0.06396484375,
-0.0034198760986328125,
-0.0777587890625,
0.01708984375,
-0.00232696533203125,
-0.0025348663330078125,
0.01496124267578125,
0.0290069580078125,
-0.05340576171875,
0.04937744140625,
0.032470703125,
-0.04351806640625,
0.0799560546875,
-0.042449951171875,
0.0271759033203125,
-0.10076904296875,
0.0290069580078125,
0.00713348388671875,
-0.0006899833679199219,
-0.03863525390625,
-0.0015544891357421875,
-0.0147552490234375,
0.00811004638671875,
-0.05096435546875,
0.05218505859375,
-0.0443115234375,
0.0159149169921875,
0.01448822021484375,
0.01007080078125,
0.00628662109375,
0.0293426513671875,
-0.019927978515625,
0.056976318359375,
0.052520751953125,
-0.0367431640625,
0.0275115966796875,
0.0219573974609375,
-0.0299530029296875,
0.034271240234375,
-0.04901123046875,
-0.007076263427734375,
-0.034210205078125,
0.0205078125,
-0.08123779296875,
-0.0333251953125,
0.059478759765625,
-0.05780029296875,
0.0169830322265625,
-0.006072998046875,
-0.017913818359375,
-0.0709228515625,
-0.047760009765625,
0.0157928466796875,
0.019805908203125,
-0.0149993896484375,
0.0391845703125,
0.0169219970703125,
-0.0179595947265625,
-0.041412353515625,
-0.031890869140625,
0.00881195068359375,
-0.005321502685546875,
-0.05804443359375,
0.031646728515625,
-0.016387939453125,
0.000560760498046875,
-0.007450103759765625,
0.006023406982421875,
-0.00394439697265625,
-0.00762939453125,
0.032257080078125,
0.0184783935546875,
-0.001499176025390625,
-0.0185394287109375,
-0.01471710205078125,
-0.01555633544921875,
0.002655029296875,
0.007434844970703125,
0.05426025390625,
-0.006877899169921875,
-0.0169525146484375,
-0.0596923828125,
0.046844482421875,
0.048492431640625,
0.0006775856018066406,
0.0599365234375,
0.043670654296875,
-0.0153350830078125,
0.0020961761474609375,
-0.036590576171875,
0.01522064208984375,
-0.038177490234375,
0.0013856887817382812,
-0.022430419921875,
-0.03936767578125,
0.048858642578125,
-0.02215576171875,
-0.03338623046875,
0.054046630859375,
0.038421630859375,
-0.025299072265625,
0.08697509765625,
0.035858154296875,
-0.00045418739318847656,
0.0294036865234375,
-0.06280517578125,
0.0214996337890625,
-0.044708251953125,
-0.060211181640625,
-0.043670654296875,
-0.0263214111328125,
-0.055023193359375,
-0.025360107421875,
0.0165863037109375,
0.0148468017578125,
-0.0208282470703125,
0.027679443359375,
-0.061492919921875,
0.01340484619140625,
0.032470703125,
0.0307769775390625,
-0.005664825439453125,
-0.019317626953125,
-0.01068115234375,
0.0133514404296875,
-0.043975830078125,
-0.03466796875,
0.073486328125,
0.0272064208984375,
0.037200927734375,
0.010833740234375,
0.036865234375,
0.0262298583984375,
-0.0102081298828125,
-0.04327392578125,
0.042877197265625,
0.005176544189453125,
-0.043060302734375,
-0.03533935546875,
-0.02581787109375,
-0.0830078125,
0.023345947265625,
-0.0235748291015625,
-0.07440185546875,
0.04193115234375,
0.00359344482421875,
-0.0220794677734375,
0.041656494140625,
-0.05120849609375,
0.055938720703125,
-0.006591796875,
-0.047760009765625,
0.009735107421875,
-0.060638427734375,
0.032958984375,
0.0002739429473876953,
0.032196044921875,
-0.01360321044921875,
0.025299072265625,
0.0618896484375,
-0.07135009765625,
0.0188446044921875,
-0.01812744140625,
0.004150390625,
0.04620361328125,
-0.0177459716796875,
0.0478515625,
0.01065826416015625,
-0.004669189453125,
0.006435394287109375,
0.0098419189453125,
-0.043182373046875,
-0.00640869140625,
0.0557861328125,
-0.07275390625,
-0.040802001953125,
-0.03863525390625,
0.00720977783203125,
0.0168304443359375,
0.0345458984375,
0.05133056640625,
0.0225677490234375,
0.0051727294921875,
0.025665283203125,
0.044952392578125,
-0.0325927734375,
0.054412841796875,
-0.0090179443359375,
-0.0093994140625,
-0.0306396484375,
0.045318603515625,
0.0123443603515625,
0.0162200927734375,
-0.005039215087890625,
0.020111083984375,
-0.0287933349609375,
-0.04571533203125,
-0.035552978515625,
0.0198516845703125,
-0.035064697265625,
-0.02581787109375,
-0.05194091796875,
-0.003917694091796875,
-0.06353759765625,
-0.02410888671875,
-0.0217132568359375,
-0.02838134765625,
-0.01078033447265625,
-0.017852783203125,
0.034698486328125,
0.0302886962890625,
-0.0288238525390625,
0.0133819580078125,
-0.053497314453125,
0.0389404296875,
0.01222991943359375,
0.03460693359375,
0.0086517333984375,
-0.0252532958984375,
-0.01419830322265625,
0.021759033203125,
-0.03717041015625,
-0.0836181640625,
0.04730224609375,
-0.032196044921875,
0.0211639404296875,
0.05230712890625,
-0.0011491775512695312,
0.053436279296875,
-0.01277923583984375,
0.06427001953125,
0.03436279296875,
-0.06304931640625,
0.055145263671875,
-0.04833984375,
0.01239776611328125,
0.040557861328125,
0.039031982421875,
-0.03851318359375,
-0.0091552734375,
-0.047515869140625,
-0.07037353515625,
0.05364990234375,
0.0229949951171875,
-0.0031108856201171875,
0.0018072128295898438,
0.00836181640625,
-0.0278167724609375,
0.01125335693359375,
-0.037750244140625,
-0.06439208984375,
-0.026092529296875,
0.01654052734375,
0.0106048583984375,
-0.009735107421875,
-0.00852203369140625,
-0.053955078125,
0.04486083984375,
0.025115966796875,
0.045867919921875,
0.047271728515625,
0.01458740234375,
-0.0286712646484375,
0.0159912109375,
0.04071044921875,
0.06109619140625,
-0.02020263671875,
-0.01317596435546875,
0.02325439453125,
-0.06292724609375,
0.038787841796875,
0.001995086669921875,
-0.037200927734375,
0.00446319580078125,
-0.0002143383026123047,
0.060516357421875,
0.0088348388671875,
-0.01861572265625,
0.05157470703125,
-0.01788330078125,
-0.032135009765625,
-0.0484619140625,
-0.01076507568359375,
0.005390167236328125,
-0.0010080337524414062,
0.040283203125,
0.012542724609375,
-0.0113067626953125,
-0.042022705078125,
0.01038360595703125,
0.026214599609375,
-0.0189056396484375,
0.005756378173828125,
0.0562744140625,
0.0012102127075195312,
-0.0142974853515625,
0.03314208984375,
-0.01047515869140625,
-0.034423828125,
0.07672119140625,
0.049468994140625,
0.046539306640625,
-0.047882080078125,
0.031982421875,
0.06591796875,
0.007205963134765625,
-0.0126495361328125,
0.029144287109375,
0.0243988037109375,
-0.037353515625,
-0.0079498291015625,
-0.061187744140625,
-0.00902557373046875,
0.03857421875,
-0.041168212890625,
0.0011138916015625,
-0.0484619140625,
-0.026641845703125,
-0.0016393661499023438,
0.00728607177734375,
-0.0596923828125,
0.030548095703125,
-0.00551605224609375,
0.06329345703125,
-0.0728759765625,
0.042388916015625,
0.056610107421875,
-0.03912353515625,
-0.08160400390625,
-0.003086090087890625,
-0.0059967041015625,
-0.04180908203125,
0.057525634765625,
0.00897216796875,
0.00882720947265625,
0.01424407958984375,
-0.03369140625,
-0.09466552734375,
0.11004638671875,
-0.00533294677734375,
-0.0297698974609375,
0.0312347412109375,
0.023651123046875,
0.021484375,
0.0045013427734375,
0.0157623291015625,
0.0285186767578125,
0.05419921875,
0.020050048828125,
-0.061614990234375,
0.042266845703125,
-0.035308837890625,
0.00632476806640625,
0.029266357421875,
-0.0711669921875,
0.06854248046875,
-0.017547607421875,
-0.00704193115234375,
-0.003326416015625,
0.0280609130859375,
0.0197601318359375,
0.03521728515625,
0.033477783203125,
0.053070068359375,
0.0241851806640625,
-0.01381683349609375,
0.07391357421875,
-0.017303466796875,
0.04229736328125,
0.0228271484375,
0.01384735107421875,
0.0280609130859375,
0.0268707275390625,
-0.050872802734375,
0.0224609375,
0.037750244140625,
-0.04327392578125,
0.0221405029296875,
0.0142822265625,
-0.041534423828125,
0.006500244140625,
0.0012331008911132812,
-0.041290283203125,
0.0164642333984375,
0.01300048828125,
-0.0307159423828125,
0.00377655029296875,
0.0226593017578125,
0.00841522216796875,
-0.0032138824462890625,
-0.0312347412109375,
0.046661376953125,
-0.0017375946044921875,
-0.03436279296875,
0.02655029296875,
0.004322052001953125,
0.0428466796875,
-0.049774169921875,
0.01100921630859375,
-0.018585205078125,
0.0077362060546875,
-0.03863525390625,
-0.056060791015625,
0.0127105712890625,
-0.008697509765625,
-0.02740478515625,
-0.005397796630859375,
0.07440185546875,
0.01385498046875,
-0.046722412109375,
0.0259246826171875,
0.0225982666015625,
0.0162353515625,
0.0052490234375,
-0.06292724609375,
-0.0076141357421875,
0.010955810546875,
-0.035064697265625,
0.03814697265625,
0.026031494140625,
0.00969696044921875,
0.061187744140625,
0.07110595703125,
0.03173828125,
0.0211181640625,
-0.0266876220703125,
0.0657958984375,
-0.0555419921875,
-0.037689208984375,
-0.053985595703125,
0.04705810546875,
-0.019622802734375,
-0.011383056640625,
0.076171875,
0.054595947265625,
0.054840087890625,
-0.01450347900390625,
0.08526611328125,
-0.041351318359375,
0.041717529296875,
-0.016815185546875,
0.05084228515625,
-0.024200439453125,
-0.00926971435546875,
-0.0284576416015625,
-0.046539306640625,
-0.0147857666015625,
0.05322265625,
-0.0179290771484375,
0.01139068603515625,
0.03155517578125,
0.047027587890625,
0.0154266357421875,
0.01108551025390625,
0.006694793701171875,
0.022216796875,
0.0027751922607421875,
0.031890869140625,
0.043701171875,
-0.055419921875,
0.0474853515625,
-0.049713134765625,
-0.01331329345703125,
-0.03131103515625,
-0.044158935546875,
-0.049285888671875,
-0.0139007568359375,
-0.0208282470703125,
-0.0200653076171875,
-0.015472412109375,
0.062103271484375,
0.0249176025390625,
-0.055389404296875,
-0.01314544677734375,
-0.0019197463989257812,
0.0124359130859375,
-0.043792724609375,
-0.0262298583984375,
0.061431884765625,
-0.00257110595703125,
-0.06475830078125,
-0.00984954833984375,
-0.00887298583984375,
0.018951416015625,
0.00653839111328125,
0.01036834716796875,
-0.0411376953125,
0.0003261566162109375,
0.04376220703125,
0.0263824462890625,
-0.0307159423828125,
-0.0225677490234375,
-0.006755828857421875,
-0.01837158203125,
0.014068603515625,
0.0303192138671875,
-0.035003662109375,
0.012908935546875,
0.052825927734375,
0.01155853271484375,
0.042724609375,
0.029144287109375,
0.00875091552734375,
-0.024139404296875,
0.0014514923095703125,
-0.004474639892578125,
0.0245819091796875,
-0.01251220703125,
-0.04949951171875,
0.0379638671875,
0.040557861328125,
-0.055267333984375,
-0.03521728515625,
-0.005580902099609375,
-0.073974609375,
-0.04534912109375,
0.07061767578125,
-0.0160064697265625,
-0.020721435546875,
-0.02484130859375,
-0.0265655517578125,
0.03369140625,
-0.03668212890625,
0.05999755859375,
0.0258941650390625,
-0.0254364013671875,
-0.00838470458984375,
-0.07183837890625,
0.0382080078125,
0.020111083984375,
-0.05926513671875,
-0.020721435546875,
0.0208587646484375,
0.018798828125,
0.0215301513671875,
0.07293701171875,
-0.0010976791381835938,
0.016693115234375,
-0.002948760986328125,
0.006633758544921875,
-0.01007843017578125,
0.0152435302734375,
-0.006664276123046875,
0.030029296875,
-0.0394287109375,
-0.045135498046875
]
] |
Open-Orca/OpenOrcaxOpenChat-Preview2-13B | 2023-08-21T06:20:09.000Z | [
"transformers",
"pytorch",
"llama",
"text-generation",
"en",
"dataset:Open-Orca/OpenOrca",
"arxiv:2306.02707",
"arxiv:2301.13688",
"arxiv:2307.09288",
"license:llama2",
"endpoints_compatible",
"has_space",
"text-generation-inference",
"region:us"
] | text-generation | Open-Orca | null | null | Open-Orca/OpenOrcaxOpenChat-Preview2-13B | 97 | 5,963 | transformers | 2023-07-31T11:08:55 | ---
license: llama2
language:
- en
library_name: transformers
pipeline_tag: text-generation
datasets:
- Open-Orca/OpenOrca
---
<p><h1>🐋 The Second OpenOrca Model Preview! 🐋</h1></p>
# OpenOrca x OpenChat - Preview2 - 13B
We have used our own [OpenOrca dataset](https://huggingface.co/datasets/Open-Orca/OpenOrca) to fine-tune Llama2-13B using [OpenChat](https://huggingface.co/openchat) packing.
This dataset is our attempt to reproduce the dataset generated for Microsoft Research's [Orca Paper](https://arxiv.org/abs/2306.02707).
This second preview release is trained on a curated filtered subset of most of our GPT-4 augmented data.
This release highlights that our dataset and training methods have surpassed performance parity with the Orca paper.
We measured this with BigBench-Hard and AGIEval results with the same methods as used in the Orca paper, finding **~103%** of original Orca's performance on average.
As well, this is done with <1/10th the compute requirement and using <20% of the dataset size from the original Orca paper.
We have run extensive evaluations internally and expect this model to **place number 1** on both the HuggingFaceH4 Open LLM Leaderboard and the GPT4ALL Leaderboard for 13B models.
"One" of [OpenChat](https://huggingface.co/openchat) has joined our team, and we'd like to provide special thanks for their training of this model!
We have utilized OpenChat [MultiPack algorithm](https://github.com/imoneoi/multipack_sampler) which achieves 99.85% bin-packing efficiency on our dataset.
This has significantly reduced training time, with efficiency improvement of 3-10X over traditional methods.
<img src="https://raw.githubusercontent.com/imoneoi/openchat/master/assets/logo_new.png" style="width: 40%">
Want to visualize our full (pre-filtering) dataset? Check out our [Nomic Atlas Map](https://atlas.nomic.ai/map/c1b88b47-2d9b-47e0-9002-b80766792582/2560fd25-52fe-42f1-a58f-ff5eccc890d2).
[<img src="https://huggingface.co/Open-Orca/OpenOrca-Preview1-13B/resolve/main/OpenOrca%20Nomic%20Atlas.png" alt="Atlas Nomic Dataset Map" width="400" height="400" />](https://atlas.nomic.ai/map/c1b88b47-2d9b-47e0-9002-b80766792582/2560fd25-52fe-42f1-a58f-ff5eccc890d2)
We are in-process with training more models, so keep a look out on our org for releases coming soon with exciting partners.
We will also give sneak-peak announcements on our Discord, which you can find here:
https://AlignmentLab.ai
# Prompt Template
We use our own prompt template which we call "`OpenChat Llama2 V1`".
The model is heavily conditioned to work using this format only and will likely encounter issues such as run-on output which emulates a chat between a user and assistant if this format is not properly followed.
Examples:
```
# Single-turn `OpenChat Llama2 V1`
tokenize("You are OpenOrcaChat.<|end_of_turn|>User: Hello<|end_of_turn|>Assistant:")
# [1, 887, 526, 4673, 2816, 1113, 1451, 271, 29889, 32000, 4911, 29901, 15043, 32000, 4007, 22137, 29901]
# Multi-turn `OpenChat Llama2 V1`
tokenize("You are OpenOrcaChat.<|end_of_turn|>User: Hello<|end_of_turn|>Assistant: Hi<|end_of_turn|>User: How are you today?<|end_of_turn|>Assistant:")
# [1, 887, 526, 4673, 2816, 1113, 1451, 271, 29889, 32000, 4911, 29901, 15043, 32000, 4007, 22137, 29901, 6324, 32000, 4911, 29901, 1128, 526, 366, 9826, 29973, 32000, 4007, 22137, 29901]
```
For UIs with Prefix and Suffix fields, these will likely work:
Prefix (include a space after colon):
```
User:
```
Suffix (space after colon):
```
<|end_of_turn|>\nAssistant:
```
**Oobabooga's text-generation-webui instructions can be found [further down the page](https://huggingface.co/Open-Orca/OpenOrcaxOpenChat-Preview2-13B#serving-with-oobabooga--text-generation-webui).**
# Evaluation
We have evaluated **OpenOrcaxOpenChat-Preview2-13B** on hard reasoning tasks from BigBench-Hard and AGIEval as outlined in the Orca paper.
Our average performance for BigBench-Hard: 0.488
Average for AGIEval: 0.447
We find our score averages to **~103%** of the total performance that was shown in the Orca paper, using the same evaluation methods as outlined in the paper.
So we are surpassing Orca performance with <20% of the dataset size and <1/10th the training budget!
As well, we have evaluated using the methodology and tools for the HuggingFace Leaderboard and GPT4ALL Leaderboard, and find that we place #1 on both for all 13B models at release time!
## AGIEval Performance
We present our results in two columns.
The column for "`(Orca Paper eval)`" uses the methods outlined in the Orca paper, so as to be a direct apples-to-apples comparison with the results from the paper.
The column for "`(HF Leaderboard eval)`" uses EleutherAI's LM Evaluation Harness with settings outlined by HuggingFace. These results are not comparable to the other columns, as the methods are different.
## BigBench-Hard Performance
We present our results in two columns.
The column for "`(Orca Paper eval)`" uses the methods outlined in the Orca paper, so as to be a direct apples-to-apples comparison with the results from the paper.
The column for "`(HF Leaderboard eval)`" uses EleutherAI's LM Evaluation Harness with settings outlined by HuggingFace. These results are not comparable to the other columns, as the methods are different.
## HuggingFaceH4 Open LLM Leaderboard Performance
We have run our own tests using parameters matching the [HuggingFaceH4 Open LLM Leaderboard](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard) evals.
We place #1 for all 13B models at release time!
**Update Aug 10th:** The official results on the leaderboard are below.
Since our release, a new model which merges an Orca-style model with a Platypus (trained on STEM and logic) model places narrowly above ours, but we were #1 at release time.
Below we also highlight how our model fits relative to models of all sizes on the current (as of Aug 10th, 2023) leaderboard.
Notably, performance is beyond falcon-40b-instruct, and close to LLaMA1-65B base.
## GPT4ALL Leaderboard Performance
We have tested using parameters matching the GPT4ALL Benchmark Suite and report our results and placement vs their official reporting below.
We place #1 for all open models and come within comparison of `text-davinci-003`, a proprietary OpenAI model an order of magnitude larger.
# Dataset
We used a curated, filtered selection of most of the GPT-4 augmented data from our OpenOrca dataset, which aims to reproduce the Orca Research Paper dataset.
Further details of our curation practices will be forthcoming with our full model releases.
# Training
We trained with 8x A100-80G GPUs for 46 hours, completing 5 epochs of full fine tuning on our dataset in one training run.
This contrasts with the 20x A100-80G GPUs for 200 hours used in the Orca paper, for only 3 epochs, and requiring stacked training (which is known to suffer catastrophic forgetting).
Our compute requirement was <1/10th that of the original Orca.
Commodity cost was ~$600.
Please await our full releases for further training details.
# Serving
This model is most easily served with [OpenChat's](https://github.com/imoneoi/openchat) customized vLLM OpenAI-compatible API server.
This is highly recommended as it is by far the fastest in terms of inference speed and is a quick and easy option for setup.
We also illustrate setup of Oobabooga/text-generation-webui below. The settings outlined there will also apply to other uses of `Transformers`.
## Serving Quantized
Pre-quantized models are now available courtesy of our friend TheBloke:
* **GGML**: https://huggingface.co/TheBloke/OpenOrcaxOpenChat-Preview2-13B-GGML
* **GPTQ**: https://huggingface.co/TheBloke/OpenOrcaxOpenChat-Preview2-13B-GPTQ
The serving instructions below only apply to the unquantized model being presented in the repository you are viewing here.
There are some notes, such as on use of the prompt format, that will still apply to the quantized models though.
## Serving with OpenChat
[Install OpenChat](https://github.com/imoneoi/openchat/#installation)
After installation, run:
```bash
python -m ochat.serving.openai_api_server \
--model-type openchat_llama2 \
--model Open-Orca/OpenOrcaxOpenChat-Preview2-13B \
--engine-use-ray --worker-use-ray --max-num-batched-tokens 5120
```
Follow the OpenChat documentation to use features such as tensor parallelism on consumer GPUs, API keys, and logging.
You may then connect to the OpenAI-compatible API endpoint with tools such as [BetterGPT.chat](https://bettergpt.chat).
## Serving with Oobabooga / text-generation-webui
The model may also be loaded via [oobabooga/text-generation-webui](https://github.com/oobabooga/text-generation-webui/) in a similar manner to other models.
See the requirements below. Note that inference with just the Transformers library is significantly slower than using the recommended OpenChat vLLM server.
### Oobabooga Key Requirements
* You will first need to download the model as you normally do to the "`models/`" folder of your `text-generation-webui` installation.
* To use the unquantized model presented here, select "`Transformers`"" in the webui's "`Model`" tab "`Model loader`" dropdown.
* You will likely want to tick "`auto-devices`". The model will require >40GB VRAM after loading in context for inference.
* The model was trained in bf16, so tick the "`bf16`" box for best performance.
* It will run safely on single GPUs with VRAM >=48GB (e.g. A6000)
* If using consumer GPUs, e.g. 2x RTX3090 24GB, you will likely want to enter "18,17" under "`tensor_split`" to split the model across both GPUs
* The model will perform significantly better if you use the appropriate prompting template
* We will submit a PR to include our prompting template into text-generation-webui soon
* For now, manually enter the settings described in the following sections:
### Oobabooga Chat Settings
In the "`Chat settings`" tab, select the following settings:
For "`User String`" ...
```
User:
```
For "`Bot string`" ...
```
Assistant:
```
For "`Context`", this is analogous to system prompt.
It is not necessary, but we have found good results with the below example.
System prompts used in the Orca training also work well. ...
```
You are a helpful assistant. Please answer truthfully and write out your thinking step by step to be sure you get the right answer. If you make a mistake or encounter an error in your thinking, say so out loud and attempt to correct it. If you don't know or aren't sure about something, say so clearly. You will act as a professional logician, mathematician, and physicist. You will also act as the most appropriate type of expert to answer any particular question or solve the relevant problem; state which expert type your are, if so. Also think of any particular named expert that would be ideal to answer the relevant question or solve the relevant problem; name and act as them, if appropriate.
```
For "`Turn template`", this is absolutely essential to have. You will get poor, mixed up output without this template ...
```
<|user|> <|user-message|><|end_of_turn|>\n<|bot|> <|bot-message|>\n
```
When done, it should look as below:
<img src="https://huggingface.co/Open-Orca/OpenOrcaxOpenChat-Preview2-13B/resolve/main/Images/OpenOrcaLlama2OobaboogaChatInstructionTemplate.png" style="width: 40%">
You may then save this as a named template preset by clicking the "Floppy" icon and giving it an appropriate name in the popup, e.g. "`OpenOrcaxOpenChat Llama2`".
### Oobabooga Text Generation Mode
In the "`Text generation`" tab, select "`instruct`" as the mode:
#### Mode Illustration
It should look as below:
<img src="https://huggingface.co/Open-Orca/OpenOrcaxOpenChat-Preview2-13B/resolve/main/Images/OpenOrcaLlama2OobaboogaInstructMode.png" style="width: 40%">
Then you should be ready to generate!
# Citation
```bibtex
@software{OpenOrcaxOpenChatPreview2,
title = {OpenOrcaxOpenChatPreview2: Llama2-13B Model Instruct-tuned on Filtered OpenOrcaV1 GPT-4 Dataset},
author = {Guan Wang and Bleys Goodson and Wing Lian and Eugene Pentland and Austin Cook and Chanvichet Vong and "Teknium"},
year = {2023},
publisher = {HuggingFace},
journal = {HuggingFace repository},
howpublished = {\url{https://https://huggingface.co/Open-Orca/OpenOrcaxOpenChat-Preview2-13B},
}
@software{openchat,
title = {{OpenChat: Advancing Open-source Language Models with Imperfect Data}},
author = {Wang, Guan and Cheng, Sijie and Yu, Qiying and Liu, Changling},
doi = {10.5281/zenodo.8105775},
url = {https://github.com/imoneoi/openchat},
version = {pre-release},
year = {2023},
month = {7},
}
@misc{mukherjee2023orca,
title={Orca: Progressive Learning from Complex Explanation Traces of GPT-4},
author={Subhabrata Mukherjee and Arindam Mitra and Ganesh Jawahar and Sahaj Agarwal and Hamid Palangi and Ahmed Awadallah},
year={2023},
eprint={2306.02707},
archivePrefix={arXiv},
primaryClass={cs.CL}
}
@misc{longpre2023flan,
title={The Flan Collection: Designing Data and Methods for Effective Instruction Tuning},
author={Shayne Longpre and Le Hou and Tu Vu and Albert Webson and Hyung Won Chung and Yi Tay and Denny Zhou and Quoc V. Le and Barret Zoph and Jason Wei and Adam Roberts},
year={2023},
eprint={2301.13688},
archivePrefix={arXiv},
primaryClass={cs.AI}
}
@misc{touvron2023llama,
title={Llama 2: Open Foundation and Fine-Tuned Chat Models},
author={Hugo Touvron and Louis Martin and Kevin Stone and Peter Albert and Amjad Almahairi and Yasmine Babaei and Nikolay Bashlykov and Soumya Batra and Prajjwal Bhargava and Shruti Bhosale and Dan Bikel and Lukas Blecher and Cristian Canton Ferrer and Moya Chen and Guillem Cucurull and David Esiobu and Jude Fernandes and Jeremy Fu and Wenyin Fu and Brian Fuller and Cynthia Gao and Vedanuj Goswami and Naman Goyal and Anthony Hartshorn and Saghar Hosseini and Rui Hou and Hakan Inan and Marcin Kardas and Viktor Kerkez and Madian Khabsa and Isabel Kloumann and Artem Korenev and Punit Singh Koura and Marie-Anne Lachaux and Thibaut Lavril and Jenya Lee and Diana Liskovich and Yinghai Lu and Yuning Mao and Xavier Martinet and Todor Mihaylov and Pushkar Mishra and Igor Molybog and Yixin Nie and Andrew Poulton and Jeremy Reizenstein and Rashi Rungta and Kalyan Saladi and Alan Schelten and Ruan Silva and Eric Michael Smith and Ranjan Subramanian and Xiaoqing Ellen Tan and Binh Tang and Ross Taylor and Adina Williams and Jian Xiang Kuan and Puxin Xu and Zheng Yan and Iliyan Zarov and Yuchen Zhang and Angela Fan and Melanie Kambadur and Sharan Narang and Aurelien Rodriguez and Robert Stojnic and Sergey Edunov and Thomas Scialom},
year={2023},
eprint={2307.09288},
archivePrefix={arXiv},
}
``` | 16,167 | [
[
-0.04168701171875,
-0.0635986328125,
0.0146484375,
0.0103302001953125,
-0.020050048828125,
-0.010589599609375,
-0.0118560791015625,
-0.0728759765625,
0.0273284912109375,
0.018585205078125,
-0.031219482421875,
-0.054473876953125,
-0.034454345703125,
-0.00995635986328125,
-0.005840301513671875,
0.08746337890625,
-0.01139068603515625,
-0.016326904296875,
0.0031871795654296875,
-0.0439453125,
-0.034698486328125,
-0.050537109375,
-0.059600830078125,
-0.0243682861328125,
0.046112060546875,
0.02569580078125,
0.052764892578125,
0.050262451171875,
0.038604736328125,
0.0185394287109375,
-0.030487060546875,
0.025634765625,
-0.0506591796875,
-0.01055908203125,
0.01068115234375,
-0.0250396728515625,
-0.0806884765625,
0.020751953125,
0.028656005859375,
0.019134521484375,
-0.02850341796875,
0.0200042724609375,
0.01357269287109375,
0.033172607421875,
-0.0458984375,
0.02911376953125,
-0.01611328125,
-0.005863189697265625,
-0.032684326171875,
0.008056640625,
-0.017822265625,
-0.041839599609375,
-0.01058197021484375,
-0.05633544921875,
-0.005062103271484375,
0.0160675048828125,
0.08892822265625,
0.003940582275390625,
-0.019073486328125,
-0.0221710205078125,
-0.026397705078125,
0.032806396484375,
-0.040435791015625,
0.0166473388671875,
0.01491546630859375,
0.0219573974609375,
-0.0287628173828125,
-0.047515869140625,
-0.0406494140625,
-0.0226593017578125,
0.0091400146484375,
0.03399658203125,
-0.011444091796875,
-0.0042724609375,
0.01343536376953125,
0.047576904296875,
-0.047149658203125,
0.0310821533203125,
-0.04443359375,
-0.00450897216796875,
0.0555419921875,
0.01123809814453125,
0.008819580078125,
0.0190582275390625,
-0.0352783203125,
-0.056243896484375,
-0.0469970703125,
0.0274200439453125,
0.038787841796875,
0.02862548828125,
-0.0462646484375,
0.044097900390625,
0.00505828857421875,
0.033416748046875,
-0.0023021697998046875,
-0.01445770263671875,
0.036834716796875,
-0.023468017578125,
-0.0301361083984375,
-0.005741119384765625,
0.0723876953125,
0.02325439453125,
-0.0029754638671875,
0.01546478271484375,
-0.0033111572265625,
0.0167694091796875,
-0.003551483154296875,
-0.0687255859375,
-0.017608642578125,
0.011871337890625,
-0.0430908203125,
-0.0224609375,
0.0135650634765625,
-0.05316162109375,
-0.0222320556640625,
0.005168914794921875,
0.00916290283203125,
-0.0284576416015625,
-0.0261993408203125,
0.0157928466796875,
0.006023406982421875,
0.034271240234375,
0.03558349609375,
-0.05572509765625,
0.035064697265625,
0.040283203125,
0.078125,
-0.01482391357421875,
-0.0202484130859375,
-0.0233001708984375,
-0.0300445556640625,
-0.031280517578125,
0.055694580078125,
-0.0229034423828125,
-0.03546142578125,
-0.0232086181640625,
-0.01611328125,
-0.00299072265625,
-0.03460693359375,
0.050567626953125,
-0.0214080810546875,
0.0214996337890625,
-0.038055419921875,
-0.0086517333984375,
-0.02142333984375,
0.01015472412109375,
-0.04632568359375,
0.091796875,
-0.001007080078125,
-0.0455322265625,
0.02197265625,
-0.07525634765625,
-0.00591278076171875,
-0.0301666259765625,
0.006687164306640625,
-0.0362548828125,
-0.00991058349609375,
0.034454345703125,
0.0196990966796875,
-0.0296173095703125,
-0.0178375244140625,
-0.046722412109375,
-0.0202484130859375,
0.006092071533203125,
0.0035724639892578125,
0.06201171875,
0.0229644775390625,
-0.01081085205078125,
0.0007309913635253906,
-0.036834716796875,
-0.0029773712158203125,
0.03729248046875,
-0.020355224609375,
-0.00432586669921875,
-0.030914306640625,
-0.0037441253662109375,
0.01776123046875,
0.0154266357421875,
-0.053497314453125,
0.044586181640625,
-0.03863525390625,
0.038604736328125,
0.05072021484375,
-0.020416259765625,
0.023681640625,
-0.026031494140625,
0.04229736328125,
0.0014495849609375,
0.02783203125,
-0.01328277587890625,
-0.06854248046875,
-0.048736572265625,
-0.02093505859375,
0.0309600830078125,
0.03131103515625,
-0.0274658203125,
0.0306854248046875,
-0.0071563720703125,
-0.06903076171875,
-0.0350341796875,
-0.0113067626953125,
0.04730224609375,
0.03887939453125,
0.0254669189453125,
-0.072998046875,
-0.0264739990234375,
-0.0484619140625,
0.00830078125,
-0.03375244140625,
0.0124969482421875,
0.038970947265625,
0.049560546875,
0.009613037109375,
0.059234619140625,
-0.043060302734375,
-0.032562255859375,
-0.005405426025390625,
0.0034942626953125,
0.0203857421875,
0.0325927734375,
0.071044921875,
-0.0369873046875,
-0.0191497802734375,
0.0040740966796875,
-0.065185546875,
0.007358551025390625,
0.0263824462890625,
-0.03216552734375,
0.04095458984375,
0.0303955078125,
-0.050933837890625,
0.05792236328125,
0.048858642578125,
-0.0369873046875,
0.033721923828125,
-0.020843505859375,
-0.0036525726318359375,
-0.05584716796875,
0.02349853515625,
-0.00249481201171875,
0.00030803680419921875,
-0.01837158203125,
0.010986328125,
-0.007358551025390625,
-0.003543853759765625,
-0.0286712646484375,
0.0577392578125,
-0.04095458984375,
-0.0177764892578125,
0.004978179931640625,
0.0163726806640625,
-0.00015819072723388672,
0.049713134765625,
-0.01324462890625,
0.053680419921875,
0.03546142578125,
-0.0229034423828125,
0.01200103759765625,
0.045196533203125,
-0.0220184326171875,
0.0293121337890625,
-0.0623779296875,
0.031280517578125,
-0.007335662841796875,
0.05535888671875,
-0.0946044921875,
-0.0170440673828125,
0.04046630859375,
-0.033843994140625,
0.0292816162109375,
0.003574371337890625,
-0.043914794921875,
-0.04931640625,
-0.034698486328125,
0.029510498046875,
0.039825439453125,
-0.05792236328125,
0.040771484375,
0.0240936279296875,
0.0011491775512695312,
-0.03753662109375,
-0.051025390625,
-0.003620147705078125,
-0.0217742919921875,
-0.06512451171875,
0.03515625,
0.0005331039428710938,
0.0052032470703125,
-0.0127105712890625,
-0.006717681884765625,
0.0116424560546875,
-0.0033283233642578125,
0.044921875,
0.032318115234375,
-0.0168609619140625,
-0.006191253662109375,
-0.0145721435546875,
-0.0093841552734375,
-0.00684356689453125,
-0.00994873046875,
0.05072021484375,
-0.03521728515625,
-0.0162811279296875,
-0.03533935546875,
-0.01221466064453125,
0.033843994140625,
-0.03826904296875,
0.0716552734375,
0.038482666015625,
-0.0036163330078125,
0.027587890625,
-0.045318603515625,
-0.01611328125,
-0.03369140625,
0.0021209716796875,
-0.0306854248046875,
-0.057525634765625,
0.064453125,
0.03546142578125,
0.0311737060546875,
0.045745849609375,
0.03515625,
0.017852783203125,
0.0723876953125,
0.049163818359375,
-0.01476287841796875,
0.034881591796875,
-0.03521728515625,
0.006023406982421875,
-0.059112548828125,
-0.039459228515625,
-0.042633056640625,
-0.0452880859375,
-0.056976318359375,
-0.0193939208984375,
0.0390625,
0.01371002197265625,
-0.030059814453125,
0.0298309326171875,
-0.048309326171875,
0.016510009765625,
0.04083251953125,
0.019134521484375,
0.0161285400390625,
0.004909515380859375,
-0.009033203125,
0.01140594482421875,
-0.050933837890625,
-0.0396728515625,
0.09564208984375,
0.031982421875,
0.050262451171875,
0.009979248046875,
0.04437255859375,
-0.01153564453125,
0.03564453125,
-0.021087646484375,
0.036773681640625,
0.010284423828125,
-0.04058837890625,
-0.0087890625,
-0.0299530029296875,
-0.0902099609375,
0.01910400390625,
-0.005523681640625,
-0.07080078125,
0.017852783203125,
0.00772857666015625,
-0.04119873046875,
0.024658203125,
-0.05364990234375,
0.0775146484375,
-0.01174163818359375,
-0.0189971923828125,
0.0027217864990234375,
-0.061187744140625,
0.03717041015625,
0.018951416015625,
-0.0011272430419921875,
0.0025482177734375,
-0.0188140869140625,
0.05987548828125,
-0.052581787109375,
0.055877685546875,
-0.0134735107421875,
-0.0078277587890625,
0.034210205078125,
-0.00354766845703125,
0.0250396728515625,
-0.006633758544921875,
-0.010955810546875,
0.04083251953125,
-0.006378173828125,
-0.026702880859375,
-0.0224761962890625,
0.0576171875,
-0.077392578125,
-0.007472991943359375,
-0.042236328125,
-0.015869140625,
0.0090789794921875,
-0.007598876953125,
0.0264892578125,
0.0289459228515625,
-0.0226287841796875,
-0.00021398067474365234,
0.0181121826171875,
-0.037109375,
0.025665283203125,
0.022186279296875,
-0.034515380859375,
-0.03656005859375,
0.04986572265625,
0.0220184326171875,
0.0011920928955078125,
0.004283905029296875,
0.0033111572265625,
-0.037872314453125,
-0.0277557373046875,
-0.0328369140625,
0.042633056640625,
-0.03082275390625,
-0.0198211669921875,
-0.0556640625,
-0.00933837890625,
-0.032928466796875,
0.011688232421875,
-0.03582763671875,
-0.032135009765625,
-0.0290374755859375,
-0.00952911376953125,
0.039947509765625,
0.06097412109375,
-0.00957489013671875,
0.03125,
-0.025390625,
0.002521514892578125,
0.007633209228515625,
0.0276641845703125,
0.0117950439453125,
-0.050567626953125,
0.00011640787124633789,
0.0059814453125,
-0.06414794921875,
-0.04632568359375,
0.0215301513671875,
0.004238128662109375,
0.0137939453125,
0.03131103515625,
-0.0053863525390625,
0.06414794921875,
-0.01140594482421875,
0.06658935546875,
0.00984954833984375,
-0.036651611328125,
0.0362548828125,
-0.0268096923828125,
0.01337432861328125,
0.032867431640625,
0.02569580078125,
-0.0059356689453125,
-0.01788330078125,
-0.07366943359375,
-0.06390380859375,
0.07958984375,
0.034027099609375,
-0.007335662841796875,
0.01085662841796875,
0.04254150390625,
0.0037136077880859375,
0.020416259765625,
-0.046630859375,
-0.0274810791015625,
-0.0161895751953125,
0.0186309814453125,
-0.0143890380859375,
-0.017913818359375,
0.012237548828125,
-0.01470184326171875,
0.04931640625,
0.00714874267578125,
0.03753662109375,
0.006000518798828125,
0.03076171875,
-0.002391815185546875,
-0.0147552490234375,
0.045440673828125,
0.04443359375,
-0.03179931640625,
-0.036834716796875,
0.01708984375,
-0.04510498046875,
-0.02447509765625,
0.0221099853515625,
0.0144805908203125,
-0.0205841064453125,
0.0182037353515625,
0.06939697265625,
-0.01264190673828125,
-0.0352783203125,
0.0361328125,
-0.0168914794921875,
-0.0089569091796875,
-0.01425933837890625,
0.01085662841796875,
0.0022373199462890625,
0.019866943359375,
0.00504302978515625,
-0.003063201904296875,
-0.0012388229370117188,
-0.05572509765625,
-0.0217132568359375,
0.01837158203125,
-0.006259918212890625,
-0.03692626953125,
0.06536865234375,
0.0032558441162109375,
-0.003940582275390625,
0.0526123046875,
-0.01445770263671875,
-0.0272369384765625,
0.05902099609375,
0.01076507568359375,
0.035491943359375,
-0.0298004150390625,
-0.0002751350402832031,
0.04150390625,
0.014007568359375,
-0.017822265625,
0.03173828125,
0.0099029541015625,
-0.0286712646484375,
-0.020599365234375,
-0.035980224609375,
-0.032989501953125,
0.01552581787109375,
-0.058197021484375,
0.0333251953125,
-0.0458984375,
-0.0281219482421875,
0.00817108154296875,
0.01552581787109375,
-0.058441162109375,
0.003997802734375,
0.00968170166015625,
0.082275390625,
-0.048980712890625,
0.041107177734375,
0.06549072265625,
-0.0489501953125,
-0.0743408203125,
-0.0284423828125,
0.005443572998046875,
-0.0662841796875,
0.0306549072265625,
0.03802490234375,
0.007904052734375,
-0.019989013671875,
-0.058380126953125,
-0.06280517578125,
0.09124755859375,
0.039337158203125,
-0.0255279541015625,
0.0034027099609375,
-0.006587982177734375,
0.0531005859375,
-0.0276031494140625,
0.06976318359375,
0.0467529296875,
0.038604736328125,
0.023468017578125,
-0.0943603515625,
0.0105438232421875,
-0.0260162353515625,
0.0031681060791015625,
0.01097869873046875,
-0.091552734375,
0.08154296875,
-0.020477294921875,
-0.01438140869140625,
0.034271240234375,
0.05426025390625,
0.0243377685546875,
0.0252838134765625,
0.0297393798828125,
0.06890869140625,
0.06085205078125,
-0.0232086181640625,
0.10888671875,
0.00295257568359375,
0.018096923828125,
0.071044921875,
-0.016876220703125,
0.05914306640625,
0.0052337646484375,
-0.01486968994140625,
0.043609619140625,
0.061370849609375,
0.00916290283203125,
0.03240966796875,
-0.006504058837890625,
0.021270751953125,
0.00008016824722290039,
-0.01360321044921875,
-0.056884765625,
0.054473876953125,
0.0162506103515625,
-0.01253509521484375,
-0.027740478515625,
-0.0033931732177734375,
0.025604248046875,
-0.0162506103515625,
-0.01342010498046875,
0.053436279296875,
0.012237548828125,
-0.05206298828125,
0.07525634765625,
0.0210113525390625,
0.0423583984375,
-0.0367431640625,
0.004253387451171875,
-0.036651611328125,
0.0126953125,
-0.0157318115234375,
-0.05413818359375,
0.004955291748046875,
-0.004581451416015625,
0.01898193359375,
-0.0167236328125,
0.026824951171875,
-0.0034084320068359375,
-0.0057220458984375,
0.030975341796875,
0.0313720703125,
0.038909912109375,
-0.019561767578125,
-0.054473876953125,
0.033111572265625,
-0.00238037109375,
-0.035064697265625,
0.038726806640625,
0.0273895263671875,
-0.01416015625,
0.040008544921875,
0.056060791015625,
-0.01255035400390625,
-0.0120849609375,
-0.0159912109375,
0.09344482421875,
-0.032470703125,
-0.041290283203125,
-0.06097412109375,
0.027130126953125,
-0.00455474853515625,
-0.063232421875,
0.052215576171875,
0.048828125,
0.06427001953125,
0.0280914306640625,
0.0222015380859375,
-0.0248565673828125,
0.0288543701171875,
-0.0290374755859375,
0.043060302734375,
-0.0576171875,
0.00571441650390625,
-0.0201416015625,
-0.082763671875,
-0.0023288726806640625,
0.051177978515625,
-0.033233642578125,
0.0269012451171875,
0.034820556640625,
0.06854248046875,
-0.0204315185546875,
0.02593994140625,
-0.00707244873046875,
0.036590576171875,
0.03363037109375,
0.05633544921875,
0.05126953125,
-0.05206298828125,
0.056854248046875,
-0.01392364501953125,
-0.040679931640625,
-0.03326416015625,
-0.05792236328125,
-0.0830078125,
-0.0254669189453125,
-0.0166473388671875,
-0.038482666015625,
0.012969970703125,
0.0504150390625,
0.049530029296875,
-0.045013427734375,
-0.0274810791015625,
0.013519287109375,
-0.01044464111328125,
-0.0188140869140625,
-0.011260986328125,
0.0352783203125,
0.01218414306640625,
-0.050445556640625,
0.020233154296875,
0.00536346435546875,
0.0142669677734375,
-0.0103912353515625,
-0.0116424560546875,
-0.004337310791015625,
-0.01312255859375,
0.0305938720703125,
0.05035400390625,
-0.03564453125,
-0.01358795166015625,
-0.001312255859375,
-0.01520538330078125,
0.028350830078125,
0.0211944580078125,
-0.06280517578125,
0.019866943359375,
0.0180206298828125,
0.008636474609375,
0.07177734375,
0.0205078125,
0.011474609375,
-0.034027099609375,
0.033203125,
0.0003962516784667969,
0.0239410400390625,
0.0183258056640625,
-0.0035572052001953125,
0.07098388671875,
-0.0033397674560546875,
-0.0455322265625,
-0.071533203125,
-0.01476287841796875,
-0.0867919921875,
0.006633758544921875,
0.069580078125,
-0.0173492431640625,
-0.029815673828125,
0.02020263671875,
-0.032135009765625,
0.00830078125,
-0.06414794921875,
0.052947998046875,
0.033782958984375,
-0.01399993896484375,
0.01497650146484375,
-0.047088623046875,
0.0090484619140625,
0.0230865478515625,
-0.05914306640625,
-0.0198974609375,
0.029022216796875,
0.0045318603515625,
0.0265350341796875,
0.047637939453125,
-0.0236663818359375,
0.016937255859375,
-0.01392364501953125,
0.0167999267578125,
-0.0316162109375,
-0.0247650146484375,
-0.01561737060546875,
0.008148193359375,
-0.00567626953125,
-0.02581787109375
]
] |
Phind/Phind-CodeLlama-34B-v1 | 2023-08-28T19:53:12.000Z | [
"transformers",
"pytorch",
"llama",
"text-generation",
"code llama",
"license:llama2",
"model-index",
"endpoints_compatible",
"has_space",
"text-generation-inference",
"region:us"
] | text-generation | Phind | null | null | Phind/Phind-CodeLlama-34B-v1 | 319 | 5,963 | transformers | 2023-08-25T20:16:25 | ---
license: llama2
model-index:
- name: Phind-CodeLlama-34B-v1
results:
- task:
type: text-generation
dataset:
type: openai_humaneval
name: HumanEval
metrics:
- name: pass@1
type: pass@1
value: 67.6%
verified: false
tags:
- code llama
---
# NOTE: We've now launched **Phind-CodeLlama-34B-v2**, which acheives **73.8% pass@1** on HumanEval. It is instruction-tuned and much easier to use than this v1 model.
# Check out Phind-CodeLlama-34B-v2 [here](https://huggingface.co/Phind/Phind-CodeLlama-34B-v2).
## **Phind-CodeLlama-34B-v1**
We've fine-tuned CodeLlama-34B and CodeLlama-34B-Python on an internal Phind dataset that achieve 67.6% and 69.5% pass@1 on HumanEval, respectively. GPT-4 achieves 67%. We've applied OpenAI's decontamination methodology to our dataset to ensure result validity.
More details can be found on our [blog post](https://www.phind.com/blog/code-llama-beats-gpt4).
## Model Details
This model is fine-tuned from CodeLlama-34B and achieves 67.6% pass@1 on HumanEval.
## Dataset Details
We fined-tuned on a proprietary dataset of ~80k high quality programming problems and solutions. This dataset consists of instruction-answer pairs instead of code completion examples, making it structurally different from HumanEval. The Phind models were trained for 2 epochs, for a total of ~160k examples shown. LoRA was not used -- both models are a native finetune. We used DeepSpeed ZeRO 3 and Flash Attention 2 to train these models in three hours on 32 A100-80GB GPUs. We used a sequence length of 4096 tokens.
## How to Get Started with the Model
Make sure to install Transformers from the main git branch:
```bash
pip install git+https://github.com/huggingface/transformers.git
```
## How to Prompt the Model
**Please note that this model is somewhat instruction-tuned, but not chat-tuned.**
Do not try to use the Llama chat markup with this model. Instead, simply tell it what you want and add "\n: " at the end of your task.
For example:
```
Write me a linked list implementation: \n
```
## How to reproduce HumanEval Results
To reproduce our results:
```python
from transformers import AutoTokenizer, LlamaForCausalLM
from human_eval.data import write_jsonl, read_problems
from tqdm import tqdm
# initialize the model
model_path = "Phind/Phind-CodeLlama-34B-v1"
model = LlamaForCausalLM.from_pretrained(model_path, device_map="auto")
tokenizer = AutoTokenizer.from_pretrained(model_path)
# HumanEval helper
def generate_one_completion(prompt: str):
tokenizer.pad_token = tokenizer.eos_token
inputs = tokenizer(prompt, return_tensors="pt", truncation=True, max_length=4096)
# Generate
generate_ids = model.generate(inputs.input_ids.to("cuda"), max_new_tokens=256, do_sample=True, top_p=0.75, top_k=40, temperature=0.1)
completion = tokenizer.batch_decode(generate_ids, skip_special_tokens=True, clean_up_tokenization_spaces=False)[0]
completion = completion.replace(prompt, "").split("\n\n\n")[0]
return completion
# perform HumanEval
problems = read_problems()
num_samples_per_task = 1
samples = [
dict(task_id=task_id, completion=generate_one_completion(problems[task_id]["prompt"]))
for task_id in tqdm(problems)
for _ in range(num_samples_per_task)
]
write_jsonl("samples.jsonl", samples)
# run `evaluate_functional_correctness samples.jsonl` in your HumanEval code sandbox
```
## Bias, Risks, and Limitations
<!-- This section is meant to convey both technical and sociotechnical limitations. -->
This model has undergone very limited testing. Additional safety testing should be performed before any real-world deployments.
## Training details
<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
- **Hardware Type:** 32x A100-80GB
- **Hours used:** 90 GPU-hours
- **Cloud Provider:** AWS
- **Compute Region:** us-east-1 | 3,974 | [
[
-0.0211944580078125,
-0.051666259765625,
0.0167236328125,
0.0155487060546875,
-0.0249786376953125,
-0.0047760009765625,
-0.0014209747314453125,
-0.033172607421875,
-0.003505706787109375,
0.0242156982421875,
-0.03387451171875,
-0.053436279296875,
-0.021575927734375,
0.01496124267578125,
-0.0018701553344726562,
0.0841064453125,
0.01177215576171875,
-0.012115478515625,
-0.004543304443359375,
-0.007251739501953125,
-0.007213592529296875,
-0.044219970703125,
-0.039093017578125,
-0.011932373046875,
0.0298614501953125,
0.0273590087890625,
0.048492431640625,
0.06298828125,
0.052154541015625,
0.0211181640625,
-0.0126495361328125,
-0.0065765380859375,
-0.0413818359375,
-0.019775390625,
-0.0042877197265625,
-0.024139404296875,
-0.059814453125,
0.00577545166015625,
0.03326416015625,
0.018707275390625,
-0.01493072509765625,
0.03399658203125,
-0.0024871826171875,
0.026336669921875,
-0.036956787109375,
0.0144805908203125,
-0.0236968994140625,
0.01177215576171875,
-0.0030307769775390625,
-0.005397796630859375,
-0.0109710693359375,
-0.0322265625,
-0.0070343017578125,
-0.05560302734375,
0.0051116943359375,
0.0007495880126953125,
0.0709228515625,
0.03857421875,
-0.01605224609375,
0.0087432861328125,
-0.029937744140625,
0.0712890625,
-0.0772705078125,
0.0220947265625,
0.054046630859375,
0.00017547607421875,
-0.0122833251953125,
-0.059722900390625,
-0.039093017578125,
-0.0267486572265625,
0.0188140869140625,
0.00949859619140625,
-0.028076171875,
0.013092041015625,
0.049835205078125,
0.03497314453125,
-0.051513671875,
0.0042877197265625,
-0.051116943359375,
-0.030975341796875,
0.048187255859375,
0.026336669921875,
0.00864410400390625,
-0.03668212890625,
-0.045257568359375,
-0.018768310546875,
-0.035491943359375,
0.03448486328125,
0.01446533203125,
0.004390716552734375,
-0.032623291015625,
0.0310516357421875,
-0.035919189453125,
0.0413818359375,
0.0063323974609375,
-0.0231781005859375,
0.026397705078125,
-0.036468505859375,
-0.0401611328125,
-0.0139007568359375,
0.0675048828125,
0.022247314453125,
-0.003009796142578125,
0.0029773712158203125,
-0.01318359375,
0.00981903076171875,
0.0218353271484375,
-0.0743408203125,
-0.0367431640625,
0.031097412109375,
-0.029541015625,
-0.0440673828125,
-0.018035888671875,
-0.044769287109375,
-0.0196990966796875,
-0.0041351318359375,
0.03570556640625,
-0.0335693359375,
-0.0272216796875,
0.0282135009765625,
0.00809478759765625,
0.033172607421875,
0.0146942138671875,
-0.04840087890625,
0.01241302490234375,
0.0428466796875,
0.058929443359375,
0.00640869140625,
-0.0206756591796875,
-0.044769287109375,
-0.01224517822265625,
-0.0121002197265625,
0.04925537109375,
-0.0203857421875,
-0.026275634765625,
-0.025421142578125,
0.00482177734375,
0.0102691650390625,
-0.032684326171875,
0.04132080078125,
-0.037261962890625,
0.01459503173828125,
-0.032073974609375,
-0.026092529296875,
-0.03564453125,
0.023193359375,
-0.034942626953125,
0.08428955078125,
0.0257568359375,
-0.07501220703125,
-0.0016307830810546875,
-0.050811767578125,
-0.019775390625,
-0.0084686279296875,
0.0039825439453125,
-0.043304443359375,
-0.022552490234375,
0.0084075927734375,
0.037872314453125,
-0.024383544921875,
0.0203399658203125,
-0.030487060546875,
-0.04229736328125,
0.0115966796875,
-0.01483154296875,
0.085205078125,
0.0214080810546875,
-0.0465087890625,
0.0251312255859375,
-0.061492919921875,
0.0006527900695800781,
0.0087738037109375,
-0.04083251953125,
0.0022754669189453125,
-0.031494140625,
0.0200958251953125,
0.029541015625,
0.0148773193359375,
-0.037567138671875,
0.0258026123046875,
-0.0357666015625,
0.03167724609375,
0.06512451171875,
0.00817108154296875,
0.0271759033203125,
-0.057403564453125,
0.058807373046875,
-0.0012683868408203125,
0.040374755859375,
0.0154876708984375,
-0.05072021484375,
-0.05853271484375,
-0.02398681640625,
0.00940704345703125,
0.053955078125,
-0.026092529296875,
0.04241943359375,
-0.0007123947143554688,
-0.04498291015625,
-0.036346435546875,
0.0052642822265625,
0.037017822265625,
0.0543212890625,
0.0379638671875,
-0.0218658447265625,
-0.046539306640625,
-0.06597900390625,
0.0214996337890625,
-0.0188751220703125,
-0.006168365478515625,
0.008026123046875,
0.051300048828125,
-0.04132080078125,
0.055908203125,
-0.0228271484375,
-0.00848388671875,
-0.00013875961303710938,
-0.00806427001953125,
0.03485107421875,
0.058349609375,
0.0308990478515625,
-0.031646728515625,
-0.003803253173828125,
-0.0210113525390625,
-0.06231689453125,
0.01448822021484375,
-0.007030487060546875,
-0.0287322998046875,
0.0135650634765625,
0.015411376953125,
-0.038909912109375,
0.03680419921875,
0.0216064453125,
-0.0202178955078125,
0.046478271484375,
-0.0111541748046875,
0.01239013671875,
-0.09210205078125,
0.0261993408203125,
-0.001094818115234375,
-0.01019287109375,
-0.027679443359375,
0.0152130126953125,
0.0023174285888671875,
-0.0020809173583984375,
-0.040283203125,
0.031494140625,
-0.0189208984375,
-0.00022327899932861328,
-0.0015058517456054688,
-0.005413055419921875,
0.00801849365234375,
0.06964111328125,
-0.01451873779296875,
0.06024169921875,
0.037628173828125,
-0.052703857421875,
0.046051025390625,
0.0274200439453125,
-0.0166473388671875,
0.0109710693359375,
-0.0728759765625,
0.0247039794921875,
0.0199737548828125,
0.0146484375,
-0.06396484375,
-0.0212860107421875,
0.042205810546875,
-0.0343017578125,
0.0020599365234375,
-0.00821685791015625,
-0.036407470703125,
-0.0311737060546875,
-0.042327880859375,
0.0343017578125,
0.06414794921875,
-0.022247314453125,
0.0167236328125,
0.02362060546875,
0.0010442733764648438,
-0.054931640625,
-0.04443359375,
-0.01312255859375,
-0.028106689453125,
-0.04693603515625,
0.0137481689453125,
-0.0219879150390625,
-0.01512908935546875,
-0.015350341796875,
-0.01493072509765625,
-0.0006384849548339844,
0.0220184326171875,
0.032562255859375,
0.0305633544921875,
-0.0139312744140625,
0.0004100799560546875,
-0.006618499755859375,
-0.013458251953125,
0.0185089111328125,
-0.01593017578125,
0.034912109375,
-0.038055419921875,
-0.01849365234375,
-0.053680419921875,
0.0133056640625,
0.046630859375,
-0.0333251953125,
0.034515380859375,
0.0648193359375,
-0.04925537109375,
0.014923095703125,
-0.04461669921875,
-0.0256805419921875,
-0.03741455078125,
0.0313720703125,
-0.01898193359375,
-0.043304443359375,
0.05438232421875,
0.01332855224609375,
0.0216522216796875,
0.04364013671875,
0.0285797119140625,
-0.005619049072265625,
0.06256103515625,
0.05511474609375,
-0.01300048828125,
0.0294647216796875,
-0.0596923828125,
-0.006992340087890625,
-0.07684326171875,
-0.0197296142578125,
-0.033416748046875,
-0.010711669921875,
-0.036163330078125,
-0.048126220703125,
0.037689208984375,
0.03460693359375,
-0.04595947265625,
0.03143310546875,
-0.059844970703125,
0.004405975341796875,
0.050811767578125,
0.030487060546875,
0.01806640625,
0.0122833251953125,
-0.013214111328125,
0.0034618377685546875,
-0.0623779296875,
-0.041534423828125,
0.087646484375,
0.036346435546875,
0.05615234375,
-0.0008568763732910156,
0.06500244140625,
0.01337432861328125,
0.021026611328125,
-0.04736328125,
0.03662109375,
0.0307769775390625,
-0.030426025390625,
-0.0205230712890625,
-0.018768310546875,
-0.063232421875,
0.0013589859008789062,
-0.00948333740234375,
-0.052886962890625,
0.02093505859375,
-0.004001617431640625,
-0.062286376953125,
0.021331787109375,
-0.0265655517578125,
0.06903076171875,
-0.004001617431640625,
-0.04638671875,
-0.0249481201171875,
-0.0499267578125,
0.038848876953125,
-0.0034542083740234375,
0.01026153564453125,
-0.01493072509765625,
0.00447845458984375,
0.0860595703125,
-0.038726806640625,
0.0645751953125,
-0.0010557174682617188,
-0.0169219970703125,
0.021484375,
-0.0038089752197265625,
0.035980224609375,
0.0283050537109375,
-0.0308837890625,
0.0217437744140625,
0.004032135009765625,
-0.0426025390625,
-0.01312255859375,
0.039764404296875,
-0.06378173828125,
-0.045806884765625,
-0.04583740234375,
-0.03765869140625,
0.004058837890625,
0.00830078125,
0.038238525390625,
0.046661376953125,
0.00609588623046875,
0.0006103515625,
0.06158447265625,
-0.036407470703125,
0.0310516357421875,
0.0268707275390625,
-0.005474090576171875,
-0.04705810546875,
0.053558349609375,
-0.010284423828125,
0.00925445556640625,
0.0095062255859375,
0.00949859619140625,
-0.03619384765625,
-0.02484130859375,
-0.040008544921875,
0.0232696533203125,
-0.04974365234375,
-0.046661376953125,
-0.0638427734375,
-0.0171966552734375,
-0.037445068359375,
-0.0068511962890625,
-0.026641845703125,
-0.0194091796875,
-0.03839111328125,
-0.006317138671875,
0.061798095703125,
0.04913330078125,
-0.026885986328125,
0.004589080810546875,
-0.06512451171875,
0.03521728515625,
0.024444580078125,
0.0187530517578125,
-0.01207733154296875,
-0.06402587890625,
-0.0224609375,
0.0301513671875,
-0.033050537109375,
-0.07586669921875,
0.0196990966796875,
-0.0026149749755859375,
0.03564453125,
0.034027099609375,
0.03302001953125,
0.05072021484375,
0.0035495758056640625,
0.05108642578125,
0.005340576171875,
-0.06707763671875,
0.0570068359375,
-0.042266845703125,
0.0213775634765625,
0.028045654296875,
0.0276947021484375,
-0.02984619140625,
-0.0308685302734375,
-0.058837890625,
-0.047088623046875,
0.054107666015625,
0.0196380615234375,
-0.00628662109375,
0.0118560791015625,
0.025054931640625,
-0.0199127197265625,
-0.00015974044799804688,
-0.058258056640625,
-0.01226043701171875,
-0.0164031982421875,
0.00267791748046875,
0.0068817138671875,
-0.0156402587890625,
-0.0171966552734375,
-0.05047607421875,
0.05682373046875,
0.00629425048828125,
0.04083251953125,
0.03143310546875,
-0.00986480712890625,
-0.0123443603515625,
0.0159912109375,
0.054046630859375,
0.048187255859375,
-0.03375244140625,
-0.0130615234375,
0.0191650390625,
-0.0367431640625,
0.01094818115234375,
0.024566650390625,
0.001453399658203125,
-0.01113128662109375,
0.0252227783203125,
0.051605224609375,
0.0130462646484375,
-0.03668212890625,
0.028167724609375,
-0.0038242340087890625,
-0.01129150390625,
-0.0196075439453125,
0.019500732421875,
0.01168060302734375,
0.034698486328125,
0.0166778564453125,
0.01273345947265625,
0.0208740234375,
-0.0247344970703125,
0.02191162109375,
0.01291656494140625,
-0.01190185546875,
-0.020294189453125,
0.07806396484375,
0.012420654296875,
-0.019317626953125,
0.0687255859375,
-0.0307464599609375,
-0.0496826171875,
0.078857421875,
0.036895751953125,
0.057769775390625,
-0.0181884765625,
0.00968170166015625,
0.0538330078125,
0.0275115966796875,
0.0002269744873046875,
0.04815673828125,
-0.019439697265625,
-0.032623291015625,
-0.0211944580078125,
-0.060699462890625,
-0.0288238525390625,
0.0106048583984375,
-0.0635986328125,
0.008514404296875,
-0.0484619140625,
-0.02459716796875,
-0.0196380615234375,
0.0274658203125,
-0.05621337890625,
0.007526397705078125,
-0.005077362060546875,
0.06707763671875,
-0.054107666015625,
0.06756591796875,
0.06280517578125,
-0.047821044921875,
-0.086669921875,
-0.0189208984375,
-0.0084381103515625,
-0.047821044921875,
0.03179931640625,
0.01113128662109375,
-0.00047206878662109375,
0.0142669677734375,
-0.043701171875,
-0.06109619140625,
0.0902099609375,
0.023712158203125,
-0.03741455078125,
-0.003093719482421875,
-0.007549285888671875,
0.03485107421875,
0.004405975341796875,
0.023956298828125,
0.03167724609375,
0.0311737060546875,
-0.01441192626953125,
-0.076416015625,
0.0244903564453125,
-0.046356201171875,
-0.0075531005859375,
-0.0140380859375,
-0.0491943359375,
0.07586669921875,
-0.05145263671875,
-0.004817962646484375,
0.0037975311279296875,
0.048095703125,
0.04364013671875,
0.0198516845703125,
0.02545166015625,
0.05206298828125,
0.07110595703125,
0.0070953369140625,
0.07110595703125,
-0.035980224609375,
0.051788330078125,
0.055877685546875,
-0.0034313201904296875,
0.064208984375,
0.0097808837890625,
-0.0306854248046875,
0.0287933349609375,
0.07080078125,
-0.0140533447265625,
0.033477783203125,
0.0280609130859375,
-0.02386474609375,
-0.0251922607421875,
-0.00975799560546875,
-0.06597900390625,
0.03167724609375,
0.0140380859375,
-0.00208282470703125,
-0.00691986083984375,
0.0177001953125,
0.003398895263671875,
-0.024200439453125,
-0.000820159912109375,
0.0421142578125,
0.004856109619140625,
-0.0253143310546875,
0.09259033203125,
0.01377105712890625,
0.0716552734375,
-0.023284912109375,
-0.01029205322265625,
-0.03485107421875,
0.0093231201171875,
-0.019775390625,
-0.0288238525390625,
0.01093292236328125,
0.0023136138916015625,
-0.0124664306640625,
-0.0084075927734375,
0.03363037109375,
-0.01446533203125,
-0.03619384765625,
0.0140380859375,
0.0177001953125,
0.0183868408203125,
-0.01543426513671875,
-0.05804443359375,
0.0125579833984375,
0.009307861328125,
-0.0205841064453125,
0.0111541748046875,
0.0044403076171875,
0.0088043212890625,
0.05413818359375,
0.05413818359375,
-0.016845703125,
0.0094146728515625,
-0.0288543701171875,
0.0753173828125,
-0.055572509765625,
-0.03704833984375,
-0.04833984375,
0.039703369140625,
0.00860595703125,
-0.03863525390625,
0.04302978515625,
0.049072265625,
0.06610107421875,
-0.010955810546875,
0.04400634765625,
0.0034656524658203125,
0.01070404052734375,
-0.026153564453125,
0.06378173828125,
-0.04315185546875,
0.0255584716796875,
-0.0291900634765625,
-0.05279541015625,
0.00241851806640625,
0.07891845703125,
-0.0070953369140625,
-0.0004858970642089844,
0.043914794921875,
0.07745361328125,
-0.0097808837890625,
-0.0014142990112304688,
0.017242431640625,
0.015960693359375,
0.01450347900390625,
0.0694580078125,
0.06024169921875,
-0.06585693359375,
0.03765869140625,
-0.058349609375,
-0.024749755859375,
-0.01264190673828125,
-0.042999267578125,
-0.06146240234375,
-0.036285400390625,
-0.04083251953125,
-0.0545654296875,
0.00574493408203125,
0.081787109375,
0.059356689453125,
-0.053955078125,
-0.012420654296875,
0.0014247894287109375,
0.01422119140625,
-0.014190673828125,
-0.0169219970703125,
0.03350830078125,
-0.01059722900390625,
-0.046539306640625,
0.0197906494140625,
0.0079498291015625,
0.004058837890625,
-0.006412506103515625,
-0.0133209228515625,
-0.01617431640625,
-0.0109100341796875,
0.026153564453125,
0.0307159423828125,
-0.05859375,
-0.01424407958984375,
0.0212249755859375,
-0.04425048828125,
0.01302337646484375,
0.027252197265625,
-0.06719970703125,
0.010528564453125,
0.036346435546875,
0.0426025390625,
0.02886962890625,
-0.007274627685546875,
0.021575927734375,
-0.0209808349609375,
0.02142333984375,
0.025482177734375,
0.0241241455078125,
0.006687164306640625,
-0.045318603515625,
0.040496826171875,
0.0231781005859375,
-0.050811767578125,
-0.06494140625,
-0.0020122528076171875,
-0.09173583984375,
0.002651214599609375,
0.09149169921875,
-0.0104827880859375,
-0.018890380859375,
-0.01303863525390625,
-0.0180816650390625,
0.050079345703125,
-0.0285797119140625,
0.06207275390625,
0.0150146484375,
-0.007171630859375,
-0.00445556640625,
-0.043975830078125,
0.046539306640625,
0.038665771484375,
-0.068603515625,
-0.017730712890625,
0.043487548828125,
0.041595458984375,
-0.0008983612060546875,
0.0699462890625,
-0.0017652511596679688,
0.0428466796875,
-0.0028934478759765625,
0.0245208740234375,
-0.0197906494140625,
-0.0027751922607421875,
-0.037933349609375,
-0.0061492919921875,
-0.00928497314453125,
-0.036895751953125
]
] |
neulab/codebert-cpp | 2023-02-27T20:56:25.000Z | [
"transformers",
"pytorch",
"roberta",
"fill-mask",
"arxiv:2302.05527",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] | fill-mask | neulab | null | null | neulab/codebert-cpp | 8 | 5,960 | transformers | 2022-09-26T13:50:02 | This is a `microsoft/codebert-base-mlm` model, trained for 1,000,000 steps (with `batch_size=32`) on **C++** code from the `codeparrot/github-code-clean` dataset, on the masked-language-modeling task.
It is intended to be used in CodeBERTScore: [https://github.com/neulab/code-bert-score](https://github.com/neulab/code-bert-score), but can be used for any other model or task.
For more information, see: [https://github.com/neulab/code-bert-score](https://github.com/neulab/code-bert-score)
## Citation
If you use this model for research, please cite:
```
@article{zhou2023codebertscore,
url = {https://arxiv.org/abs/2302.05527},
author = {Zhou, Shuyan and Alon, Uri and Agarwal, Sumit and Neubig, Graham},
title = {CodeBERTScore: Evaluating Code Generation with Pretrained Models of Code},
publisher = {arXiv},
year = {2023},
}
``` | 850 | [
[
-0.01450347900390625,
-0.041412353515625,
-0.004119873046875,
0.037384033203125,
-0.006378173828125,
-0.0004360675811767578,
-0.012664794921875,
-0.01499176025390625,
0.00641632080078125,
0.047332763671875,
-0.04168701171875,
-0.053253173828125,
-0.0302581787109375,
-0.01169586181640625,
-0.043701171875,
0.09735107421875,
0.013824462890625,
0.029205322265625,
-0.001979827880859375,
-0.0018787384033203125,
-0.0298309326171875,
-0.051025390625,
-0.030487060546875,
-0.0117950439453125,
0.037017822265625,
0.013458251953125,
0.039154052734375,
0.03466796875,
0.0280914306640625,
0.007022857666015625,
0.01236724853515625,
-0.01465606689453125,
-0.053680419921875,
-0.012054443359375,
0.0206756591796875,
-0.039703369140625,
-0.0687255859375,
0.025238037109375,
0.022247314453125,
0.05560302734375,
0.00799560546875,
0.04779052734375,
0.0140533447265625,
0.06024169921875,
-0.032257080078125,
0.006805419921875,
-0.051116943359375,
0.0009784698486328125,
0.024169921875,
0.0073394775390625,
-0.0413818359375,
-0.03399658203125,
-0.0002359151840209961,
-0.0157318115234375,
0.0268707275390625,
-0.004730224609375,
0.072509765625,
0.0017852783203125,
-0.01100921630859375,
-0.022705078125,
-0.043548583984375,
0.0589599609375,
-0.046600341796875,
0.008941650390625,
0.024322509765625,
0.014190673828125,
0.0100250244140625,
-0.061767578125,
-0.0193634033203125,
-0.032806396484375,
0.01922607421875,
-0.0220184326171875,
-0.0270843505859375,
-0.01424407958984375,
0.049774169921875,
0.0013637542724609375,
-0.06524658203125,
-0.0142822265625,
-0.072265625,
-0.018157958984375,
0.040740966796875,
0.013153076171875,
-0.0082550048828125,
-0.004070281982421875,
-0.051361083984375,
0.0046234130859375,
-0.056884765625,
0.0213775634765625,
0.040557861328125,
0.0184783935546875,
-0.0186309814453125,
0.0301361083984375,
0.002262115478515625,
0.06500244140625,
-0.0094146728515625,
0.0161590576171875,
0.04718017578125,
-0.0129241943359375,
-0.037017822265625,
-0.0046844482421875,
0.05718994140625,
0.0207061767578125,
0.05548095703125,
-0.00518798828125,
-0.024322509765625,
-0.0024204254150390625,
0.038818359375,
-0.0784912109375,
-0.055328369140625,
0.0225067138671875,
-0.039794921875,
-0.036041259765625,
0.029052734375,
-0.0163726806640625,
0.002529144287109375,
-0.020904541015625,
0.03558349609375,
-0.0212554931640625,
-0.0140228271484375,
0.0043792724609375,
0.0127105712890625,
0.022857666015625,
0.0160064697265625,
-0.0268096923828125,
0.003948211669921875,
0.041717529296875,
0.060455322265625,
-0.005481719970703125,
-0.0178680419921875,
-0.03125,
-0.04327392578125,
-0.046905517578125,
0.0132293701171875,
-0.02215576171875,
-0.017333984375,
0.0006504058837890625,
0.01611328125,
-0.0126800537109375,
-0.049896240234375,
0.0227508544921875,
-0.058746337890625,
0.0082244873046875,
0.005802154541015625,
-0.038330078125,
-0.020904541015625,
0.0272216796875,
-0.0611572265625,
0.079345703125,
0.024078369140625,
-0.018524169921875,
0.035003662109375,
-0.055023193359375,
-0.0026092529296875,
0.0207366943359375,
-0.009063720703125,
-0.034912109375,
-0.0087432861328125,
-0.0131683349609375,
0.034088134765625,
0.007244110107421875,
0.045989990234375,
-0.01788330078125,
-0.04766845703125,
0.0246124267578125,
-0.0325927734375,
0.07476806640625,
0.0309600830078125,
-0.0278472900390625,
0.007232666015625,
-0.07672119140625,
0.0211029052734375,
-0.004974365234375,
-0.037261962890625,
0.004726409912109375,
-0.0248870849609375,
0.0276947021484375,
0.03582763671875,
0.0384521484375,
-0.0423583984375,
0.033203125,
-0.015838623046875,
0.0201416015625,
0.04486083984375,
-0.01263427734375,
0.02520751953125,
-0.039031982421875,
0.061431884765625,
-0.005458831787109375,
0.016693115234375,
-0.0245208740234375,
-0.050323486328125,
-0.06439208984375,
-0.02349853515625,
0.056976318359375,
0.0293121337890625,
-0.033355712890625,
0.051300048828125,
-0.018951416015625,
-0.046234130859375,
-0.044403076171875,
0.0159912109375,
0.025115966796875,
0.016845703125,
0.0280303955078125,
-0.0230712890625,
-0.04302978515625,
-0.05975341796875,
-0.0178680419921875,
0.0124664306640625,
-0.01806640625,
0.0059967041015625,
0.06982421875,
-0.0312347412109375,
0.07574462890625,
-0.031982421875,
-0.0179290771484375,
-0.009185791015625,
0.0179443359375,
0.0467529296875,
0.0675048828125,
0.0399169921875,
-0.04718017578125,
-0.0389404296875,
-0.03936767578125,
-0.044952392578125,
0.0079193115234375,
-0.01551055908203125,
-0.01322174072265625,
0.02398681640625,
0.04193115234375,
-0.01146697998046875,
0.038543701171875,
0.06903076171875,
-0.02532958984375,
0.04779052734375,
-0.004764556884765625,
0.01444244384765625,
-0.06817626953125,
0.02618408203125,
-0.0017795562744140625,
-0.0309600830078125,
-0.0440673828125,
-0.002857208251953125,
0.0172576904296875,
-0.027557373046875,
-0.040496826171875,
0.017791748046875,
-0.0248565673828125,
0.0203704833984375,
-0.02508544921875,
-0.026397705078125,
-0.002895355224609375,
0.0654296875,
-0.0182952880859375,
0.0469970703125,
0.043701171875,
-0.045867919921875,
0.0131072998046875,
-0.0004239082336425781,
-0.0482177734375,
-0.00490570068359375,
-0.06890869140625,
0.03143310546875,
0.023040771484375,
0.00396728515625,
-0.0716552734375,
0.0029506683349609375,
0.02264404296875,
-0.056884765625,
0.00878143310546875,
-0.0180206298828125,
-0.055023193359375,
-0.004329681396484375,
-0.0182647705078125,
0.04437255859375,
0.048858642578125,
-0.0249786376953125,
0.0270233154296875,
0.0142822265625,
0.009735107421875,
-0.038055419921875,
-0.05126953125,
-0.0015964508056640625,
0.00016415119171142578,
-0.044097900390625,
-0.0008740425109863281,
-0.0211944580078125,
0.00626373291015625,
-0.0049591064453125,
-0.0104522705078125,
-0.0093994140625,
-0.0024356842041015625,
0.043609619140625,
0.03173828125,
-0.01502227783203125,
0.022552490234375,
-0.026397705078125,
-0.003833770751953125,
0.02630615234375,
-0.01483917236328125,
0.05682373046875,
-0.01177215576171875,
-0.0193634033203125,
-0.00598907470703125,
0.01499176025390625,
0.037353515625,
-0.00940704345703125,
0.0784912109375,
0.040771484375,
-0.03216552734375,
-0.032623291015625,
-0.0238494873046875,
-0.011016845703125,
-0.0300445556640625,
0.0333251953125,
-0.01395416259765625,
-0.055908203125,
0.030059814453125,
0.0059967041015625,
0.004894256591796875,
0.036712646484375,
0.04791259765625,
-0.0031070709228515625,
0.059417724609375,
0.04351806640625,
-0.030029296875,
0.032745361328125,
-0.04608154296875,
0.01142120361328125,
-0.0355224609375,
-0.024017333984375,
-0.046112060546875,
-0.0224761962890625,
-0.048095703125,
-0.0472412109375,
0.013275146484375,
0.02294921875,
-0.046417236328125,
0.054840087890625,
-0.039581298828125,
0.0161285400390625,
0.05450439453125,
0.024322509765625,
0.00919342041015625,
0.009368896484375,
-0.0184783935546875,
0.00806427001953125,
-0.057647705078125,
-0.054168701171875,
0.1009521484375,
0.035400390625,
0.06396484375,
-0.007472991943359375,
0.055389404296875,
0.03643798828125,
0.033172607421875,
-0.029449462890625,
0.036712646484375,
0.0016145706176757812,
-0.06500244140625,
-0.0002512931823730469,
-0.038726806640625,
-0.08111572265625,
-0.0080413818359375,
-0.01132965087890625,
-0.04595947265625,
-0.00238037109375,
0.0168914794921875,
-0.01131439208984375,
0.005680084228515625,
-0.052947998046875,
0.07879638671875,
-0.0144500732421875,
-0.0021514892578125,
0.006572723388671875,
-0.03814697265625,
0.023223876953125,
-0.0210418701171875,
0.0252685546875,
-0.0064239501953125,
0.01006317138671875,
0.06982421875,
-0.0244140625,
0.059234619140625,
-0.0228424072265625,
-0.00868988037109375,
0.0204925537109375,
0.00787353515625,
0.040191650390625,
-0.01546478271484375,
-0.0085906982421875,
0.044891357421875,
-0.009429931640625,
-0.03948974609375,
-0.01708984375,
0.0628662109375,
-0.047210693359375,
-0.004856109619140625,
-0.03643798828125,
-0.043853759765625,
0.00970458984375,
0.004665374755859375,
0.021087646484375,
0.04058837890625,
0.0022125244140625,
0.043701171875,
0.041595458984375,
-0.029266357421875,
0.0252685546875,
0.0423583984375,
-0.026123046875,
-0.0168914794921875,
0.0711669921875,
0.004505157470703125,
0.0303192138671875,
0.01343536376953125,
-0.026123046875,
0.004497528076171875,
-0.0234222412109375,
-0.01336669921875,
0.00240325927734375,
-0.05059814453125,
-0.0161285400390625,
-0.044830322265625,
-0.0343017578125,
-0.027801513671875,
-0.00392913818359375,
-0.0201568603515625,
-0.0145263671875,
-0.0330810546875,
-0.0002446174621582031,
0.013214111328125,
0.03533935546875,
0.012664794921875,
-0.00038433074951171875,
-0.0643310546875,
0.0169525146484375,
-0.003910064697265625,
0.029327392578125,
-0.005340576171875,
-0.0531005859375,
-0.055633544921875,
0.01412200927734375,
-0.0194244384765625,
-0.0537109375,
0.04327392578125,
-0.0083465576171875,
0.047332763671875,
0.004810333251953125,
0.00901031494140625,
0.02899169921875,
-0.024505615234375,
0.05487060546875,
0.01322174072265625,
-0.059844970703125,
0.0308380126953125,
-0.01361083984375,
0.051910400390625,
0.04595947265625,
0.051910400390625,
-0.00589752197265625,
-0.039398193359375,
-0.054962158203125,
-0.07696533203125,
0.04254150390625,
0.0343017578125,
0.027130126953125,
0.014801025390625,
0.00411224365234375,
-0.0030536651611328125,
0.037445068359375,
-0.0936279296875,
-0.037200927734375,
0.0028743743896484375,
-0.0177459716796875,
-0.005962371826171875,
-0.01465606689453125,
-0.00792694091796875,
-0.026214599609375,
0.06231689453125,
-0.0017633438110351562,
0.028228759765625,
-0.0079193115234375,
-0.038116455078125,
-0.0026874542236328125,
0.00008058547973632812,
0.06396484375,
0.064208984375,
-0.039459228515625,
-0.02294921875,
-0.0140533447265625,
-0.047637939453125,
-0.01476287841796875,
0.00807952880859375,
0.009735107421875,
0.007541656494140625,
0.0372314453125,
0.046844482421875,
0.01219940185546875,
-0.060638427734375,
0.045928955078125,
0.00806427001953125,
-0.04931640625,
-0.036712646484375,
0.02081298828125,
0.006923675537109375,
0.027435302734375,
0.043365478515625,
0.032745361328125,
-0.00501251220703125,
-0.016754150390625,
0.035003662109375,
0.0203857421875,
-0.06396484375,
0.00016188621520996094,
0.0538330078125,
0.00807952880859375,
-0.03369140625,
0.04949951171875,
-0.0306243896484375,
-0.0347900390625,
0.064697265625,
0.004894256591796875,
0.060791015625,
0.0189666748046875,
-0.00377655029296875,
0.04638671875,
0.0389404296875,
0.01293182373046875,
0.0268707275390625,
0.001434326171875,
-0.041748046875,
-0.0230712890625,
-0.06561279296875,
-0.007656097412109375,
0.0252685546875,
-0.05438232421875,
0.0239410400390625,
-0.01461029052734375,
-0.00728607177734375,
-0.00008606910705566406,
0.002101898193359375,
-0.06475830078125,
0.0205841064453125,
0.00887298583984375,
0.084228515625,
-0.05181884765625,
0.0870361328125,
0.0347900390625,
-0.035980224609375,
-0.07568359375,
-0.004673004150390625,
-0.0191650390625,
-0.0645751953125,
0.077392578125,
0.0202484130859375,
0.00998687744140625,
0.0015964508056640625,
-0.064453125,
-0.0233612060546875,
0.06878662109375,
0.0191802978515625,
-0.062286376953125,
0.005374908447265625,
0.007556915283203125,
0.0443115234375,
-0.05218505859375,
0.00705718994140625,
0.0194244384765625,
-0.0030384063720703125,
-0.0093536376953125,
-0.056884765625,
-0.00861358642578125,
-0.035308837890625,
-0.00635528564453125,
-0.003265380859375,
-0.00795745849609375,
0.09942626953125,
-0.0191802978515625,
0.01372528076171875,
0.0197601318359375,
0.01715087890625,
0.04345703125,
0.0090789794921875,
0.039886474609375,
0.0223846435546875,
0.0203399658203125,
0.0008301734924316406,
0.0555419921875,
-0.06256103515625,
0.07342529296875,
0.08282470703125,
0.004261016845703125,
0.052154541015625,
0.014495849609375,
-0.0206298828125,
0.0496826171875,
0.040496826171875,
-0.04449462890625,
0.0241241455078125,
0.05010986328125,
-0.0041656494140625,
-0.00969696044921875,
0.033447265625,
-0.041351318359375,
0.0173797607421875,
0.0028667449951171875,
-0.07672119140625,
-0.005153656005859375,
-0.004650115966796875,
0.007343292236328125,
-0.0146636962890625,
-0.023284912109375,
0.0343017578125,
0.0023365020751953125,
-0.037384033203125,
0.07391357421875,
-0.0031032562255859375,
0.0465087890625,
-0.046356201171875,
-0.0260009765625,
-0.006587982177734375,
0.0183563232421875,
-0.0063629150390625,
-0.02911376953125,
-0.014007568359375,
0.02801513671875,
-0.02532958984375,
-0.026275634765625,
0.034027099609375,
-0.03570556640625,
-0.0621337890625,
0.023834228515625,
0.028045654296875,
0.034515380859375,
-0.0214691162109375,
-0.07025146484375,
0.003673553466796875,
0.0140533447265625,
-0.0237274169921875,
0.0284881591796875,
0.007228851318359375,
0.0157318115234375,
0.046600341796875,
0.053131103515625,
-0.0078887939453125,
0.0145111083984375,
0.01678466796875,
0.06500244140625,
-0.05670166015625,
-0.02398681640625,
-0.0552978515625,
0.05120849609375,
0.009918212890625,
-0.037139892578125,
0.06097412109375,
0.07379150390625,
0.067138671875,
-0.033905029296875,
0.049896240234375,
-0.00830078125,
0.0218048095703125,
-0.02825927734375,
0.060028076171875,
-0.032989501953125,
0.0243377685546875,
-0.0303192138671875,
-0.071533203125,
-0.00847625732421875,
0.03936767578125,
0.0005273818969726562,
0.0308990478515625,
0.037841796875,
0.08477783203125,
0.0159454345703125,
-0.0152130126953125,
0.0394287109375,
0.00415802001953125,
0.01513671875,
0.042938232421875,
0.03936767578125,
-0.05303955078125,
0.054473876953125,
-0.00820159912109375,
-0.0128021240234375,
-0.0236053466796875,
-0.0550537109375,
-0.067626953125,
-0.036529541015625,
-0.03216552734375,
-0.0572509765625,
-0.0135498046875,
0.079833984375,
0.06915283203125,
-0.08056640625,
-0.032440185546875,
-0.0182342529296875,
-0.0165557861328125,
-0.0170440673828125,
-0.0206146240234375,
0.00811767578125,
-0.03216552734375,
-0.0574951171875,
0.00214385986328125,
-0.004810333251953125,
-0.0183258056640625,
-0.036956787109375,
-0.0130157470703125,
-0.026275634765625,
-0.01544952392578125,
0.0301361083984375,
-0.0019054412841796875,
-0.0301055908203125,
-0.0028438568115234375,
0.01885986328125,
-0.036529541015625,
0.006954193115234375,
0.060211181640625,
-0.05755615234375,
0.041595458984375,
0.0182952880859375,
0.0181121826171875,
0.035400390625,
0.0016117095947265625,
0.044525146484375,
-0.061920166015625,
-0.00579833984375,
0.020477294921875,
0.018798828125,
0.0016222000122070312,
-0.0104217529296875,
0.035675048828125,
0.0131683349609375,
-0.046234130859375,
-0.057708740234375,
-0.00766754150390625,
-0.0833740234375,
-0.0191497802734375,
0.0980224609375,
-0.02728271484375,
0.0021190643310546875,
-0.00302886962890625,
-0.00875091552734375,
0.005611419677734375,
-0.034637451171875,
0.031158447265625,
0.040740966796875,
0.0025844573974609375,
-0.0088958740234375,
-0.0626220703125,
0.026092529296875,
0.0013599395751953125,
-0.04345703125,
-0.0266876220703125,
0.02130126953125,
0.043609619140625,
0.01800537109375,
0.0421142578125,
0.0052947998046875,
0.024383544921875,
0.01151275634765625,
0.036041259765625,
-0.0244140625,
-0.031585693359375,
-0.036651611328125,
0.034210205078125,
-0.0022792816162109375,
-0.043914794921875
]
] |
CalderaAI/13B-Thorns-l2 | 2023-09-06T11:04:31.000Z | [
"transformers",
"pytorch",
"llama",
"text-generation",
"alpaca",
"cot",
"vicuna",
"uncensored",
"merge",
"mix",
"endpoints_compatible",
"has_space",
"text-generation-inference",
"region:us"
] | text-generation | CalderaAI | null | null | CalderaAI/13B-Thorns-l2 | 6 | 5,953 | transformers | 2023-09-06T03:47:04 | ---
tags:
- llama
- alpaca
- cot
- vicuna
- uncensored
- merge
- mix
---
## 13B-Thorns [An Instruct Based LLaMAv2-13B Ensemble Merge | Alpaca Format]
# WARNING - This Model Is Uncensored And Has Not Been Fully Tested For Toxicity. This Is A Research Artifact Intended For Responsible Use. May Generate Offensive And Misleading Content. Do Not Treat Language Sythesized By This Research Artifact As Advice Or As Factual In Any Domain. CalderaAI Strictly Does Not Condone Use Of This Release Outside The Domain Of Research Or Entertainment.
# Composition:
13B-Thorns-l2 utilizes a new merge method called Spherical Linear Interpolation. By merging data as a spherical vector store concept, a combined pair of models have a smoother transition between feature spaces that are characteristic of each model, resulting in a more coherent fusion of both model's unique strengths.
## Our implementation of Spherical Linear Interpolation for LLM merging: https://github.com/Digitous/LLM-SLERP-Merge
## Note: Skip to the TL;DR section for the finalized design this model is comprised of.
Thorns' design is based on the concept of purposed segmentation, in this case we have two-
--Logic Segment (MK1):
Fine-Tuned parent models were hand selected and reviewed for datasets, performance, least restrictive censorship, and community perception of coherence and utility. Ultimately we decided on four models to merge in pairs of two, then combine those offspring for a quad merged logic cluster.
All four models were merged using the SLERP method. Yes the name is annoyingly funny. SLERP.
--Creativity and Imagination Segment (MK1):
Flawed first approach (a takeaway on LoRAs);
We then decided the creativity and imagination segment could be as simple as one model, especially if its dataset design, tagging, training quality, and proven track record is above and beyond. KoboldAI's Holodeck model is the result of a dataset that is years of collected, organized, tagged, deduped, and cleaned data. Holodeck alone would be beyond sufficient for the segment we view as the 'subconscious' segment of the model ensemble, however we applied the LIMA RP PEFT to it for extended variety of a different kind.
That's where we got carried away. LoRAs offer unique augmentation to model merge possibilities, and the decision was made to take the result of that segment and add two more LoRAs to see if they further extended Holodeck, settling on Kimiko and Janine; two very different RP and conversational LoRAs.
This was a bad move, as when we SLERP merged that version of the imagination segment to the logic segment the result was a ranting mess that followed instructions but was the equivalent of a child scribbling all over the place and ignoring obvious chains of logic and a mushy amalgam of awkward creative behavior that had no semblance of coherency.
The composite model was slated to be named 13B-Astronomicon; after all the work that went into it and the flatly bland result, the name was abandoned and the next move, which is a byproduct experiment of Astronomicon is what became Thorn.. because this project became a thorn in our side.
Because pain is fun, and persistence in design iteration is the only way forward, we reworked our approach to both segment ensembles following one idea - all three Roleplay and Conversational LoRAs stay no matter what because sure why not add arbitrary rules to the redesign phase at this point.
## TL;DR Section
--Finalized Logic and Creativity Segments (MK2):
After a few meetings with our top teams of model hacking memegineers we drafted Thorns MK2, which was prompty fast tracked for production by the Roko's Basilisk Shadow Council.
..Actually I just redid the merge like this:
```
-Model Merge Ensemble Key-
{} = SLERP Merge | [] = PEFT Merge | () = Composite Model
({({NousHermes+Chronos}[Kimiko])+({Platupus+AiroborosM2.0}[Janine])}{Holodeck[LIMA RP]})
```
## Findings:
-Strategically fusing LoRAs to models that stand to gain the most from them and then merging the result into the ensemble is exceptionally effective.
-Stacking the exact same LoRAs onto one model then merging that into the ensemble results in noisy garbage.
## Language Models and LoRAs Used Credits:
All models and adapters used are LLaMAv2-13B.
# Models:
Nous-Hermes
Chronos
Platypus
Airoboros
Holodeck
# Adapters:
Kimiko
Janine
LIMA RP
Also thanks to Meta for LLaMAv2 and deciding to allow the research community at large to benefit from their incredible work.
Each model and LoRA was hand picked and considered for what it could contribute to this ensemble.
Thanks to each and every one of you for your incredible work developing some of the best things
to come out of this community. | 4,749 | [
[
-0.04339599609375,
-0.041290283203125,
0.03375244140625,
0.0183868408203125,
-0.04266357421875,
0.0170440673828125,
0.007465362548828125,
-0.0745849609375,
0.052581787109375,
0.058990478515625,
-0.0631103515625,
-0.03179931640625,
-0.03997802734375,
-0.016632080078125,
-0.039337158203125,
0.0885009765625,
0.01488494873046875,
-0.0298309326171875,
0.0012607574462890625,
-0.013519287109375,
-0.033782958984375,
-0.026031494140625,
-0.0419921875,
-0.03192138671875,
0.0540771484375,
0.0228729248046875,
0.042999267578125,
0.04315185546875,
0.051483154296875,
0.021392822265625,
-0.031097412109375,
0.036224365234375,
-0.046630859375,
0.005950927734375,
-0.0048065185546875,
-0.0280303955078125,
-0.06488037109375,
0.007175445556640625,
0.0294952392578125,
0.041290283203125,
-0.0096893310546875,
0.00675201416015625,
0.00821685791015625,
0.031982421875,
-0.0318603515625,
-0.0178680419921875,
-0.046783447265625,
0.0123443603515625,
-0.0217437744140625,
0.00020837783813476562,
-0.0269927978515625,
-0.01548004150390625,
-0.008026123046875,
-0.06280517578125,
-0.0118255615234375,
0.0022182464599609375,
0.06378173828125,
0.039764404296875,
-0.014312744140625,
-0.01055145263671875,
-0.061004638671875,
0.060882568359375,
-0.04986572265625,
0.00469970703125,
0.0019550323486328125,
0.0097198486328125,
-0.01122283935546875,
-0.054901123046875,
-0.0255279541015625,
0.0109710693359375,
0.020843505859375,
0.0290374755859375,
-0.03509521484375,
-0.0168609619140625,
0.0012950897216796875,
0.0307464599609375,
-0.033172607421875,
0.0300140380859375,
-0.06695556640625,
-0.0118255615234375,
0.061065673828125,
0.00655364990234375,
-0.0035648345947265625,
-0.00949859619140625,
-0.037017822265625,
-0.0231475830078125,
-0.04736328125,
-0.00940704345703125,
0.051971435546875,
0.0035648345947265625,
-0.01236724853515625,
0.0576171875,
0.01861572265625,
0.035888671875,
0.025238037109375,
-0.0195770263671875,
0.012298583984375,
-0.007030487060546875,
-0.01395416259765625,
0.0032672882080078125,
0.062255859375,
0.0302886962890625,
-0.012908935546875,
-0.00920867919921875,
0.00263214111328125,
0.036651611328125,
0.021759033203125,
-0.039337158203125,
-0.0215301513671875,
0.0310516357421875,
-0.0283355712890625,
-0.05615234375,
-0.007030487060546875,
-0.0595703125,
-0.026397705078125,
-0.01068115234375,
0.029022216796875,
-0.04559326171875,
-0.015228271484375,
0.00742340087890625,
-0.0240478515625,
-0.003009796142578125,
0.023681640625,
-0.04913330078125,
0.037841796875,
0.0662841796875,
0.0458984375,
-0.03521728515625,
-0.0269927978515625,
-0.002044677734375,
-0.0055999755859375,
-0.03948974609375,
0.06622314453125,
-0.020172119140625,
-0.049285888671875,
-0.04583740234375,
-0.01229095458984375,
0.0166168212890625,
-0.050811767578125,
0.0421142578125,
-0.034149169921875,
0.0258331298828125,
-0.013092041015625,
-0.0303802490234375,
-0.019866943359375,
-0.00299835205078125,
-0.041107177734375,
0.07562255859375,
0.0053253173828125,
-0.043670654296875,
0.0276947021484375,
-0.045745849609375,
-0.004703521728515625,
0.0086822509765625,
-0.002162933349609375,
-0.0273284912109375,
0.00428009033203125,
0.0001857280731201172,
0.008819580078125,
-0.0306854248046875,
0.0168609619140625,
-0.03192138671875,
-0.0272369384765625,
0.0193023681640625,
-0.01428985595703125,
0.055572509765625,
0.0235443115234375,
-0.0145111083984375,
0.004932403564453125,
-0.044830322265625,
-0.0081939697265625,
0.011505126953125,
-0.00672149658203125,
-0.026580810546875,
-0.01279449462890625,
0.005046844482421875,
0.009765625,
0.0117645263671875,
-0.02984619140625,
0.036407470703125,
-0.038543701171875,
0.03948974609375,
0.06134033203125,
0.01499176025390625,
0.050750732421875,
-0.04010009765625,
0.0288238525390625,
0.01258087158203125,
0.02960205078125,
0.008392333984375,
-0.06317138671875,
-0.0694580078125,
-0.02099609375,
0.01488494873046875,
0.028076171875,
-0.0423583984375,
0.0467529296875,
-0.0009050369262695312,
-0.08203125,
-0.03485107421875,
-0.005542755126953125,
0.0277557373046875,
0.0268707275390625,
0.0197296142578125,
-0.035797119140625,
-0.052001953125,
-0.06622314453125,
0.006237030029296875,
-0.021209716796875,
0.02069091796875,
0.02880859375,
0.043060302734375,
-0.033599853515625,
0.05987548828125,
-0.042205810546875,
-0.025909423828125,
-0.0198211669921875,
0.004871368408203125,
0.04937744140625,
0.049102783203125,
0.059417724609375,
-0.0355224609375,
-0.0198974609375,
0.00714874267578125,
-0.056915283203125,
-0.00872039794921875,
0.00591278076171875,
-0.0297088623046875,
-0.003948211669921875,
0.01058197021484375,
-0.05462646484375,
0.03521728515625,
0.04571533203125,
-0.0243988037109375,
0.036468505859375,
-0.0037288665771484375,
0.00484466552734375,
-0.08489990234375,
0.0030059814453125,
-0.0255279541015625,
0.01442718505859375,
-0.04583740234375,
0.0197906494140625,
-0.0226287841796875,
-0.001720428466796875,
-0.03466796875,
0.04833984375,
-0.017303466796875,
-0.01021575927734375,
-0.02496337890625,
0.01425933837890625,
0.00641632080078125,
0.049835205078125,
-0.01995849609375,
0.034271240234375,
0.03369140625,
-0.0526123046875,
0.03759765625,
0.02288818359375,
-0.020477294921875,
0.045074462890625,
-0.050048828125,
0.0160675048828125,
-0.00970458984375,
0.0183868408203125,
-0.0648193359375,
-0.0282135009765625,
0.037841796875,
-0.0242462158203125,
0.0027923583984375,
-0.0158233642578125,
-0.0170745849609375,
-0.04541015625,
-0.0251922607421875,
0.04144287109375,
0.05670166015625,
-0.03802490234375,
0.054931640625,
0.045013427734375,
-0.01361083984375,
-0.037261962890625,
-0.062408447265625,
-0.003154754638671875,
-0.054779052734375,
-0.06988525390625,
0.06494140625,
-0.01474761962890625,
-0.0445556640625,
0.0019426345825195312,
-0.0004477500915527344,
-0.0021839141845703125,
-0.01056671142578125,
0.035369873046875,
0.04840087890625,
-0.005344390869140625,
-0.00833892822265625,
0.025909423828125,
-0.0040130615234375,
-0.01904296875,
-0.0252532958984375,
0.06475830078125,
-0.00228118896484375,
-0.039306640625,
-0.032745361328125,
0.010101318359375,
0.060821533203125,
-0.003643035888671875,
0.02569580078125,
0.0386962890625,
-0.0242462158203125,
0.0024013519287109375,
-0.06060791015625,
-0.0017518997192382812,
-0.0345458984375,
0.0182342529296875,
-0.01171875,
-0.06329345703125,
0.0537109375,
0.01548004150390625,
0.032379150390625,
0.053924560546875,
0.0216064453125,
-0.018096923828125,
0.042755126953125,
0.037384033203125,
-0.008392333984375,
0.0182647705078125,
-0.054779052734375,
0.0096588134765625,
-0.08258056640625,
-0.035675048828125,
-0.00945281982421875,
-0.05657958984375,
-0.034759521484375,
-0.044281005859375,
0.037017822265625,
0.009033203125,
-0.0034637451171875,
0.0506591796875,
-0.052947998046875,
0.032470703125,
0.028778076171875,
0.01580810546875,
0.0176849365234375,
0.0104522705078125,
0.0027217864990234375,
-0.017120361328125,
-0.056884765625,
-0.010528564453125,
0.0772705078125,
0.0384521484375,
0.07061767578125,
0.0160064697265625,
0.076416015625,
0.0163421630859375,
0.0028591156005859375,
-0.054962158203125,
0.05474853515625,
-0.0217742919921875,
-0.0335693359375,
-0.0187225341796875,
-0.0147705078125,
-0.07757568359375,
0.0300140380859375,
-0.0171356201171875,
-0.06756591796875,
0.048828125,
0.01177978515625,
-0.06201171875,
0.021697998046875,
-0.03765869140625,
0.04736328125,
-0.01580810546875,
-0.03802490234375,
-0.0286712646484375,
-0.035888671875,
0.06427001953125,
-0.01366424560546875,
0.0239105224609375,
-0.0019521713256835938,
0.01503753662109375,
0.07763671875,
-0.03839111328125,
0.056671142578125,
0.0134735107421875,
-0.01287078857421875,
0.053314208984375,
0.0215301513671875,
0.033935546875,
0.0018510818481445312,
0.01235198974609375,
0.0142059326171875,
-0.00485992431640625,
0.0007982254028320312,
-0.031768798828125,
0.07049560546875,
-0.07867431640625,
-0.02740478515625,
-0.035064697265625,
-0.040130615234375,
0.0037384033203125,
-0.018707275390625,
0.04541015625,
0.034698486328125,
-0.0161590576171875,
0.0109710693359375,
0.041473388671875,
-0.0028705596923828125,
0.025543212890625,
0.01373291015625,
-0.053253173828125,
-0.057861328125,
0.059661865234375,
0.0019855499267578125,
0.01299285888671875,
-0.004726409912109375,
0.00044155120849609375,
-0.00499725341796875,
-0.00273895263671875,
-0.004253387451171875,
0.041900634765625,
-0.048431396484375,
-0.01092529296875,
-0.038330078125,
-0.032989501953125,
-0.037689208984375,
-0.00949859619140625,
-0.036712646484375,
-0.057342529296875,
-0.033355712890625,
0.0008544921875,
0.028350830078125,
0.06890869140625,
-0.0220184326171875,
0.0413818359375,
-0.04766845703125,
0.0025501251220703125,
0.017242431640625,
-0.011993408203125,
-0.001956939697265625,
-0.0716552734375,
0.0035877227783203125,
-0.00783538818359375,
-0.040374755859375,
-0.0655517578125,
0.046142578125,
-0.006809234619140625,
0.02459716796875,
0.041290283203125,
-0.025360107421875,
0.0714111328125,
-0.01503753662109375,
0.0540771484375,
0.0226593017578125,
-0.047760009765625,
0.0361328125,
-0.03326416015625,
0.008697509765625,
0.025665283203125,
0.005161285400390625,
-0.035888671875,
-0.01226043701171875,
-0.08367919921875,
-0.06854248046875,
0.06298828125,
0.035491943359375,
-0.024444580078125,
0.013946533203125,
0.026336669921875,
0.0067596435546875,
0.01486968994140625,
-0.061370849609375,
-0.01122283935546875,
-0.0087890625,
0.0126190185546875,
-0.0019741058349609375,
-0.0160369873046875,
-0.031036376953125,
-0.01520538330078125,
0.05804443359375,
0.009429931640625,
0.03509521484375,
0.0010461807250976562,
0.0108489990234375,
-0.0207977294921875,
0.0025730133056640625,
0.03515625,
0.01495361328125,
-0.0112152099609375,
0.004627227783203125,
0.0216827392578125,
-0.0279998779296875,
0.0019245147705078125,
-0.007537841796875,
-0.01349639892578125,
-0.01020050048828125,
0.038116455078125,
0.07373046875,
0.0292205810546875,
-0.0413818359375,
0.01425933837890625,
0.0247955322265625,
-0.031280517578125,
-0.0235443115234375,
0.0214996337890625,
0.0014371871948242188,
0.03729248046875,
0.01091766357421875,
0.0526123046875,
0.00499725341796875,
-0.055267333984375,
-0.000028073787689208984,
0.027557373046875,
-0.0272979736328125,
-0.03399658203125,
0.045684814453125,
0.004062652587890625,
-0.0240020751953125,
0.0298614501953125,
-0.01125335693359375,
-0.03436279296875,
0.05316162109375,
0.038055419921875,
0.052398681640625,
-0.01519012451171875,
0.017303466796875,
0.024566650390625,
0.0281524658203125,
-0.007175445556640625,
0.0186767578125,
0.005157470703125,
-0.049560546875,
-0.034912109375,
-0.038330078125,
-0.0179443359375,
0.0149078369140625,
-0.042877197265625,
0.02850341796875,
-0.047119140625,
-0.02703857421875,
-0.0186767578125,
-0.01451873779296875,
-0.018951416015625,
0.005504608154296875,
0.009979248046875,
0.07568359375,
-0.08203125,
0.045013427734375,
0.03460693359375,
-0.023590087890625,
-0.052215576171875,
-0.0253753662109375,
0.0048370361328125,
-0.060546875,
0.040496826171875,
-0.00605010986328125,
0.01364898681640625,
-0.01517486572265625,
-0.056976318359375,
-0.0938720703125,
0.104736328125,
0.0014505386352539062,
-0.035430908203125,
-0.01511383056640625,
-0.004245758056640625,
0.0413818359375,
-0.034942626953125,
0.004871368408203125,
0.0491943359375,
0.040191650390625,
0.051910400390625,
-0.062408447265625,
-0.024688720703125,
-0.0183868408203125,
-0.01491546630859375,
0.0181884765625,
-0.069580078125,
0.0810546875,
-0.01029205322265625,
-0.0015478134155273438,
0.0399169921875,
0.061279296875,
0.046142578125,
0.01373291015625,
0.037689208984375,
0.0751953125,
0.060882568359375,
-0.022216796875,
0.06121826171875,
-0.0192718505859375,
0.01453399658203125,
0.071044921875,
-0.0205841064453125,
0.058380126953125,
0.05462646484375,
-0.0201873779296875,
0.034759521484375,
0.05511474609375,
-0.0033779144287109375,
0.055511474609375,
-0.0163116455078125,
-0.002552032470703125,
-0.0270843505859375,
-0.0128173828125,
-0.032684326171875,
0.035736083984375,
0.031951904296875,
-0.00905609130859375,
-0.0007791519165039062,
-0.006526947021484375,
-0.004421234130859375,
-0.01399993896484375,
-0.00960540771484375,
0.0399169921875,
0.0168914794921875,
-0.07275390625,
0.018798828125,
-0.000446319580078125,
0.0635986328125,
-0.05401611328125,
0.0039005279541015625,
-0.020660400390625,
0.0189361572265625,
-0.021087646484375,
-0.05670166015625,
0.0309295654296875,
-0.0170745849609375,
0.00876617431640625,
-0.0023651123046875,
0.05224609375,
-0.0174713134765625,
-0.038421630859375,
0.02777099609375,
0.01125335693359375,
0.038360595703125,
0.005329132080078125,
-0.038970947265625,
0.037261962890625,
0.01125335693359375,
0.01422119140625,
0.0214691162109375,
0.018341064453125,
0.005252838134765625,
0.043670654296875,
0.051910400390625,
0.006534576416015625,
-0.0103607177734375,
-0.0036563873291015625,
0.080810546875,
-0.0240478515625,
-0.0313720703125,
-0.05316162109375,
0.0531005859375,
-0.0018663406372070312,
-0.0428466796875,
0.04302978515625,
0.0537109375,
0.029754638671875,
-0.007274627685546875,
0.041229248046875,
-0.0101318359375,
0.03277587890625,
-0.043304443359375,
0.03912353515625,
-0.0543212890625,
0.033477783203125,
-0.03729248046875,
-0.09588623046875,
-0.001995086669921875,
0.044219970703125,
-0.00496673583984375,
-0.0018053054809570312,
0.04766845703125,
0.05206298828125,
-0.01045989990234375,
-0.003810882568359375,
0.026153564453125,
0.032135009765625,
0.009674072265625,
0.0509033203125,
0.07611083984375,
-0.057647705078125,
0.033966064453125,
-0.034271240234375,
-0.0301971435546875,
-0.0284271240234375,
-0.043365478515625,
-0.058807373046875,
-0.038543701171875,
-0.0206298828125,
-0.0347900390625,
-0.002559661865234375,
0.0753173828125,
0.039337158203125,
-0.034423828125,
-0.0386962890625,
-0.0018110275268554688,
-0.00708770751953125,
-0.01184844970703125,
-0.01288604736328125,
0.0051422119140625,
0.0145111083984375,
-0.05023193359375,
0.058837890625,
0.046783447265625,
0.019744873046875,
-0.0281219482421875,
-0.01364898681640625,
-0.0029773712158203125,
0.0011119842529296875,
0.038726806640625,
0.022735595703125,
-0.06903076171875,
-0.0011539459228515625,
0.0276947021484375,
0.0102386474609375,
-0.0089569091796875,
0.0361328125,
-0.0570068359375,
0.00014603137969970703,
0.0278472900390625,
0.00844573974609375,
0.055694580078125,
-0.004795074462890625,
0.02032470703125,
-0.0362548828125,
0.027923583984375,
0.013763427734375,
0.046142578125,
0.0044097900390625,
-0.0253143310546875,
0.05511474609375,
0.037994384765625,
-0.04718017578125,
-0.0667724609375,
0.0204925537109375,
-0.106201171875,
-0.020843505859375,
0.08966064453125,
0.022186279296875,
-0.02099609375,
0.0218963623046875,
-0.0386962890625,
-0.0028057098388671875,
-0.0176849365234375,
0.055633544921875,
0.042755126953125,
-0.028656005859375,
-0.0025691986083984375,
0.004344940185546875,
0.01369476318359375,
0.02301025390625,
-0.057342529296875,
-0.028045654296875,
0.02947998046875,
0.001033782958984375,
0.056671142578125,
0.0330810546875,
-0.01519012451171875,
0.026123046875,
0.00194549560546875,
0.01314544677734375,
0.00640869140625,
-0.042327880859375,
-0.009429931640625,
-0.0003864765167236328,
-0.00036787986755371094,
0.0103607177734375
]
] |
Corianas/111m | 2023-03-31T13:44:17.000Z | [
"transformers",
"pytorch",
"safetensors",
"gpt2",
"text-generation",
"dataset:tatsu-lab/alpaca",
"dataset:the_pile",
"arxiv:1910.09700",
"license:cc-by-nc-sa-4.0",
"endpoints_compatible",
"has_space",
"text-generation-inference",
"region:us"
] | text-generation | Corianas | null | null | Corianas/111m | 2 | 5,951 | transformers | 2023-03-29T14:01:30 | ---
license: cc-by-nc-sa-4.0
datasets:
- tatsu-lab/alpaca
- the_pile
---
# Model Card for Cerebras 111M Dollyfied.
This is a finetuned model of Cerebras 111M model. using DataBricksLabs Dolly Framework
## Model Details
### Model Description
This is a finetuned version of cerebras' 111million paramater model that has been trained to follow instructions.
It was accomplished using DataBricks Dolly training tools and the alpaca dataset, and was trained for 2 epochs.
- **Developed by:** Finetuned by Corianas (me) using open source tools
- **Shared by [optional]:** [More Information Needed]
- **Model type:** [More Information Needed]
- **Language(s) (NLP):** EN
- **License:** cc-by-nc-4.0
- **Finetuned from model:** https://huggingface.co/cerebras/Cerebras-GPT-111m
- **Finetuned using:** https://www.databricks.com/blog/2023/03/24/hello-dolly-democratizing-magic-chatgpt-open-models.html
## Uses
This is a simple GPT chatbot that has been finetuned to understand instructions.
Its knowledge about facts about the world is should be considered suspect at best.
### Direct Use
If you have a use you put it to, Please let me know.
[More Information Needed]
### Downstream Use [optional]
<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
[More Information Needed]
### Out-of-Scope Use
Any form of use where any form of accuracy is needed.
FOR THE LOVE OF GOD DO NOT FOLLOW MEDICAL ADVICE FROM THIS.
or financial advice.
[More Information Needed]
## Bias, Risks, and Limitations
Limitations... Yes, I am sure there are so so many.
[More Information Needed]
## How to Get Started with the Model
Use the code below to get started with the model.
[More Information Needed]
## Training Details
### Training Data
<!-- This should link to a Data Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
[More Information Needed]
### Training Procedure
<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
#### Preprocessing [optional]
[More Information Needed]
#### Training Hyperparameters
- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
#### Speeds, Sizes, Times [optional]
<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
[More Information Needed]
## Evaluation
<!-- This section describes the evaluation protocols and provides the results. -->
### Testing Data, Factors & Metrics
#### Testing Data
<!-- This should link to a Data Card if possible. -->
[More Information Needed]
#### Factors
<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
[More Information Needed]
#### Metrics
<!-- These are the evaluation metrics being used, ideally with a description of why. -->
[More Information Needed]
### Results
[More Information Needed]
#### Summary
## Model Examination [optional]
<!-- Relevant interpretability work for the model goes here -->
[More Information Needed]
## Environmental Impact
<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
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).
- **Hardware Type:** 8xA100s (accomplished while I was downloading the model I was actually training.)
- **Minutes used:** 7.5
- **Cloud Provider:** LambdaGPU
- **Compute Region:** USA
- **Carbon Emitted:** [More Information Needed]
## Technical Specifications [optional]
### Model Architecture and Objective
[More Information Needed]
### Compute Infrastructure
[More Information Needed]
#### Hardware
[More Information Needed]
#### Software
[More Information Needed]
## Citation [optional]
<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
**BibTeX:**
[More Information Needed]
**APA:**
[More Information Needed]
## Glossary [optional]
<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
[More Information Needed]
## More Information [optional]
[More Information Needed]
## Model Card Authors [optional]
[More Information Needed]
## Model Card Contact
[More Information Needed] | 4,779 | [
[
-0.0423583984375,
-0.06451416015625,
0.0218963623046875,
0.006069183349609375,
-0.0161590576171875,
-0.02734375,
0.006122589111328125,
-0.031463623046875,
0.01194000244140625,
0.05059814453125,
-0.043548583984375,
-0.039337158203125,
-0.04962158203125,
-0.0151519775390625,
-0.0255279541015625,
0.09576416015625,
0.01104736328125,
0.018310546875,
-0.0211029052734375,
0.00688934326171875,
-0.039703369140625,
-0.051513671875,
-0.0550537109375,
-0.029541015625,
0.028350830078125,
0.028076171875,
0.057342529296875,
0.05828857421875,
0.048736572265625,
0.023651123046875,
-0.031646728515625,
-0.00611114501953125,
-0.039764404296875,
-0.037811279296875,
-0.01103973388671875,
-0.0169525146484375,
-0.06640625,
0.005695343017578125,
0.0408935546875,
0.04840087890625,
-0.0129547119140625,
0.04730224609375,
0.0098724365234375,
0.034698486328125,
-0.045867919921875,
0.025115966796875,
-0.043609619140625,
0.0048675537109375,
-0.002685546875,
-0.0083160400390625,
-0.02239990234375,
-0.0216522216796875,
-0.00946807861328125,
-0.04425048828125,
0.017303466796875,
0.007198333740234375,
0.08026123046875,
0.0137176513671875,
-0.030303955078125,
-0.02398681640625,
-0.0673828125,
0.051788330078125,
-0.036834716796875,
0.017822265625,
0.035736083984375,
0.035797119140625,
-0.01284027099609375,
-0.06378173828125,
-0.033203125,
-0.023834228515625,
-0.00901031494140625,
0.017181396484375,
-0.01045989990234375,
0.0063323974609375,
0.042327880859375,
0.04083251953125,
-0.044525146484375,
-0.00916290283203125,
-0.048858642578125,
-0.0186309814453125,
0.04888916015625,
0.037567138671875,
0.0020809173583984375,
-0.022003173828125,
-0.031005859375,
-0.0270538330078125,
-0.03106689453125,
0.0122528076171875,
0.038665771484375,
0.022125244140625,
-0.037811279296875,
0.045440673828125,
-0.013092041015625,
0.055419921875,
-0.005168914794921875,
-0.0102386474609375,
0.014312744140625,
-0.029144287109375,
-0.03411865234375,
-0.008697509765625,
0.051300048828125,
0.0219573974609375,
-0.01267242431640625,
0.00817108154296875,
-0.006473541259765625,
-0.003032684326171875,
0.0243988037109375,
-0.06927490234375,
-0.0301055908203125,
0.0218048095703125,
-0.04730224609375,
-0.026885986328125,
-0.00693511962890625,
-0.07550048828125,
-0.00196075439453125,
-0.04058837890625,
0.043853759765625,
-0.0256500244140625,
-0.01031494140625,
-0.0006546974182128906,
-0.00885009765625,
0.0194244384765625,
0.02557373046875,
-0.06695556640625,
0.042083740234375,
0.036895751953125,
0.04638671875,
0.0120086669921875,
-0.014312744140625,
-0.007709503173828125,
-0.0032501220703125,
-0.007049560546875,
0.04119873046875,
-0.0256805419921875,
-0.048431396484375,
-0.0194854736328125,
0.025360107421875,
-0.00611114501953125,
-0.0235748291015625,
0.056610107421875,
-0.03533935546875,
0.0258026123046875,
-0.01329803466796875,
-0.0472412109375,
-0.031982421875,
0.022186279296875,
-0.05340576171875,
0.08270263671875,
0.004756927490234375,
-0.05914306640625,
0.0124359130859375,
-0.06488037109375,
-0.005176544189453125,
0.00412750244140625,
0.01181793212890625,
-0.045440673828125,
-0.019378662109375,
-0.0005612373352050781,
0.040374755859375,
-0.028350830078125,
0.0201416015625,
-0.024566650390625,
-0.015289306640625,
-0.011566162109375,
-0.0256500244140625,
0.08685302734375,
0.0260467529296875,
-0.0094146728515625,
0.004180908203125,
-0.07574462890625,
0.00919342041015625,
0.0211639404296875,
-0.0280914306640625,
0.01305389404296875,
-0.0289154052734375,
0.035003662109375,
0.01070404052734375,
0.0282440185546875,
-0.028900146484375,
0.0084228515625,
0.006267547607421875,
0.0194091796875,
0.033203125,
0.015716552734375,
0.01030731201171875,
-0.029632568359375,
0.046234130859375,
-0.0125274658203125,
0.0533447265625,
0.006450653076171875,
-0.047088623046875,
-0.06256103515625,
-0.001842498779296875,
0.02520751953125,
0.043853759765625,
-0.01419830322265625,
0.05316162109375,
-0.0014438629150390625,
-0.0745849609375,
-0.019744873046875,
0.00016736984252929688,
0.034149169921875,
0.05487060546875,
0.043853759765625,
-0.0107879638671875,
-0.04193115234375,
-0.065185546875,
0.0181884765625,
0.0009450912475585938,
0.007732391357421875,
0.031280517578125,
0.066162109375,
-0.0146026611328125,
0.0528564453125,
-0.048248291015625,
-0.00878143310546875,
-0.01177215576171875,
0.013092041015625,
0.0089111328125,
0.054290771484375,
0.042205810546875,
-0.0625,
-0.01544189453125,
-0.01473236083984375,
-0.04998779296875,
0.0227508544921875,
-0.0034008026123046875,
-0.01084136962890625,
-0.007694244384765625,
0.0252227783203125,
-0.04290771484375,
0.050079345703125,
0.041259765625,
-0.01873779296875,
0.043914794921875,
-0.029541015625,
0.005771636962890625,
-0.0853271484375,
0.02984619140625,
-0.004230499267578125,
-0.00858306884765625,
-0.031280517578125,
0.001155853271484375,
0.00279998779296875,
-0.03302001953125,
-0.053985595703125,
0.04913330078125,
-0.02301025390625,
-0.003887176513671875,
-0.02655029296875,
-0.010986328125,
0.007518768310546875,
0.05169677734375,
0.01473236083984375,
0.041412353515625,
0.03851318359375,
-0.064697265625,
0.02471923828125,
0.0309600830078125,
-0.0176544189453125,
0.0289459228515625,
-0.061279296875,
0.01493072509765625,
-0.00463104248046875,
0.027984619140625,
-0.0465087890625,
-0.0276641845703125,
0.02374267578125,
-0.0308685302734375,
0.0259552001953125,
-0.0147705078125,
-0.02972412109375,
-0.037811279296875,
0.00783538818359375,
0.021636962890625,
0.044921875,
-0.03033447265625,
0.0303802490234375,
0.037628173828125,
0.015594482421875,
-0.025360107421875,
-0.0372314453125,
-0.00011867284774780273,
-0.028076171875,
-0.034393310546875,
0.033599853515625,
-0.00647735595703125,
0.0023632049560546875,
-0.00856781005859375,
0.020782470703125,
-0.020477294921875,
0.006961822509765625,
0.030059814453125,
0.02734375,
0.00975799560546875,
0.00800323486328125,
-0.019775390625,
-0.0177001953125,
0.008819580078125,
0.005008697509765625,
0.03887939453125,
-0.01666259765625,
-0.0120849609375,
-0.057647705078125,
0.0190277099609375,
0.03564453125,
-0.0057525634765625,
0.062042236328125,
0.056396484375,
-0.053375244140625,
0.01270294189453125,
-0.03143310546875,
-0.017425537109375,
-0.03277587890625,
0.02728271484375,
-0.01629638671875,
-0.026031494140625,
0.043792724609375,
0.002956390380859375,
-0.0070343017578125,
0.061431884765625,
0.046417236328125,
0.006847381591796875,
0.07183837890625,
0.053558349609375,
0.01052093505859375,
0.03546142578125,
-0.050262451171875,
-0.0011720657348632812,
-0.0682373046875,
-0.037750244140625,
-0.06268310546875,
-0.006099700927734375,
-0.048736572265625,
-0.021331787109375,
0.01320648193359375,
0.0171661376953125,
-0.049957275390625,
0.050445556640625,
-0.04559326171875,
0.01629638671875,
0.038238525390625,
0.0219573974609375,
0.0008158683776855469,
-0.02349853515625,
0.0031890869140625,
0.01058197021484375,
-0.049774169921875,
-0.057037353515625,
0.08721923828125,
0.041900634765625,
0.049652099609375,
-0.01251220703125,
0.04229736328125,
0.0159759521484375,
0.018707275390625,
-0.041656494140625,
0.04461669921875,
0.00246429443359375,
-0.0673828125,
-0.009124755859375,
-0.019012451171875,
-0.07232666015625,
-0.0017461776733398438,
-0.0249786376953125,
-0.06640625,
0.0216827392578125,
0.022552490234375,
-0.031585693359375,
0.0243682861328125,
-0.048553466796875,
0.09527587890625,
-0.026611328125,
-0.0218048095703125,
-0.019134521484375,
-0.04620361328125,
0.02862548828125,
0.00016236305236816406,
0.0109710693359375,
-0.00726318359375,
0.01325225830078125,
0.06805419921875,
-0.06243896484375,
0.061553955078125,
-0.02947998046875,
0.02703857421875,
0.040191650390625,
-0.0211944580078125,
0.047454833984375,
0.0004458427429199219,
-0.00539398193359375,
0.0259246826171875,
0.015045166015625,
-0.040496826171875,
-0.0262298583984375,
0.047943115234375,
-0.06927490234375,
-0.0207061767578125,
-0.0321044921875,
-0.041595458984375,
-0.00994110107421875,
0.004253387451171875,
0.0276947021484375,
0.02496337890625,
-0.024749755859375,
0.021087646484375,
0.0465087890625,
-0.0201568603515625,
0.0221710205078125,
0.0218963623046875,
-0.0169525146484375,
-0.035888671875,
0.057952880859375,
0.004108428955078125,
0.0158843994140625,
0.0265045166015625,
0.01837158203125,
-0.035858154296875,
-0.037353515625,
-0.0288238525390625,
0.0008959770202636719,
-0.05322265625,
-0.0109100341796875,
-0.0633544921875,
-0.0273590087890625,
-0.0307769775390625,
0.0013170242309570312,
-0.03375244140625,
-0.0276947021484375,
-0.042510986328125,
-0.007038116455078125,
0.047088623046875,
0.043060302734375,
-0.00847625732421875,
0.052276611328125,
-0.0531005859375,
0.0093536376953125,
0.03179931640625,
0.00856781005859375,
0.00437164306640625,
-0.051361083984375,
-0.0287017822265625,
0.01666259765625,
-0.05462646484375,
-0.057830810546875,
0.0235595703125,
-0.003570556640625,
0.03533935546875,
0.0259246826171875,
-0.0095062255859375,
0.0621337890625,
-0.0200347900390625,
0.0731201171875,
0.0138702392578125,
-0.05743408203125,
0.049774169921875,
-0.032958984375,
0.00171661376953125,
0.0474853515625,
0.042327880859375,
-0.018829345703125,
0.01000213623046875,
-0.07476806640625,
-0.0439453125,
0.0487060546875,
0.0235443115234375,
0.0156097412109375,
0.00439453125,
0.033477783203125,
-0.00803375244140625,
0.00872039794921875,
-0.069580078125,
-0.021636962890625,
-0.0202484130859375,
0.01105499267578125,
-0.00287628173828125,
-0.007740020751953125,
-0.01422119140625,
-0.041534423828125,
0.07159423828125,
0.01187896728515625,
0.0367431640625,
0.0006122589111328125,
0.00861358642578125,
0.00995635986328125,
0.0005474090576171875,
0.03680419921875,
0.044036865234375,
-0.048797607421875,
-0.0272216796875,
0.024688720703125,
-0.053985595703125,
0.00310516357421875,
0.0169219970703125,
-0.00794219970703125,
-0.006763458251953125,
0.008148193359375,
0.07196044921875,
0.008270263671875,
-0.024871826171875,
0.025115966796875,
-0.0009527206420898438,
-0.0173187255859375,
-0.034393310546875,
0.0067901611328125,
0.0115509033203125,
-0.0010557174682617188,
-0.0060272216796875,
0.008575439453125,
0.0160369873046875,
-0.03924560546875,
0.00560760498046875,
0.028045654296875,
-0.038421630859375,
-0.006378173828125,
0.07421875,
0.01861572265625,
-0.03387451171875,
0.045806884765625,
-0.02044677734375,
-0.0142669677734375,
0.08251953125,
0.0369873046875,
0.060791015625,
-0.01136016845703125,
0.010040283203125,
0.054718017578125,
0.0130615234375,
0.0022373199462890625,
0.0226898193359375,
-0.0007696151733398438,
-0.031585693359375,
-0.007617950439453125,
-0.05535888671875,
-0.047576904296875,
0.034210205078125,
-0.069091796875,
0.0577392578125,
-0.04522705078125,
-0.0159912109375,
0.019866943359375,
0.016021728515625,
-0.08282470703125,
0.049041748046875,
0.008209228515625,
0.07818603515625,
-0.06890869140625,
0.07080078125,
0.05072021484375,
-0.04803466796875,
-0.0679931640625,
-0.0267333984375,
-0.0030956268310546875,
-0.060577392578125,
0.0286712646484375,
0.00519561767578125,
0.01495361328125,
-0.005290985107421875,
-0.03936767578125,
-0.06475830078125,
0.1033935546875,
0.0148162841796875,
-0.047576904296875,
0.01546478271484375,
0.00600433349609375,
0.041412353515625,
-0.03515625,
0.052398681640625,
0.0313720703125,
0.032135009765625,
0.02142333984375,
-0.062042236328125,
-0.000522613525390625,
-0.01163482666015625,
0.0028629302978515625,
0.0022220611572265625,
-0.06280517578125,
0.06719970703125,
-0.01090240478515625,
0.0067901611328125,
0.013092041015625,
0.034210205078125,
0.0112762451171875,
0.031158447265625,
0.02972412109375,
0.056121826171875,
0.065673828125,
0.006992340087890625,
0.08770751953125,
-0.0489501953125,
0.0504150390625,
0.0985107421875,
0.0055694580078125,
0.060028076171875,
0.0181884765625,
-0.03936767578125,
0.02142333984375,
0.08367919921875,
-0.019866943359375,
0.0266571044921875,
0.0186309814453125,
-0.006908416748046875,
-0.016021728515625,
-0.016845703125,
-0.04278564453125,
0.033935546875,
0.0250396728515625,
-0.04681396484375,
-0.019989013671875,
-0.00035834312438964844,
0.006694793701171875,
-0.0229034423828125,
-0.03369140625,
0.04547119140625,
-0.006023406982421875,
-0.03533935546875,
0.044647216796875,
0.01078033447265625,
0.0164642333984375,
-0.055572509765625,
-0.01276397705078125,
-0.00605010986328125,
0.010345458984375,
-0.0270538330078125,
-0.038055419921875,
0.0325927734375,
-0.003070831298828125,
-0.020782470703125,
-0.0061187744140625,
0.04156494140625,
-0.012359619140625,
-0.059661865234375,
0.0247955322265625,
0.01995849609375,
0.037353515625,
-0.005889892578125,
-0.07666015625,
-0.0017223358154296875,
0.0019330978393554688,
-0.0125579833984375,
0.0208587646484375,
0.0026187896728515625,
-0.0006928443908691406,
0.03955078125,
0.041107177734375,
-0.00966644287109375,
-0.00867462158203125,
-0.002285003662109375,
0.0677490234375,
-0.054901123046875,
-0.028900146484375,
-0.038482666015625,
0.044189453125,
-0.00746917724609375,
-0.0439453125,
0.047760009765625,
0.0626220703125,
0.05474853515625,
-0.00780487060546875,
0.04913330078125,
-0.01377105712890625,
0.0294036865234375,
-0.0311126708984375,
0.038421630859375,
-0.04150390625,
0.0028076171875,
-0.0214080810546875,
-0.0845947265625,
0.0024700164794921875,
0.03289794921875,
-0.0271453857421875,
0.0107269287109375,
0.0428466796875,
0.0721435546875,
-0.01232147216796875,
0.027984619140625,
0.0221710205078125,
0.008209228515625,
0.008148193359375,
0.0301513671875,
0.041534423828125,
-0.0601806640625,
0.046417236328125,
-0.0465087890625,
-0.0268096923828125,
-0.0097808837890625,
-0.0736083984375,
-0.053314208984375,
-0.03662109375,
-0.046722412109375,
-0.0285491943359375,
0.004863739013671875,
0.05877685546875,
0.0672607421875,
-0.06097412109375,
-0.023101806640625,
-0.022552490234375,
-0.0040740966796875,
-0.01812744140625,
-0.0171051025390625,
0.047943115234375,
-0.011199951171875,
-0.06396484375,
0.004131317138671875,
-0.01412200927734375,
0.027435302734375,
-0.021697998046875,
-0.010589599609375,
-0.01849365234375,
-0.0014553070068359375,
0.02685546875,
0.03021240234375,
-0.034454345703125,
-0.01739501953125,
-0.0159912109375,
-0.0046844482421875,
-0.00540924072265625,
0.0390625,
-0.043701171875,
0.0289154052734375,
0.0255584716796875,
0.02130126953125,
0.06292724609375,
0.0037097930908203125,
0.01666259765625,
-0.0218963623046875,
0.0173492431640625,
0.0196990966796875,
0.03857421875,
0.01520538330078125,
-0.045135498046875,
0.041046142578125,
0.0226593017578125,
-0.06787109375,
-0.042449951171875,
-0.0092620849609375,
-0.08612060546875,
-0.0009975433349609375,
0.08819580078125,
-0.002567291259765625,
-0.024017333984375,
-0.0065460205078125,
-0.025604248046875,
0.01331329345703125,
-0.014312744140625,
0.050811767578125,
0.05615234375,
-0.023193359375,
0.0105743408203125,
-0.059234619140625,
0.042144775390625,
0.01084136962890625,
-0.07470703125,
-0.00926971435546875,
0.036651611328125,
0.024017333984375,
0.0132904052734375,
0.032623291015625,
-0.01654052734375,
0.018829345703125,
0.0272064208984375,
0.0328369140625,
-0.00021648406982421875,
-0.02752685546875,
-0.027130126953125,
0.0022525787353515625,
-0.01499176025390625,
-0.044708251953125
]
] |
CalderaAI/13B-Ouroboros | 2023-07-29T06:38:16.000Z | [
"transformers",
"pytorch",
"llama",
"text-generation",
"alpaca",
"vicuna",
"uncensored",
"merge",
"mix",
"airoboros",
"openorca",
"orcamini",
"orca",
"instruct",
"mixtune",
"en",
"dataset:Open-Orca/OpenOrca",
"dataset:anon8231489123/ShareGPT_Vicuna_unfiltered",
"dataset:jondurbin/airoboros-uncensored",
"endpoints_compatible",
"has_space",
"text-generation-inference",
"region:us"
] | text-generation | CalderaAI | null | null | CalderaAI/13B-Ouroboros | 7 | 5,951 | transformers | 2023-07-20T21:13:34 | ---
tags:
- llama
- alpaca
- vicuna
- uncensored
- merge
- mix
- airoboros
- openorca
- orcamini
- orca
- instruct
- mixtune
datasets:
- Open-Orca/OpenOrca
- anon8231489123/ShareGPT_Vicuna_unfiltered
- jondurbin/airoboros-uncensored
language:
- en
metrics:
- accuracy
pipeline_tag: text-generation
---
## 13B-Ouroboros
Ouroboros is an experimental model based on Meta's LLaMA [v1] 13B base model using a custom merging technique, tweaking
each layer's merge % based on internal tests against the PTB dataset, scoring ~26.31 according to internal evaluation
(6 samples, sequence length 1024; this testing is not empirical, it's a quick way to find near-optimum values). Testing,
evaluating, and remixing this model is absolutely permissible and even encouraged (within the bounds of Meta's LLaMAv1
license agreement); the more feedback the better we can tune our process! 😊
## Composition:
Ouroboros is comprised of 40 layers [LLaMAv1 13B standard] mixed at optimized
ratios VS the PTB dataset for lowest perplexity score. Listed below are the
paired models and ratios merged per layer.
Tier One Merge:
13B-airoboros-gpt4-1.4 > 13B-orca_mini_v2
[0.22, 0.85, 0.89, 0.98, 0.3, 0.41, 0.71, 0.83, 0.32, 0.1, 0.44, 0.6, 0.53, 0.15, 0.86, 0.79, 0.93, 0.02, 0.19, 0.82, 0.01, 0.52, 0.07, 0.27, 0.73, 0.86, 0.08, 0.67, 0.42, 0.28, 0.37, 0.08, 0.95, 0.68, 0.45, 0.08, 0.7, 0.93, 0.96, 0.43]
13B-gpt4-x-alpaca > 13B-Vicuna-cocktail
[0.65, 0.94, 0.98, 0.87, 0.28, 0.64, 0.73, 0.7, 0.95, 0.89, 0.84, 0.9, 0.59, 0.92, 0.28, 0.61, 0.88, 0.73, 0.34, 0.85, 0.98, 0.05, 0.74, 0.92, 0.5, 0.78, 0.26, 0.4, 0.27, 0.65, 0.71, 0.7, 0.8, 0.93, 0.36, 0.03, 0.45, 0.39, 0.77, 0.06]
Tier Two Merge:
[13B-airoboros-gpt4-1.4 + 13B-orca_mini_v2] offspring > [13B-gpt4-x-alpaca + 13B-Vicuna-cocktail] offspring
[0.2, 0.83, 0.24, 0.03, 0.37, 0.62, 0.02, 0.82, 0.65, 0.63, 0.45, 0.65, 0.48, 0.45, 0.24, 0.76, 0.06, 0.31, 0.45, 0.86, 0.23, 0.99, 0.93, 0.84, 0.96, 0.53, 0.95, 0.32, 0.19, 0.06, 0.4, 0.08, 0.62, 0.4, 0.26, 0.12, 0.16, 0.91, 0.14, 0.0]
Result:
13B-Ouroboros, a model that seems uncensored and highly competent. So far only Alpaca instruction prompting has been tested and seems to work solidly well.
## Use:
Alpaca's instruct format can be used to do many things, including control of the terms of behavior
between a user and a response from an agent in chat. Below is an example of a command injected into
memory.
```
### Instruction:
Make Narrator function as a text based adventure game that responds with verbose, detailed, and creative descriptions of what happens next after Player's response.
Make Player function as the player input for Narrator's text based adventure game, controlling a character named (insert character name here, their short bio, and
whatever quest or other information to keep consistent in the interaction).
### Response:
{an empty new line here}
```
## Language Models Used Credits:
13B-airoboros-gpt4-1.4 by jondurbin
https://huggingface.co/jondurbin/airoboros-13b-gpt4-1.4
13B-orca_mini_v2 by psmathur
https://huggingface.co/psmathur/orca_mini_v2_13b
13B-gpt4-x-alpaca by chavinlo
https://huggingface.co/chavinlo/gpt4-x-alpaca
13B-Vicuna-cocktail by reeducator
https://huggingface.co/reeducator/vicuna-13b-cocktail
Also thanks to Meta for LLaMA.
Each model was hand picked and considered for what it could contribute to this ensemble.
Thanks to each and every one of you for your incredible work developing some of the best things
to come out of this community. | 3,503 | [
[
-0.041168212890625,
-0.053955078125,
0.01611328125,
0.0282745361328125,
-0.0282135009765625,
0.004329681396484375,
0.0085906982421875,
-0.046783447265625,
0.033416748046875,
0.044830322265625,
-0.0450439453125,
-0.046966552734375,
-0.045806884765625,
0.0107574462890625,
-0.0154266357421875,
0.08123779296875,
0.0207977294921875,
-0.024932861328125,
0.0152587890625,
-0.01666259765625,
-0.047393798828125,
-0.0289459228515625,
-0.056121826171875,
-0.0272674560546875,
0.042572021484375,
0.0263519287109375,
0.06463623046875,
0.0552978515625,
0.027923583984375,
0.0232391357421875,
-0.035308837890625,
0.0235137939453125,
-0.040679931640625,
-0.0125885009765625,
-0.00433349609375,
-0.0239715576171875,
-0.044769287109375,
0.00926971435546875,
0.0264739990234375,
0.039886474609375,
-0.02978515625,
0.017364501953125,
-0.0022792816162109375,
0.022186279296875,
-0.0362548828125,
0.007732391357421875,
-0.01861572265625,
0.0115814208984375,
-0.00853729248046875,
-0.018463134765625,
-0.0219268798828125,
-0.036163330078125,
0.007568359375,
-0.054779052734375,
-0.0087127685546875,
0.017120361328125,
0.08026123046875,
0.01904296875,
-0.041046142578125,
-0.038482666015625,
-0.0305023193359375,
0.0595703125,
-0.0665283203125,
-0.0025997161865234375,
0.035980224609375,
-0.0005960464477539062,
-0.0361328125,
-0.046173095703125,
-0.04974365234375,
-0.0194091796875,
-0.01499176025390625,
0.0182952880859375,
-0.038299560546875,
-0.0078582763671875,
0.01885986328125,
0.0250091552734375,
-0.03643798828125,
0.0160064697265625,
-0.04931640625,
-0.0121612548828125,
0.0440673828125,
0.0178070068359375,
0.0184326171875,
-0.01334381103515625,
-0.0491943359375,
-0.025848388671875,
-0.046966552734375,
0.0274810791015625,
0.040191650390625,
0.00347900390625,
-0.031341552734375,
0.052215576171875,
-0.005870819091796875,
0.05096435546875,
0.01373291015625,
-0.03076171875,
0.0282440185546875,
-0.021392822265625,
-0.02716064453125,
0.0008406639099121094,
0.07147216796875,
0.034027099609375,
-0.01324462890625,
0.0205535888671875,
-0.007617950439453125,
0.0022029876708984375,
0.00559234619140625,
-0.055419921875,
-0.0013399124145507812,
0.02239990234375,
-0.048431396484375,
-0.041839599609375,
-0.0013179779052734375,
-0.052398681640625,
-0.01279449462890625,
0.01068878173828125,
0.0499267578125,
-0.040435791015625,
-0.0122222900390625,
0.0128326416015625,
-0.0247039794921875,
0.04595947265625,
0.006870269775390625,
-0.06353759765625,
0.03668212890625,
0.03546142578125,
0.04693603515625,
0.001132965087890625,
-0.02508544921875,
-0.01235198974609375,
0.01611328125,
-0.039398193359375,
0.0662841796875,
-0.0343017578125,
-0.05230712890625,
-0.025909423828125,
0.004520416259765625,
0.0092315673828125,
-0.03369140625,
0.030670166015625,
-0.0278472900390625,
0.02508544921875,
-0.042510986328125,
-0.0362548828125,
-0.04156494140625,
0.0085601806640625,
-0.051177978515625,
0.074951171875,
0.0092926025390625,
-0.047088623046875,
0.0189361572265625,
-0.052825927734375,
-0.015716552734375,
-0.0158843994140625,
0.00440216064453125,
-0.0171356201171875,
-0.021270751953125,
0.020721435546875,
0.01340484619140625,
-0.035552978515625,
0.009185791015625,
-0.0110015869140625,
-0.03863525390625,
0.0265960693359375,
-0.0206756591796875,
0.09197998046875,
0.01384735107421875,
-0.0284271240234375,
-0.00125885009765625,
-0.060546875,
-0.004543304443359375,
0.0208282470703125,
-0.0228729248046875,
-0.00859832763671875,
-0.02099609375,
-0.019775390625,
-0.005115509033203125,
0.02197265625,
-0.0186309814453125,
0.036102294921875,
-0.00962066650390625,
0.040008544921875,
0.04913330078125,
0.0204925537109375,
0.032989501953125,
-0.046966552734375,
0.053009033203125,
0.0020198822021484375,
0.030120849609375,
-0.02471923828125,
-0.058074951171875,
-0.0648193359375,
-0.0282135009765625,
0.0205230712890625,
0.054718017578125,
-0.025146484375,
0.057037353515625,
0.00550079345703125,
-0.061370849609375,
-0.031280517578125,
0.0024929046630859375,
0.04266357421875,
0.052215576171875,
0.0272064208984375,
-0.0498046875,
-0.03668212890625,
-0.0687255859375,
0.041595458984375,
-0.035369873046875,
0.0037689208984375,
0.029876708984375,
0.03472900390625,
-0.0311737060546875,
0.058135986328125,
-0.04925537109375,
-0.0194549560546875,
-0.0246429443359375,
0.0123748779296875,
0.024688720703125,
0.044189453125,
0.0654296875,
-0.035400390625,
-0.0007462501525878906,
-0.007671356201171875,
-0.055419921875,
-0.0208740234375,
0.004810333251953125,
-0.0301513671875,
0.001209259033203125,
0.0109710693359375,
-0.055450439453125,
0.038421630859375,
0.04425048828125,
-0.03277587890625,
0.0258026123046875,
-0.018463134765625,
0.0153045654296875,
-0.07183837890625,
0.002178192138671875,
-0.0225677490234375,
0.00909423828125,
-0.0311431884765625,
0.01058197021484375,
-0.00905609130859375,
0.0170745849609375,
-0.037322998046875,
0.04345703125,
-0.03179931640625,
-0.013885498046875,
-0.0047760009765625,
-0.003955841064453125,
0.016265869140625,
0.0562744140625,
-0.00241851806640625,
0.039093017578125,
0.04522705078125,
-0.04364013671875,
0.037811279296875,
0.03179931640625,
-0.012603759765625,
0.0264739990234375,
-0.05487060546875,
0.0175323486328125,
-0.0009074211120605469,
0.042877197265625,
-0.07281494140625,
-0.010894775390625,
0.0504150390625,
-0.0309295654296875,
0.01678466796875,
-0.006336212158203125,
-0.01383209228515625,
-0.033447265625,
-0.03326416015625,
0.0240631103515625,
0.050140380859375,
-0.040985107421875,
0.053680419921875,
0.0211334228515625,
-0.00701141357421875,
-0.05914306640625,
-0.061309814453125,
-0.0007071495056152344,
-0.030426025390625,
-0.042755126953125,
0.0367431640625,
-0.031402587890625,
-0.005970001220703125,
0.012298583984375,
-0.007732391357421875,
-0.01690673828125,
-0.01364898681640625,
0.026092529296875,
0.038818359375,
-0.022186279296875,
-0.0143585205078125,
0.0210113525390625,
0.00695037841796875,
-0.0192718505859375,
0.00836181640625,
0.052947998046875,
-0.01357269287109375,
-0.008453369140625,
-0.03448486328125,
0.02838134765625,
0.022369384765625,
-0.018585205078125,
0.044219970703125,
0.03179931640625,
-0.01161956787109375,
0.0063018798828125,
-0.05609130859375,
0.005718231201171875,
-0.037445068359375,
-0.0018014907836914062,
-0.0203857421875,
-0.051116943359375,
0.077392578125,
0.023468017578125,
0.008453369140625,
0.04168701171875,
0.031341552734375,
-0.00485992431640625,
0.061004638671875,
0.05072021484375,
-0.01483154296875,
0.0263824462890625,
-0.055908203125,
-0.01416778564453125,
-0.06494140625,
-0.049560546875,
-0.029937744140625,
-0.036376953125,
-0.0447998046875,
-0.027069091796875,
0.01462554931640625,
0.00641632080078125,
-0.01666259765625,
0.052978515625,
-0.037841796875,
0.041168212890625,
0.045867919921875,
0.01166534423828125,
0.0198516845703125,
-0.018707275390625,
-0.007373809814453125,
0.0217437744140625,
-0.049713134765625,
-0.03326416015625,
0.08026123046875,
0.0195770263671875,
0.052459716796875,
0.0170135498046875,
0.0689697265625,
0.012115478515625,
0.023529052734375,
-0.057159423828125,
0.047454833984375,
0.005916595458984375,
-0.053802490234375,
-0.015655517578125,
-0.024658203125,
-0.0875244140625,
0.02532958984375,
-0.00975799560546875,
-0.054779052734375,
0.037750244140625,
0.00041365623474121094,
-0.05511474609375,
0.0016222000122070312,
-0.052490234375,
0.05487060546875,
-0.0164947509765625,
-0.0035495758056640625,
-0.009124755859375,
-0.055877685546875,
0.0494384765625,
0.00521087646484375,
0.0125274658203125,
-0.00638580322265625,
-0.0311431884765625,
0.067626953125,
-0.04241943359375,
0.06280517578125,
0.01322174072265625,
-0.0161895751953125,
0.059112548828125,
0.01015472412109375,
0.01210784912109375,
0.01027679443359375,
-0.0016689300537109375,
0.017242431640625,
0.009490966796875,
-0.0240936279296875,
-0.031524658203125,
0.04779052734375,
-0.0657958984375,
-0.045257568359375,
-0.04803466796875,
-0.03192138671875,
0.01163482666015625,
0.008148193359375,
0.039154052734375,
0.039398193359375,
-0.00945281982421875,
0.01134490966796875,
0.03936767578125,
-0.03155517578125,
0.038818359375,
0.05462646484375,
-0.01904296875,
-0.045257568359375,
0.046966552734375,
0.006488800048828125,
0.0124664306640625,
0.0262451171875,
0.02581787109375,
-0.01557159423828125,
-0.0204010009765625,
-0.0322265625,
0.0206756591796875,
-0.046905517578125,
-0.036834716796875,
-0.0447998046875,
-0.0230865478515625,
-0.027069091796875,
-0.01187896728515625,
-0.01422882080078125,
-0.032806396484375,
-0.0303192138671875,
-0.021820068359375,
0.032440185546875,
0.0582275390625,
-0.0150299072265625,
0.046417236328125,
-0.0638427734375,
0.0179595947265625,
0.03594970703125,
-0.005672454833984375,
-0.007236480712890625,
-0.055419921875,
0.01526641845703125,
-0.001216888427734375,
-0.043365478515625,
-0.09674072265625,
0.04388427734375,
0.0135345458984375,
0.051666259765625,
0.04302978515625,
-0.018402099609375,
0.076416015625,
-0.01904296875,
0.0762939453125,
0.01120758056640625,
-0.08233642578125,
0.052825927734375,
-0.0207061767578125,
-0.0009412765502929688,
0.021636962890625,
0.03387451171875,
-0.0238189697265625,
-0.0280914306640625,
-0.0697021484375,
-0.0599365234375,
0.0628662109375,
0.03167724609375,
0.007083892822265625,
-0.005077362060546875,
0.0226593017578125,
0.0103607177734375,
0.0177459716796875,
-0.06451416015625,
-0.035125732421875,
-0.0159454345703125,
0.006526947021484375,
-0.0055389404296875,
-0.01953125,
-0.00991058349609375,
-0.0196533203125,
0.035675048828125,
0.01385498046875,
0.007320404052734375,
0.00820159912109375,
0.014434814453125,
-0.0143890380859375,
-0.0032978057861328125,
0.051483154296875,
0.039703369140625,
-0.0291748046875,
-0.00414276123046875,
0.0235748291015625,
-0.035552978515625,
0.01546478271484375,
-0.0014019012451171875,
0.0004987716674804688,
-0.0185394287109375,
0.051422119140625,
0.04571533203125,
0.024749755859375,
-0.052825927734375,
0.0299835205078125,
-0.007171630859375,
-0.0011835098266601562,
-0.0061187744140625,
0.02435302734375,
0.005584716796875,
0.0277252197265625,
0.0237274169921875,
0.004787445068359375,
0.024505615234375,
-0.0628662109375,
-0.0089263916015625,
0.01934814453125,
0.00626373291015625,
-0.02484130859375,
0.033203125,
0.0073394775390625,
-0.01346588134765625,
0.034637451171875,
-0.03155517578125,
-0.0279388427734375,
0.0811767578125,
0.053497314453125,
0.042266845703125,
-0.0496826171875,
0.01323699951171875,
0.03228759765625,
0.032440185546875,
-0.0110321044921875,
0.031524658203125,
-0.0028591156005859375,
-0.052642822265625,
-0.00957489013671875,
-0.03790283203125,
-0.033233642578125,
0.021759033203125,
-0.05438232421875,
0.01898193359375,
-0.0400390625,
-0.0291748046875,
-0.0012712478637695312,
0.001384735107421875,
-0.041412353515625,
0.0177459716796875,
0.0099639892578125,
0.0633544921875,
-0.07122802734375,
0.06121826171875,
0.035919189453125,
-0.054534912109375,
-0.09014892578125,
-0.028717041015625,
0.00609588623046875,
-0.06689453125,
0.032012939453125,
-0.0009088516235351562,
0.003910064697265625,
-0.0161590576171875,
-0.060882568359375,
-0.08331298828125,
0.1092529296875,
0.0239410400390625,
-0.01314544677734375,
0.0044097900390625,
-0.01441192626953125,
0.0377197265625,
-0.039642333984375,
0.03912353515625,
0.05853271484375,
0.03436279296875,
0.036834716796875,
-0.069580078125,
0.010467529296875,
-0.0248260498046875,
0.0109710693359375,
-0.0003204345703125,
-0.07476806640625,
0.07867431640625,
-0.0209503173828125,
0.002574920654296875,
0.0193634033203125,
0.057769775390625,
0.041046142578125,
0.0147247314453125,
0.04351806640625,
0.049896240234375,
0.055755615234375,
0.00473785400390625,
0.0703125,
-0.001178741455078125,
0.0083160400390625,
0.062225341796875,
-0.007415771484375,
0.06158447265625,
0.019287109375,
-0.024322509765625,
0.053070068359375,
0.06787109375,
-0.00911712646484375,
0.045562744140625,
0.0034122467041015625,
-0.000469207763671875,
0.007129669189453125,
-0.0078277587890625,
-0.046844482421875,
0.0430908203125,
0.039215087890625,
-0.0229034423828125,
0.00977325439453125,
-0.018829345703125,
0.0164794921875,
-0.025238037109375,
-0.00878143310546875,
0.04681396484375,
0.004810333251953125,
-0.054931640625,
0.051177978515625,
0.005779266357421875,
0.059722900390625,
-0.064697265625,
-0.0084991455078125,
-0.041046142578125,
-0.0150299072265625,
-0.0229034423828125,
-0.0531005859375,
0.01279449462890625,
0.0023822784423828125,
0.01035308837890625,
0.0089111328125,
0.030609130859375,
-0.0200653076171875,
-0.0323486328125,
0.0290069580078125,
0.016510009765625,
0.0182037353515625,
0.033050537109375,
-0.04168701171875,
0.0280914306640625,
0.0095367431640625,
-0.03314208984375,
0.00992584228515625,
0.03546142578125,
0.0010833740234375,
0.06494140625,
0.04669189453125,
0.0015277862548828125,
0.0115814208984375,
-0.01165008544921875,
0.0751953125,
-0.05389404296875,
-0.037841796875,
-0.054290771484375,
0.029205322265625,
0.00827789306640625,
-0.038238525390625,
0.041839599609375,
0.040191650390625,
0.04364013671875,
-0.00467681884765625,
0.038116455078125,
-0.0226898193359375,
0.0081787109375,
-0.042236328125,
0.036407470703125,
-0.038787841796875,
0.0264129638671875,
-0.0211181640625,
-0.08111572265625,
0.0030460357666015625,
0.0546875,
-0.0011320114135742188,
0.01212310791015625,
0.051849365234375,
0.0687255859375,
-0.0006880760192871094,
-0.002101898193359375,
0.00504302978515625,
0.0196380615234375,
0.02288818359375,
0.06256103515625,
0.071044921875,
-0.0310211181640625,
0.058349609375,
-0.052947998046875,
-0.031890869140625,
-0.01044464111328125,
-0.07940673828125,
-0.05938720703125,
-0.0281524658203125,
-0.0297393798828125,
-0.0151214599609375,
-0.0034198760986328125,
0.077392578125,
0.046112060546875,
-0.04718017578125,
-0.0400390625,
0.0224151611328125,
0.004070281982421875,
-0.02813720703125,
-0.0142974853515625,
0.02484130859375,
0.01004791259765625,
-0.048126220703125,
0.03863525390625,
0.00518798828125,
0.0338134765625,
-0.0117645263671875,
-0.0037860870361328125,
0.0052642822265625,
0.032440185546875,
0.03143310546875,
0.033294677734375,
-0.06072998046875,
-0.0258331298828125,
-0.0058746337890625,
-0.0098876953125,
0.003627777099609375,
0.0196533203125,
-0.054779052734375,
0.0065155029296875,
0.0304412841796875,
0.002689361572265625,
0.035491943359375,
0.00939178466796875,
0.0229339599609375,
-0.03717041015625,
0.031768798828125,
0.00890350341796875,
0.0335693359375,
0.0128326416015625,
-0.030426025390625,
0.050140380859375,
-0.0013484954833984375,
-0.04217529296875,
-0.0572509765625,
0.0012350082397460938,
-0.11810302734375,
-0.0038623809814453125,
0.08038330078125,
-0.0086669921875,
-0.0280609130859375,
0.0262298583984375,
-0.04522705078125,
0.0248565673828125,
-0.04937744140625,
0.0538330078125,
0.04510498046875,
-0.027191162109375,
0.004161834716796875,
-0.024383544921875,
0.0263519287109375,
0.0111083984375,
-0.0635986328125,
-0.02947998046875,
0.03399658203125,
0.0271759033203125,
0.031890869140625,
0.05206298828125,
-0.0129547119140625,
0.0144195556640625,
0.00803375244140625,
0.0093994140625,
-0.0108795166015625,
-0.00228118896484375,
-0.0200958251953125,
-0.005962371826171875,
-0.01165008544921875,
-0.0243377685546875
]
] |
AlekseyKorshuk/chatml-pyg-v1 | 2023-06-10T16:59:31.000Z | [
"transformers",
"pytorch",
"gptj",
"text-generation",
"generated_from_trainer",
"license:creativeml-openrail-m",
"endpoints_compatible",
"has_space",
"region:us"
] | text-generation | AlekseyKorshuk | null | null | AlekseyKorshuk/chatml-pyg-v1 | 1 | 5,950 | transformers | 2023-06-06T18:44:29 | ---
license: creativeml-openrail-m
tags:
- generated_from_trainer
model-index:
- name: chatml-pyg-v1
results: []
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# chatml-pyg-v1
This model is a fine-tuned version of [PygmalionAI/pygmalion-6b](https://huggingface.co/PygmalionAI/pygmalion-6b) on the None dataset.
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 5e-06
- train_batch_size: 4
- eval_batch_size: 2
- seed: 42
- distributed_type: multi-GPU
- num_devices: 8
- total_train_batch_size: 32
- total_eval_batch_size: 16
- optimizer: Adam with betas=(0.9,0.95) and epsilon=1e-08
- lr_scheduler_type: cosine
- lr_scheduler_warmup_ratio: 0.03
- num_epochs: 1
### Training results
### Framework versions
- Transformers 4.28.1
- Pytorch 2.0.1
- Datasets 2.10.1
- Tokenizers 0.13.3
| 1,175 | [
[
-0.0290069580078125,
-0.049957275390625,
-0.004730224609375,
0.01611328125,
-0.030364990234375,
-0.04022216796875,
-0.0218963623046875,
-0.02398681640625,
0.0140380859375,
0.01384735107421875,
-0.049652099609375,
-0.03228759765625,
-0.045654296875,
-0.00004696846008300781,
-0.005886077880859375,
0.0975341796875,
0.0136871337890625,
0.021270751953125,
-0.020233154296875,
-0.007289886474609375,
-0.0263519287109375,
-0.042877197265625,
-0.06689453125,
-0.06219482421875,
0.039306640625,
0.0082550048828125,
0.0611572265625,
0.06842041015625,
0.031890869140625,
0.020172119140625,
-0.034637451171875,
-0.0064544677734375,
-0.04522705078125,
-0.02911376953125,
0.0057830810546875,
-0.0281829833984375,
-0.060882568359375,
0.0174560546875,
0.041656494140625,
0.0283966064453125,
-0.0243377685546875,
0.0465087890625,
0.01387786865234375,
0.0107269287109375,
-0.04052734375,
0.041290283203125,
-0.042816162109375,
0.018798828125,
-0.0045623779296875,
-0.0177154541015625,
-0.0225372314453125,
-0.00975799560546875,
0.0157623291015625,
-0.052276611328125,
0.0311126708984375,
-0.0115509033203125,
0.089599609375,
0.0271453857421875,
-0.014495849609375,
0.0128936767578125,
-0.0382080078125,
0.040924072265625,
-0.0626220703125,
0.00208282470703125,
0.033233642578125,
0.032928466796875,
-0.0033283233642578125,
-0.06591796875,
-0.0241851806640625,
-0.0135040283203125,
0.0010824203491210938,
0.01226043701171875,
-0.0020751953125,
0.011016845703125,
0.05303955078125,
0.0318603515625,
-0.039886474609375,
0.00899505615234375,
-0.051300048828125,
-0.019500732421875,
0.05126953125,
0.032867431640625,
0.0006647109985351562,
-0.0085601806640625,
-0.02386474609375,
-0.0027103424072265625,
-0.0260009765625,
0.004093170166015625,
0.040740966796875,
0.0135650634765625,
-0.033203125,
0.05181884765625,
-0.0241546630859375,
0.059967041015625,
0.006977081298828125,
-0.021453857421875,
0.039703369140625,
0.00045418739318847656,
-0.033447265625,
-0.0005741119384765625,
0.07135009765625,
0.04754638671875,
0.0253448486328125,
0.00522613525390625,
-0.01177215576171875,
-0.01511383056640625,
0.0198822021484375,
-0.07122802734375,
-0.04779052734375,
0.006145477294921875,
-0.053619384765625,
-0.04754638671875,
-0.014434814453125,
-0.033447265625,
-0.007724761962890625,
-0.0277252197265625,
0.044097900390625,
-0.0322265625,
-0.0279083251953125,
-0.0009279251098632812,
-0.0161895751953125,
0.0166473388671875,
0.016448974609375,
-0.06365966796875,
0.0191192626953125,
0.0238800048828125,
0.04034423828125,
0.015960693359375,
-0.04754638671875,
-0.0007243156433105469,
-0.0019359588623046875,
-0.01708984375,
0.037933349609375,
-0.01047515869140625,
-0.0293121337890625,
-0.01629638671875,
0.01102447509765625,
-0.029205322265625,
-0.039306640625,
0.05889892578125,
-0.00749969482421875,
0.0308685302734375,
0.00682830810546875,
-0.0582275390625,
-0.018280029296875,
0.031219482421875,
-0.044158935546875,
0.07647705078125,
0.0079193115234375,
-0.064208984375,
0.0389404296875,
-0.04583740234375,
-0.010955810546875,
0.0187225341796875,
-0.004241943359375,
-0.055816650390625,
-0.0028629302978515625,
0.004913330078125,
0.040924072265625,
-0.007904052734375,
0.025115966796875,
-0.045623779296875,
-0.040313720703125,
0.0118255615234375,
-0.04693603515625,
0.049713134765625,
0.0156707763671875,
-0.01543426513671875,
0.0201263427734375,
-0.0841064453125,
0.024566650390625,
0.02386474609375,
-0.046600341796875,
0.01503753662109375,
-0.0182037353515625,
0.03302001953125,
0.01849365234375,
0.041168212890625,
-0.03485107421875,
0.01837158203125,
-0.0190582275390625,
0.033599853515625,
0.03900146484375,
0.0013399124145507812,
0.005168914794921875,
-0.0364990234375,
0.0078125,
0.00417327880859375,
0.049957275390625,
0.01261138916015625,
-0.053314208984375,
-0.057220458984375,
-0.0218048095703125,
0.01479339599609375,
0.025909423828125,
-0.0277557373046875,
0.05462646484375,
-0.0052490234375,
-0.06640625,
-0.0238494873046875,
0.00344085693359375,
0.029541015625,
0.03680419921875,
0.029876708984375,
-0.001316070556640625,
-0.031494140625,
-0.07330322265625,
-0.0170440673828125,
-0.0118560791015625,
0.01543426513671875,
0.031707763671875,
0.045196533203125,
-0.0086517333984375,
0.0440673828125,
-0.03729248046875,
0.00403594970703125,
-0.0193023681640625,
-0.00027251243591308594,
0.041107177734375,
0.06658935546875,
0.046142578125,
-0.0283660888671875,
-0.016357421875,
-0.015899658203125,
-0.0645751953125,
0.016448974609375,
-0.010040283203125,
-0.0159454345703125,
-0.0030307769775390625,
0.00982666015625,
-0.059478759765625,
0.05108642578125,
0.0221405029296875,
-0.018096923828125,
0.038787841796875,
-0.038421630859375,
-0.01392364501953125,
-0.0811767578125,
0.01535797119140625,
0.0202484130859375,
0.00408935546875,
-0.0257110595703125,
0.004253387451171875,
0.00969696044921875,
-0.0178375244140625,
-0.03387451171875,
0.045166015625,
-0.01290130615234375,
0.0271759033203125,
-0.0196075439453125,
-0.024444580078125,
-0.01148223876953125,
0.051025390625,
0.015899658203125,
0.0296478271484375,
0.059906005859375,
-0.052581787109375,
0.04205322265625,
0.040496826171875,
-0.01015472412109375,
0.03253173828125,
-0.07403564453125,
0.01043701171875,
0.00916290283203125,
0.003955841064453125,
-0.053619384765625,
-0.0284576416015625,
0.05108642578125,
-0.048370361328125,
0.0297393798828125,
-0.039886474609375,
-0.0211639404296875,
-0.027374267578125,
-0.0008320808410644531,
0.0316162109375,
0.042144775390625,
-0.05645751953125,
0.028228759765625,
0.0009565353393554688,
0.030487060546875,
-0.02630615234375,
-0.041229248046875,
-0.01398468017578125,
-0.0186309814453125,
-0.03656005859375,
0.0020313262939453125,
-0.007904052734375,
0.0205230712890625,
-0.00632476806640625,
-0.01094818115234375,
-0.01922607421875,
-0.0031871795654296875,
0.0350341796875,
0.0316162109375,
-0.0157318115234375,
-0.0142669677734375,
-0.01534271240234375,
-0.01611328125,
0.02325439453125,
-0.01218414306640625,
0.03466796875,
0.0122222900390625,
-0.003932952880859375,
-0.06805419921875,
-0.015838623046875,
0.040557861328125,
-0.009063720703125,
0.064208984375,
0.06494140625,
-0.043731689453125,
0.005672454833984375,
-0.029388427734375,
-0.0016918182373046875,
-0.031341552734375,
0.034637451171875,
-0.035919189453125,
-0.01442718505859375,
0.035064697265625,
0.00609588623046875,
-0.0023593902587890625,
0.055694580078125,
0.042022705078125,
0.01544952392578125,
0.0927734375,
0.0142364501953125,
-0.0086822509765625,
0.038177490234375,
-0.045501708984375,
-0.00644683837890625,
-0.0606689453125,
-0.04718017578125,
-0.0382080078125,
-0.0151824951171875,
-0.06597900390625,
0.0024871826171875,
0.0081329345703125,
0.015655517578125,
-0.040740966796875,
0.0283660888671875,
-0.040496826171875,
0.0294647216796875,
0.055938720703125,
0.035858154296875,
-0.0193939208984375,
0.00923919677734375,
-0.0012502670288085938,
0.005718231201171875,
-0.07659912109375,
-0.0307159423828125,
0.1002197265625,
0.0302276611328125,
0.059478759765625,
-0.021453857421875,
0.04730224609375,
-0.0222015380859375,
0.00689697265625,
-0.04486083984375,
0.036041259765625,
0.0148162841796875,
-0.0611572265625,
-0.01000213623046875,
-0.037750244140625,
-0.057373046875,
0.01806640625,
-0.03802490234375,
-0.050445556640625,
-0.004451751708984375,
0.0170135498046875,
-0.021209716796875,
0.024322509765625,
-0.057037353515625,
0.099609375,
-0.01250457763671875,
-0.0292205810546875,
-0.0161895751953125,
-0.03204345703125,
0.01053619384765625,
0.0273284912109375,
-0.0285797119140625,
-0.00797271728515625,
0.01158905029296875,
0.061920166015625,
-0.042877197265625,
0.052581787109375,
-0.0304718017578125,
0.0325927734375,
0.028717041015625,
-0.0182952880859375,
0.035125732421875,
0.0226287841796875,
-0.0023555755615234375,
0.020843505859375,
0.0094451904296875,
-0.05560302734375,
-0.0234832763671875,
0.05987548828125,
-0.0955810546875,
-0.0150604248046875,
-0.03533935546875,
-0.035400390625,
-0.00711822509765625,
0.0086822509765625,
0.04608154296875,
0.047119140625,
-0.01020050048828125,
0.0161895751953125,
0.0325927734375,
-0.01006317138671875,
0.030609130859375,
0.018707275390625,
0.00554656982421875,
-0.048187255859375,
0.06744384765625,
-0.00449371337890625,
0.01043701171875,
-0.0169677734375,
0.0162811279296875,
-0.0266265869140625,
-0.041168212890625,
-0.0302734375,
0.0186767578125,
-0.0419921875,
-0.01177978515625,
-0.0292205810546875,
-0.0438232421875,
-0.0283203125,
0.023193359375,
-0.0305633544921875,
-0.0223541259765625,
-0.050567626953125,
0.0006899833679199219,
0.030914306640625,
0.041900634765625,
0.0008459091186523438,
0.044677734375,
-0.049224853515625,
0.002544403076171875,
0.0108642578125,
0.03955078125,
-0.0013837814331054688,
-0.0653076171875,
-0.0345458984375,
0.013275146484375,
-0.02899169921875,
-0.036651611328125,
0.0303192138671875,
0.01218414306640625,
0.062286376953125,
0.036834716796875,
-0.019500732421875,
0.07135009765625,
-0.018402099609375,
0.06591796875,
0.024078369140625,
-0.030914306640625,
0.034912109375,
-0.022491455078125,
0.025360107421875,
0.0233612060546875,
0.0335693359375,
-0.00933837890625,
-0.00836944580078125,
-0.093994140625,
-0.057220458984375,
0.0660400390625,
0.03387451171875,
0.022674560546875,
0.00905609130859375,
0.046417236328125,
-0.003589630126953125,
0.0143585205078125,
-0.056427001953125,
-0.03466796875,
-0.036468505859375,
-0.0103912353515625,
-0.01213836669921875,
-0.02154541015625,
-0.00977325439453125,
-0.058929443359375,
0.08197021484375,
-0.006748199462890625,
0.017242431640625,
0.004199981689453125,
0.00811004638671875,
-0.01148223876953125,
-0.01885986328125,
0.036376953125,
0.05560302734375,
-0.04339599609375,
-0.0188751220703125,
0.01421356201171875,
-0.04522705078125,
-0.0172119140625,
0.0208740234375,
-0.006256103515625,
0.010223388671875,
0.0298004150390625,
0.089599609375,
0.005008697509765625,
-0.0071868896484375,
0.0284576416015625,
-0.0238037109375,
-0.02423095703125,
-0.023162841796875,
0.030609130859375,
-0.006092071533203125,
0.0218963623046875,
0.0017518997192382812,
0.0297698974609375,
-0.0004227161407470703,
-0.004669189453125,
0.0008611679077148438,
0.0109100341796875,
-0.0251617431640625,
-0.0240020751953125,
0.06689453125,
0.00954437255859375,
-0.0234222412109375,
0.057861328125,
-0.00078582763671875,
-0.005031585693359375,
0.055389404296875,
0.046630859375,
0.058929443359375,
-0.007717132568359375,
-0.0008974075317382812,
0.06805419921875,
0.00490570068359375,
-0.0174407958984375,
0.027679443359375,
-0.0005984306335449219,
-0.041107177734375,
-0.0086822509765625,
-0.0391845703125,
-0.01255035400390625,
0.048675537109375,
-0.07720947265625,
0.027130126953125,
-0.042510986328125,
-0.0299530029296875,
0.024200439453125,
0.0150604248046875,
-0.057647705078125,
0.042877197265625,
0.017425537109375,
0.0816650390625,
-0.0579833984375,
0.06585693359375,
0.06524658203125,
-0.041168212890625,
-0.07684326171875,
-0.01419830322265625,
-0.01387786865234375,
-0.06805419921875,
0.0296478271484375,
0.005229949951171875,
0.041046142578125,
0.0086669921875,
-0.061431884765625,
-0.052642822265625,
0.0797119140625,
0.023590087890625,
-0.035125732421875,
0.00778961181640625,
0.00469970703125,
0.04779052734375,
-0.0148468017578125,
0.052093505859375,
0.0166015625,
0.007442474365234375,
0.022918701171875,
-0.08050537109375,
-0.01239013671875,
-0.038787841796875,
0.0175018310546875,
-0.002227783203125,
-0.0506591796875,
0.0811767578125,
0.0017614364624023438,
0.027130126953125,
0.032958984375,
0.0367431640625,
0.0194854736328125,
0.01422882080078125,
0.0170135498046875,
0.05413818359375,
0.040771484375,
-0.00685882568359375,
0.07135009765625,
-0.05560302734375,
0.05810546875,
0.08447265625,
0.00981903076171875,
0.03619384765625,
0.0301361083984375,
-0.005615234375,
-0.0026760101318359375,
0.07568359375,
-0.032470703125,
0.027130126953125,
0.019805908203125,
-0.006256103515625,
-0.03985595703125,
0.01345062255859375,
-0.050872802734375,
0.0263214111328125,
0.00036215782165527344,
-0.047271728515625,
-0.0220184326171875,
-0.0223388671875,
-0.0004730224609375,
-0.0311126708984375,
-0.03851318359375,
0.050872802734375,
-0.020111083984375,
-0.035552978515625,
0.0660400390625,
0.0015306472778320312,
0.035919189453125,
-0.0435791015625,
-0.006317138671875,
-0.01486968994140625,
0.03271484375,
-0.0290985107421875,
-0.042633056640625,
0.004108428955078125,
-0.0113372802734375,
-0.002651214599609375,
0.00952911376953125,
0.04791259765625,
-0.0260009765625,
-0.042510986328125,
0.00586700439453125,
0.0302734375,
0.025146484375,
-0.0110321044921875,
-0.07977294921875,
-0.01216888427734375,
-0.01425933837890625,
-0.034210205078125,
0.0263214111328125,
0.027435302734375,
0.01129150390625,
0.04827880859375,
0.04571533203125,
0.0037250518798828125,
0.0164337158203125,
0.01078033447265625,
0.0645751953125,
-0.037567138671875,
-0.032196044921875,
-0.0631103515625,
0.0288238525390625,
-0.01727294921875,
-0.07598876953125,
0.05230712890625,
0.073974609375,
0.075439453125,
-0.0211639404296875,
0.038604736328125,
0.0010585784912109375,
0.01192474365234375,
-0.0275115966796875,
0.0576171875,
-0.0307159423828125,
-0.0097198486328125,
-0.01050567626953125,
-0.05462646484375,
-0.00034236907958984375,
0.06549072265625,
-0.00827789306640625,
0.01090240478515625,
0.03338623046875,
0.06170654296875,
-0.0093231201171875,
0.01560211181640625,
0.02239990234375,
0.0082550048828125,
0.002552032470703125,
0.0316162109375,
0.058624267578125,
-0.062286376953125,
0.044219970703125,
-0.0386962890625,
-0.00628662109375,
-0.0006451606750488281,
-0.04046630859375,
-0.08917236328125,
-0.02154541015625,
-0.0526123046875,
-0.047607421875,
0.0017786026000976562,
0.07855224609375,
0.068359375,
-0.044586181640625,
-0.034759521484375,
0.0030307769775390625,
-0.0157012939453125,
-0.017974853515625,
-0.01338958740234375,
0.013824462890625,
-0.022430419921875,
-0.0504150390625,
-0.01392364501953125,
-0.0196533203125,
0.0218048095703125,
-0.001922607421875,
-0.02252197265625,
-0.0145263671875,
-0.0233306884765625,
0.0148468017578125,
0.0046539306640625,
-0.03448486328125,
-0.021484375,
-0.0226287841796875,
-0.00897979736328125,
0.01861572265625,
0.02044677734375,
-0.0374755859375,
0.0198516845703125,
0.0182037353515625,
0.01092529296875,
0.056671142578125,
-0.016876220703125,
0.0175628662109375,
-0.045928955078125,
0.0352783203125,
0.0027294158935546875,
0.0307769775390625,
0.00388336181640625,
-0.0254364013671875,
0.03656005859375,
0.041168212890625,
-0.04388427734375,
-0.050689697265625,
-0.02001953125,
-0.06964111328125,
0.0015439987182617188,
0.095458984375,
-0.00673675537109375,
-0.047760009765625,
0.029205322265625,
-0.03338623046875,
0.03961181640625,
-0.01416778564453125,
0.04669189453125,
0.04327392578125,
-0.01248931884765625,
-0.0023021697998046875,
-0.0384521484375,
0.037567138671875,
0.0036163330078125,
-0.05224609375,
-0.0189056396484375,
0.0296478271484375,
0.053558349609375,
-0.009521484375,
0.02813720703125,
-0.003589630126953125,
0.0292205810546875,
0.0175018310546875,
0.0286712646484375,
-0.029632568359375,
-0.006725311279296875,
-0.028778076171875,
-0.00571441650390625,
0.00875091552734375,
-0.0496826171875
]
] |
lmsys/vicuna-7b-delta-v1.1 | 2023-08-01T18:23:16.000Z | [
"transformers",
"pytorch",
"llama",
"text-generation",
"arxiv:2302.13971",
"arxiv:2306.05685",
"has_space",
"text-generation-inference",
"region:us"
] | text-generation | lmsys | null | null | lmsys/vicuna-7b-delta-v1.1 | 199 | 5,948 | transformers | 2023-04-12T04:15:00 | ---
inference: false
---
**NOTE: New version available**
Please check out a newer version of the weights [here](https://github.com/lm-sys/FastChat/blob/main/docs/vicuna_weights_version.md).
**NOTE: This "delta model" cannot be used directly.**
Users have to apply it on top of the original LLaMA weights to get actual Vicuna weights. See [instructions](https://github.com/lm-sys/FastChat/blob/main/docs/vicuna_weights_version.md#how-to-apply-delta-weights-for-weights-v11-and-v0).
<br>
<br>
# Vicuna Model Card
## Model Details
Vicuna is a chat assistant trained by fine-tuning LLaMA on user-shared conversations collected from ShareGPT.
- **Developed by:** [LMSYS](https://lmsys.org/)
- **Model type:** An auto-regressive language model based on the transformer architecture.
- **License:** Non-commercial license
- **Finetuned from model:** [LLaMA](https://arxiv.org/abs/2302.13971).
### Model Sources
- **Repository:** https://github.com/lm-sys/FastChat
- **Blog:** https://lmsys.org/blog/2023-03-30-vicuna/
- **Paper:** https://arxiv.org/abs/2306.05685
- **Demo:** https://chat.lmsys.org/
## Uses
The primary use of Vicuna is research on large language models and chatbots.
The primary intended users of the model are researchers and hobbyists in natural language processing, machine learning, and artificial intelligence.
## How to Get Started with the Model
Command line interface: https://github.com/lm-sys/FastChat#vicuna-weights.
APIs (OpenAI API, Huggingface API): https://github.com/lm-sys/FastChat/tree/main#api.
## Training Details
Vicuna v1.1 is fine-tuned from LLaMA with supervised instruction fine-tuning.
The training data is around 70K conversations collected from ShareGPT.com.
See more details in the "Training Details of Vicuna Models" section in the appendix of this [paper](https://arxiv.org/pdf/2306.05685.pdf).
## Evaluation
Vicuna is evaluated with standard benchmarks, human preference, and LLM-as-a-judge. See more details in this [paper](https://arxiv.org/pdf/2306.05685.pdf) and [leaderboard](https://huggingface.co/spaces/lmsys/chatbot-arena-leaderboard).
## Difference between different versions of Vicuna
See [vicuna_weights_version.md](https://github.com/lm-sys/FastChat/blob/main/docs/vicuna_weights_version.md) | 2,273 | [
[
-0.01529693603515625,
-0.06463623046875,
0.025390625,
0.03692626953125,
-0.043121337890625,
-0.016021728515625,
-0.0173797607421875,
-0.042694091796875,
0.03179931640625,
0.03094482421875,
-0.045501708984375,
-0.03997802734375,
-0.046173095703125,
-0.0007748603820800781,
-0.01038360595703125,
0.06427001953125,
0.00455474853515625,
0.015533447265625,
-0.00091552734375,
-0.023406982421875,
-0.06396484375,
-0.039886474609375,
-0.0714111328125,
-0.0228271484375,
0.04779052734375,
0.034881591796875,
0.04144287109375,
0.037353515625,
0.024810791015625,
0.0290679931640625,
-0.00942230224609375,
0.0157012939453125,
-0.040863037109375,
0.00800323486328125,
0.0260162353515625,
-0.06500244140625,
-0.05096435546875,
-0.0184783935546875,
0.045501708984375,
0.0078582763671875,
-0.0198516845703125,
0.0131378173828125,
-0.0029754638671875,
0.03582763671875,
-0.02838134765625,
0.0276641845703125,
-0.042724609375,
-0.0162506103515625,
-0.0173797607421875,
-0.03759765625,
-0.0171661376953125,
-0.029632568359375,
-0.00803375244140625,
-0.034637451171875,
-0.007678985595703125,
-0.0095062255859375,
0.08489990234375,
0.036712646484375,
-0.0343017578125,
-0.0166473388671875,
-0.043365478515625,
0.04962158203125,
-0.07122802734375,
0.033966064453125,
0.03265380859375,
0.04620361328125,
-0.0198211669921875,
-0.04736328125,
-0.041534423828125,
-0.027923583984375,
0.0108184814453125,
-0.00566864013671875,
-0.021484375,
0.0020923614501953125,
0.0030536651611328125,
0.036529541015625,
-0.023406982421875,
0.029632568359375,
-0.042022705078125,
0.0004863739013671875,
0.044647216796875,
0.0228118896484375,
0.01239776611328125,
-0.01474761962890625,
-0.0318603515625,
-0.030303955078125,
-0.0264739990234375,
0.0011720657348632812,
0.0243377685546875,
0.042694091796875,
-0.042510986328125,
0.039154052734375,
-0.0214691162109375,
0.03680419921875,
-0.0026607513427734375,
-0.0147705078125,
0.0341796875,
-0.013397216796875,
-0.03753662109375,
-0.016693115234375,
0.08294677734375,
0.035369873046875,
0.0019817352294921875,
0.0120086669921875,
0.002246856689453125,
-0.01308441162109375,
0.00830078125,
-0.0577392578125,
0.004535675048828125,
0.03900146484375,
-0.017608642578125,
-0.03424072265625,
-0.0185699462890625,
-0.026763916015625,
-0.037384033203125,
-0.020263671875,
0.029632568359375,
-0.031890869140625,
-0.030487060546875,
0.024017333984375,
0.0017223358154296875,
0.0269927978515625,
0.04058837890625,
-0.04608154296875,
0.032318115234375,
0.033447265625,
0.07452392578125,
0.0036144256591796875,
-0.0297088623046875,
-0.0129547119140625,
-0.032318115234375,
-0.014190673828125,
0.05926513671875,
0.00785064697265625,
-0.019683837890625,
-0.00514984130859375,
0.0161590576171875,
-0.0032806396484375,
-0.03912353515625,
0.05133056640625,
-0.0298309326171875,
0.0238037109375,
-0.01971435546875,
-0.038482666015625,
-0.004016876220703125,
0.0135650634765625,
-0.048187255859375,
0.0877685546875,
0.007781982421875,
-0.055267333984375,
0.006122589111328125,
-0.0364990234375,
0.000457763671875,
0.002765655517578125,
0.005619049072265625,
-0.036285400390625,
-0.00782012939453125,
0.0026378631591796875,
0.0390625,
-0.028350830078125,
0.02484130859375,
-0.017547607421875,
-0.03790283203125,
0.02032470703125,
-0.031585693359375,
0.08685302734375,
0.016754150390625,
-0.0249176025390625,
0.027191162109375,
-0.050079345703125,
-0.00928497314453125,
0.0244293212890625,
-0.028717041015625,
-0.0250396728515625,
-0.01428985595703125,
-0.0005693435668945312,
0.0005774497985839844,
0.038238525390625,
-0.021697998046875,
0.0288238525390625,
-0.0048675537109375,
0.01123046875,
0.0560302734375,
-0.0023193359375,
0.02130126953125,
-0.0322265625,
0.0294952392578125,
0.001190185546875,
0.04644775390625,
0.0166473388671875,
-0.033721923828125,
-0.0848388671875,
-0.0340576171875,
0.006015777587890625,
0.04534912109375,
-0.05670166015625,
0.0509033203125,
-0.0307159423828125,
-0.07684326171875,
-0.06158447265625,
0.019683837890625,
0.024932861328125,
0.003490447998046875,
0.024658203125,
-0.039947509765625,
-0.05926513671875,
-0.06744384765625,
-0.00885009765625,
-0.0223541259765625,
-0.00489044189453125,
0.0287017822265625,
0.0231475830078125,
-0.0396728515625,
0.06353759765625,
-0.03466796875,
-0.024688720703125,
-0.009735107421875,
-0.0013017654418945312,
0.0089569091796875,
0.02520751953125,
0.045989990234375,
-0.039947509765625,
-0.024932861328125,
-0.0118408203125,
-0.05596923828125,
-0.0029926300048828125,
-0.00600433349609375,
-0.0318603515625,
0.0104217529296875,
0.0305938720703125,
-0.052947998046875,
0.024993896484375,
0.0538330078125,
-0.0307464599609375,
0.0386962890625,
-0.01168060302734375,
-0.0029926300048828125,
-0.10003662109375,
-0.0038585662841796875,
0.01025390625,
-0.0335693359375,
-0.0450439453125,
0.003093719482421875,
-0.00130462646484375,
0.035308837890625,
-0.05474853515625,
0.0732421875,
-0.032196044921875,
0.0103302001953125,
-0.03533935546875,
-0.007137298583984375,
-0.0104522705078125,
0.0592041015625,
-0.004852294921875,
0.045166015625,
0.03173828125,
-0.06793212890625,
0.033843994140625,
0.0095062255859375,
-0.0269012451171875,
0.0139923095703125,
-0.0618896484375,
0.0211181640625,
0.0007905960083007812,
0.030120849609375,
-0.061737060546875,
-0.00022220611572265625,
0.0450439453125,
-0.037841796875,
0.015228271484375,
-0.0024566650390625,
-0.0298004150390625,
-0.01342010498046875,
-0.024871826171875,
0.0146484375,
0.0265655517578125,
-0.03350830078125,
0.0206451416015625,
0.0361328125,
0.0030117034912109375,
-0.047332763671875,
-0.04534912109375,
0.002033233642578125,
-0.0303497314453125,
-0.0081329345703125,
0.002765655517578125,
-0.019989013671875,
-0.0204010009765625,
-0.0198516845703125,
0.00554656982421875,
-0.01068115234375,
0.00804901123046875,
0.0194549560546875,
0.00397491455078125,
-0.0019588470458984375,
0.00934600830078125,
-0.00579071044921875,
0.0020580291748046875,
-0.011016845703125,
-0.00247955322265625,
0.07342529296875,
-0.037628173828125,
0.0082550048828125,
-0.059906005859375,
-0.012176513671875,
0.046630859375,
0.01107025146484375,
0.0927734375,
0.05682373046875,
-0.019866943359375,
0.0139312744140625,
-0.053070068359375,
-0.0199432373046875,
-0.0364990234375,
0.03363037109375,
-0.0131378173828125,
-0.060516357421875,
0.043365478515625,
0.0253753662109375,
0.0253753662109375,
0.03289794921875,
0.061981201171875,
0.0014934539794921875,
0.03338623046875,
0.06884765625,
-0.0131378173828125,
0.07366943359375,
-0.0195770263671875,
-0.01165008544921875,
-0.06243896484375,
-0.020477294921875,
-0.0438232421875,
-0.006988525390625,
-0.048614501953125,
-0.04791259765625,
0.0016489028930664062,
0.007633209228515625,
-0.0300750732421875,
0.05487060546875,
-0.031646728515625,
0.0015544891357421875,
0.041290283203125,
0.0255126953125,
0.0240020751953125,
-0.00004738569259643555,
0.0174407958984375,
0.007274627685546875,
-0.044189453125,
-0.044036865234375,
0.0792236328125,
0.050811767578125,
0.04931640625,
0.0140228271484375,
0.04473876953125,
0.0208282470703125,
0.040374755859375,
-0.067626953125,
0.040130615234375,
0.0221099853515625,
-0.051361083984375,
-0.032379150390625,
-0.054595947265625,
-0.08428955078125,
0.030426025390625,
-0.011932373046875,
-0.058349609375,
0.0125885009765625,
0.0029697418212890625,
-0.007457733154296875,
0.0202178955078125,
-0.05023193359375,
0.05078125,
-0.0273590087890625,
-0.0102386474609375,
-0.0018358230590820312,
-0.03668212890625,
0.04681396484375,
0.004276275634765625,
0.011932373046875,
-0.01316070556640625,
-0.0064849853515625,
0.059661865234375,
-0.050506591796875,
0.0908203125,
-0.01245880126953125,
-0.033233642578125,
0.018402099609375,
-0.007965087890625,
0.0219573974609375,
-0.0030231475830078125,
0.0052490234375,
0.03717041015625,
0.005733489990234375,
-0.035400390625,
-0.04144287109375,
0.04296875,
-0.08441162109375,
-0.03265380859375,
-0.0211029052734375,
-0.0297088623046875,
0.00981903076171875,
0.01107025146484375,
0.0245361328125,
0.0124969482421875,
-0.0098724365234375,
0.0182647705078125,
0.03582763671875,
-0.0284423828125,
0.00592041015625,
0.0394287109375,
-0.024688720703125,
-0.03424072265625,
0.044036865234375,
0.001651763916015625,
0.01055908203125,
0.0406494140625,
0.0133819580078125,
-0.01247406005859375,
-0.00917816162109375,
-0.013427734375,
0.0322265625,
-0.040802001953125,
-0.01483154296875,
-0.057373046875,
-0.017578125,
-0.03240966796875,
0.0341796875,
-0.062042236328125,
-0.03173828125,
-0.021148681640625,
-0.00213623046875,
0.055419921875,
0.0296783447265625,
0.0171356201171875,
0.0601806640625,
-0.042755126953125,
0.01416015625,
0.017486572265625,
0.0217132568359375,
0.0053253173828125,
-0.056304931640625,
-0.040191650390625,
0.0178375244140625,
-0.0243988037109375,
-0.06439208984375,
0.038604736328125,
-0.0084991455078125,
0.03424072265625,
0.0260162353515625,
0.0027637481689453125,
0.0552978515625,
-0.01404571533203125,
0.046173095703125,
0.0131378173828125,
-0.043060302734375,
0.02996826171875,
-0.019683837890625,
0.0254669189453125,
0.042236328125,
0.033843994140625,
-0.05218505859375,
-0.021026611328125,
-0.052947998046875,
-0.060455322265625,
0.034332275390625,
0.02685546875,
0.025604248046875,
0.003513336181640625,
0.035400390625,
0.004604339599609375,
0.0200653076171875,
-0.0687255859375,
-0.0400390625,
-0.0074920654296875,
-0.0164642333984375,
-0.0160980224609375,
-0.0202484130859375,
0.0003104209899902344,
-0.0300750732421875,
0.053924560546875,
-0.00685882568359375,
0.040496826171875,
0.003971099853515625,
-0.002521514892578125,
-0.005950927734375,
0.01470947265625,
0.048858642578125,
0.0226287841796875,
-0.033721923828125,
-0.0234222412109375,
0.018402099609375,
-0.039825439453125,
-0.005016326904296875,
0.0081787109375,
0.0021877288818359375,
0.007678985595703125,
0.0241241455078125,
0.10443115234375,
0.025604248046875,
-0.035247802734375,
0.0253753662109375,
-0.055419921875,
-0.01427459716796875,
-0.034912109375,
0.01024627685546875,
0.0159912109375,
0.0350341796875,
0.015533447265625,
-0.01508331298828125,
-0.01116180419921875,
-0.049346923828125,
-0.015228271484375,
0.0247802734375,
-0.027374267578125,
-0.0182647705078125,
0.03814697265625,
0.01145172119140625,
-0.03363037109375,
0.03350830078125,
0.009033203125,
-0.0242462158203125,
0.03619384765625,
0.0123138427734375,
0.06658935546875,
-0.0206756591796875,
0.006511688232421875,
0.03387451171875,
0.0214691162109375,
-0.007564544677734375,
0.0124664306640625,
-0.01470184326171875,
-0.0626220703125,
-0.0018510818481445312,
-0.0511474609375,
-0.04632568359375,
0.0233306884765625,
-0.04833984375,
0.03521728515625,
-0.0252685546875,
-0.041259765625,
-0.031585693359375,
0.038787841796875,
-0.06634521484375,
-0.0004057884216308594,
-0.0069580078125,
0.0672607421875,
-0.055511474609375,
0.07464599609375,
0.04632568359375,
-0.038970947265625,
-0.068603515625,
-0.021636962890625,
-0.0025482177734375,
-0.0616455078125,
0.01678466796875,
0.0014791488647460938,
0.00005882978439331055,
-0.00872039794921875,
-0.0504150390625,
-0.061492919921875,
0.107177734375,
0.0305938720703125,
-0.0435791015625,
-0.0135650634765625,
-0.00799560546875,
0.044708251953125,
-0.00629425048828125,
0.0443115234375,
0.0300750732421875,
0.01666259765625,
0.005252838134765625,
-0.09124755859375,
0.003093719482421875,
-0.03314208984375,
-0.0007109642028808594,
-0.0193023681640625,
-0.0826416015625,
0.072021484375,
0.002132415771484375,
-0.0070953369140625,
0.0197296142578125,
0.0634765625,
0.042755126953125,
0.004810333251953125,
0.039764404296875,
0.023223876953125,
0.07501220703125,
0.0079498291015625,
0.08929443359375,
-0.0115509033203125,
0.0205078125,
0.08404541015625,
0.0048370361328125,
0.0679931640625,
0.0308837890625,
-0.005489349365234375,
0.04791259765625,
0.06304931640625,
0.0212554931640625,
0.018341064453125,
0.00258636474609375,
-0.002246856689453125,
0.002849578857421875,
0.0036792755126953125,
-0.03558349609375,
0.036163330078125,
0.0197601318359375,
-0.0129241943359375,
0.005382537841796875,
-0.007503509521484375,
0.028045654296875,
-0.029876708984375,
-0.0032806396484375,
0.051971435546875,
0.0272369384765625,
-0.038787841796875,
0.08551025390625,
0.01366424560546875,
0.082763671875,
-0.055419921875,
0.0176239013671875,
-0.041961669921875,
0.025634765625,
-0.0007696151733398438,
-0.0147705078125,
0.0010862350463867188,
0.0083465576171875,
0.016754150390625,
0.004505157470703125,
0.03582763671875,
-0.0263214111328125,
-0.02362060546875,
0.0269012451171875,
0.041290283203125,
0.038421630859375,
-0.000762939453125,
-0.054718017578125,
0.032470703125,
-0.00627899169921875,
-0.04638671875,
0.01800537109375,
0.024017333984375,
-0.0156707763671875,
0.0810546875,
0.0390625,
0.007091522216796875,
0.0007662773132324219,
0.024810791015625,
0.0709228515625,
-0.041259765625,
-0.0298919677734375,
-0.0570068359375,
0.0284576416015625,
0.00016427040100097656,
-0.034576416015625,
0.06781005859375,
0.0296783447265625,
0.051025390625,
0.00824737548828125,
0.03363037109375,
-0.00125885009765625,
0.016387939453125,
-0.03973388671875,
0.055816650390625,
-0.060516357421875,
0.018218994140625,
-0.028045654296875,
-0.066650390625,
-0.0091705322265625,
0.04644775390625,
-0.01018524169921875,
0.01385498046875,
0.03857421875,
0.059661865234375,
0.01342010498046875,
-0.016021728515625,
0.016326904296875,
0.0286865234375,
0.0322265625,
0.03802490234375,
0.04510498046875,
-0.061431884765625,
0.03558349609375,
-0.01427459716796875,
-0.02398681640625,
-0.034576416015625,
-0.052764892578125,
-0.079345703125,
-0.048858642578125,
-0.015838623046875,
-0.024993896484375,
0.012786865234375,
0.075927734375,
0.050445556640625,
-0.0230560302734375,
-0.050537109375,
0.00567626953125,
-0.00104522705078125,
-0.0160369873046875,
-0.0158538818359375,
0.0160675048828125,
-0.0021190643310546875,
-0.0704345703125,
0.0160064697265625,
-0.0253448486328125,
0.0196075439453125,
-0.026092529296875,
-0.0292510986328125,
-0.023773193359375,
0.007801055908203125,
0.0328369140625,
0.043121337890625,
-0.04058837890625,
0.0022220611572265625,
-0.00624847412109375,
-0.03582763671875,
0.01206207275390625,
0.0216064453125,
-0.05438232421875,
0.0023250579833984375,
0.0244293212890625,
0.0125732421875,
0.051666259765625,
-0.00980377197265625,
0.036224365234375,
-0.05303955078125,
0.036590576171875,
-0.0078887939453125,
0.032257080078125,
0.036895751953125,
-0.024993896484375,
0.0310516357421875,
-0.00144195556640625,
-0.0257110595703125,
-0.06756591796875,
-0.01160430908203125,
-0.0765380859375,
-0.017364501953125,
0.094970703125,
0.0164947509765625,
-0.04193115234375,
0.0081939697265625,
-0.040802001953125,
0.052490234375,
-0.0262451171875,
0.059661865234375,
0.035888671875,
0.0189971923828125,
-0.0419921875,
-0.0518798828125,
0.036956787109375,
0.020599365234375,
-0.063720703125,
0.0032329559326171875,
0.020599365234375,
0.03533935546875,
-0.0005273818969726562,
0.0941162109375,
-0.0026798248291015625,
0.0012416839599609375,
-0.009918212890625,
0.04193115234375,
-0.017578125,
-0.0237274169921875,
-0.02044677734375,
-0.0292816162109375,
0.012054443359375,
-0.0245819091796875
]
] |
CHIH-HUNG/llama-2-13b-dolphin_5w | 2023-09-06T04:55:31.000Z | [
"transformers",
"pytorch",
"llama",
"text-generation",
"dataset:ehartford/dolphin",
"license:llama2",
"endpoints_compatible",
"has_space",
"text-generation-inference",
"region:us"
] | text-generation | CHIH-HUNG | null | null | CHIH-HUNG/llama-2-13b-dolphin_5w | 0 | 5,948 | transformers | 2023-08-25T00:46:40 | ---
license: llama2
datasets:
- ehartford/dolphin
---
# Model Card for Model ID
<!-- Provide a quick summary of what the model is/does. -->
在llama-2-13b上使用dolphin前5萬筆資料集進行訓練
# Fine-Tuning Information
- **GPU:** RTX4090 (single core / 24564MiB)
- **model:** meta-llama/Llama-2-13b-hf
- **dataset:** ehartford/dolphin (取前5w筆訓練集)
- **peft_type:** LoRA
- **lora_rank:** 8
- **lora_target:** q_proj, v_proj
- **per_device_train_batch_size:** 8
- **gradient_accumulation_steps:** 8
- **learning_rate :** 5e-5
- **epoch:** 1
- **precision:** bf16
- **quantization:** load_in_4bit
# Fine-Tuning Detail
- **train_loss:** 0.8799
- **train_runtime:** 7:11:23 (use deepspeed)
# Evaluation
- 評估結果來自**HuggingFaceH4/open_llm_leaderboard**
- 與Llama-2-13b和其他使用dolphin的模型比較4種Benchmark
- Benchmark包含**ARC**、**HellaSwag**、**MMLU**、**TruthfulQA**
- **注意**：ehartford/dolphin-llama-13b使用的是llama-1
| Model |Average| ARC |HellaSwag| MMLU | TruthfulQA |
|----------------------------------|-------|-------|---------|-------|------------|
|meta-llama/Llama-2-13b-hf | 56.9 | 58.11 | 80.97 | 54.34 | 34.17 |
|meta-llama/Llama-2-13b-chat-hf | 59.93 | 59.04 | 81.94 | 54.64 | 44.12 |
|ehartford/dolphin-llama-13b | 59.26 | 55.55 | 77.11 | 52.16 | 52.23 |
|CHIH-HUNG/llama-2-13b-dolphin_20w | 60.17 | 59.56 | 82.55 | 55.89 | 42.67 |
|CHIH-HUNG/llama-2-13b-dolphin_5w | 61 | 60.67 | 82.69 | 56.23 | 44.41 |
# How to convert dataset to json
- 在**load_dataset**中輸入資料集名稱，並且在**take**中輸入要取前幾筆資料
- 觀察該資料集的欄位名稱，填入**example**欄位中(例如instruction、input、output)
- 最後指定json檔儲存位置 (**json_filename**)
```py
import json
from datasets import load_dataset
# 讀取數據集，take可以取得該數據集前n筆資料
dataset = load_dataset("ehartford/dolphin", split="train", streaming=True).take(50000)
# 提取所需欄位並建立新的字典列表
extracted_data = []
for example in dataset:
extracted_example = {
### dolphin
"instruction": example["instruction"],
"input": example["input"],
"output": example["output"]
}
extracted_data.append(extracted_example)
# 指定 JSON 文件名稱
json_filename = "dolphin.json"
# 寫入 JSON 文件
with open(json_filename, "w") as json_file:
json.dump(extracted_data, json_file, indent=4)
print(f"數據已提取並保存為 {json_filename}")
``` | 2,289 | [
[
-0.04949951171875,
-0.04412841796875,
0.00992584228515625,
0.01491546630859375,
-0.05621337890625,
0.003231048583984375,
-0.005046844482421875,
-0.0260162353515625,
0.0176239013671875,
0.034515380859375,
-0.044677734375,
-0.038818359375,
-0.044464111328125,
0.0192413330078125,
-0.005146026611328125,
0.0750732421875,
-0.00719451904296875,
-0.01351165771484375,
0.02410888671875,
0.00004845857620239258,
-0.039154052734375,
-0.023834228515625,
-0.0572509765625,
-0.0291290283203125,
0.023162841796875,
0.008697509765625,
0.04876708984375,
0.078857421875,
0.051116943359375,
0.0226898193359375,
-0.00966644287109375,
0.024200439453125,
-0.041259765625,
-0.017059326171875,
0.0120391845703125,
-0.038055419921875,
-0.0487060546875,
-0.007381439208984375,
0.045318603515625,
0.030426025390625,
0.00911712646484375,
0.040924072265625,
0.0101318359375,
0.0518798828125,
-0.0283355712890625,
0.03143310546875,
-0.0223388671875,
0.01010894775390625,
-0.0245513916015625,
-0.02227783203125,
-0.00186920166015625,
-0.0219879150390625,
-0.00904083251953125,
-0.07281494140625,
0.00146484375,
0.0134429931640625,
0.1033935546875,
0.028594970703125,
-0.0277252197265625,
0.0024242401123046875,
-0.0306549072265625,
0.0694580078125,
-0.07568359375,
-0.003734588623046875,
0.0223388671875,
0.0250244140625,
-0.0171661376953125,
-0.044464111328125,
-0.049072265625,
0.006000518798828125,
-0.00628662109375,
0.0171966552734375,
-0.0049896240234375,
-0.0202178955078125,
0.0239410400390625,
0.031280517578125,
-0.028564453125,
0.00506591796875,
-0.032989501953125,
0.0119171142578125,
0.0660400390625,
0.0306854248046875,
0.00963592529296875,
-0.0165557861328125,
-0.0184326171875,
-0.0284881591796875,
-0.040374755859375,
0.0170745849609375,
0.034942626953125,
0.037567138671875,
-0.035247802734375,
0.041595458984375,
-0.034454345703125,
0.032135009765625,
0.006134033203125,
-0.034881591796875,
0.045623779296875,
-0.013397216796875,
-0.034210205078125,
0.005062103271484375,
0.0762939453125,
0.051605224609375,
-0.00113677978515625,
0.0212554931640625,
-0.01288604736328125,
-0.023193359375,
0.00038433074951171875,
-0.0592041015625,
-0.0222320556640625,
0.034881591796875,
-0.056854248046875,
-0.037445068359375,
0.0026645660400390625,
-0.0682373046875,
-0.00567626953125,
0.0023860931396484375,
0.0254058837890625,
-0.0236663818359375,
-0.04559326171875,
0.006378173828125,
-0.0198822021484375,
0.0205230712890625,
0.0253448486328125,
-0.0589599609375,
0.01708984375,
0.038848876953125,
0.05255126953125,
0.0091094970703125,
-0.024749755859375,
-0.0004203319549560547,
0.022979736328125,
-0.030120849609375,
0.045501708984375,
-0.0051727294921875,
-0.035858154296875,
-0.0193939208984375,
0.0192718505859375,
-0.005565643310546875,
-0.036346435546875,
0.046630859375,
-0.0290679931640625,
-0.00458526611328125,
-0.041168212890625,
-0.0227508544921875,
-0.041473388671875,
0.031982421875,
-0.057891845703125,
0.080322265625,
0.00710296630859375,
-0.06915283203125,
0.0245819091796875,
-0.05810546875,
-0.022430419921875,
0.002941131591796875,
0.00836944580078125,
-0.04144287109375,
-0.0203094482421875,
0.02850341796875,
0.04168701171875,
-0.0364990234375,
0.00962066650390625,
-0.0241546630859375,
-0.0390625,
0.0225372314453125,
-0.0184326171875,
0.068359375,
0.030517578125,
-0.0160064697265625,
-0.0030689239501953125,
-0.0684814453125,
0.00347900390625,
0.0523681640625,
-0.041961669921875,
-0.0079193115234375,
-0.002429962158203125,
-0.00251007080078125,
-0.0055999755859375,
0.033477783203125,
-0.015838623046875,
0.0312347412109375,
-0.010406494140625,
0.029937744140625,
0.062225341796875,
0.004352569580078125,
0.0064544677734375,
-0.041961669921875,
0.0197601318359375,
0.007843017578125,
0.02362060546875,
-0.005573272705078125,
-0.043487548828125,
-0.06549072265625,
-0.0247039794921875,
0.0030517578125,
0.0312347412109375,
-0.036590576171875,
0.060028076171875,
-0.0236968994140625,
-0.05224609375,
-0.047943115234375,
0.00006949901580810547,
0.019439697265625,
0.046112060546875,
0.0399169921875,
0.004833221435546875,
-0.05560302734375,
-0.066650390625,
0.008880615234375,
-0.006732940673828125,
0.010986328125,
0.034423828125,
0.053680419921875,
-0.01666259765625,
0.03082275390625,
-0.0345458984375,
-0.017059326171875,
-0.0291748046875,
0.001743316650390625,
0.0673828125,
0.0452880859375,
0.049102783203125,
-0.03411865234375,
-0.0237884521484375,
-0.00013959407806396484,
-0.0869140625,
0.01448822021484375,
-0.00296783447265625,
-0.01462554931640625,
-0.00200653076171875,
-0.0017328262329101562,
-0.04351806640625,
0.0308990478515625,
0.0270843505859375,
-0.01015472412109375,
0.041351318359375,
-0.0012674331665039062,
0.026702880859375,
-0.0762939453125,
0.015380859375,
-0.0175933837890625,
0.0170440673828125,
-0.02410888671875,
0.0138702392578125,
-0.01351165771484375,
0.0219268798828125,
-0.028900146484375,
0.0248870849609375,
-0.0264434814453125,
0.0001024007797241211,
-0.01285552978515625,
-0.0038280487060546875,
0.002857208251953125,
0.038421630859375,
-0.0016775131225585938,
0.045074462890625,
0.0401611328125,
-0.0589599609375,
0.040374755859375,
0.0310516357421875,
-0.023223876953125,
0.00331878662109375,
-0.044189453125,
0.002162933349609375,
0.007236480712890625,
0.0223541259765625,
-0.0731201171875,
-0.024444580078125,
0.041778564453125,
-0.0233001708984375,
0.021331787109375,
-0.033447265625,
-0.0265655517578125,
-0.04974365234375,
-0.04193115234375,
0.0208740234375,
0.0290069580078125,
-0.04913330078125,
0.0167083740234375,
0.0142974853515625,
0.0159912109375,
-0.059173583984375,
-0.058349609375,
-0.0049896240234375,
-0.016143798828125,
-0.03289794921875,
0.0182952880859375,
-0.008544921875,
0.00038051605224609375,
0.006336212158203125,
0.0018310546875,
0.003635406494140625,
0.00811767578125,
0.01125335693359375,
0.044769287109375,
-0.0284423828125,
-0.028106689453125,
0.007213592529296875,
-0.00537872314453125,
0.002017974853515625,
0.01203155517578125,
0.059417724609375,
-0.01666259765625,
-0.01427459716796875,
-0.050201416015625,
-0.0012826919555664062,
0.03216552734375,
0.01044464111328125,
0.042266845703125,
0.05780029296875,
-0.0120697021484375,
0.0032482147216796875,
-0.0190277099609375,
0.00339508056640625,
-0.03759765625,
0.0271453857421875,
-0.050750732421875,
-0.043243408203125,
0.05218505859375,
-0.0052337646484375,
0.0132293701171875,
0.062744140625,
0.0192413330078125,
-0.01520538330078125,
0.08404541015625,
0.0178985595703125,
-0.0214996337890625,
0.01485443115234375,
-0.0709228515625,
0.004741668701171875,
-0.081298828125,
-0.031524658203125,
-0.0322265625,
-0.03759765625,
-0.047088623046875,
-0.00876617431640625,
0.01251983642578125,
0.02423095703125,
-0.04949951171875,
0.027557373046875,
-0.0640869140625,
0.020843505859375,
0.04156494140625,
0.0164642333984375,
0.00818634033203125,
-0.004962921142578125,
0.004241943359375,
0.0015583038330078125,
-0.038665771484375,
-0.035675048828125,
0.1026611328125,
0.02362060546875,
0.0533447265625,
0.0022907257080078125,
0.05438232421875,
0.0115966796875,
0.005954742431640625,
-0.04229736328125,
0.049560546875,
-0.005542755126953125,
-0.047882080078125,
-0.0189361572265625,
-0.0186920166015625,
-0.055999755859375,
0.0300445556640625,
-0.01276397705078125,
-0.06036376953125,
0.0013332366943359375,
-0.0001493692398071289,
-0.022857666015625,
0.047210693359375,
-0.0256500244140625,
0.049560546875,
-0.031829833984375,
-0.0258941650390625,
0.000053822994232177734,
-0.037841796875,
0.056549072265625,
0.002498626708984375,
0.01050567626953125,
-0.0297088623046875,
-0.006824493408203125,
0.0806884765625,
-0.0472412109375,
0.04425048828125,
-0.0268096923828125,
0.007648468017578125,
0.0341796875,
0.00217437744140625,
0.051788330078125,
0.0200653076171875,
-0.006404876708984375,
0.04034423828125,
0.0019102096557617188,
-0.01934814453125,
-0.018768310546875,
0.05029296875,
-0.09161376953125,
-0.03985595703125,
-0.04534912109375,
-0.02520751953125,
0.009552001953125,
0.028167724609375,
0.0333251953125,
-0.0091552734375,
0.015869140625,
0.01194000244140625,
0.0306854248046875,
-0.015655517578125,
0.0440673828125,
0.0316162109375,
-0.0189056396484375,
-0.05572509765625,
0.06329345703125,
0.0018434524536132812,
-0.0088348388671875,
0.029022216796875,
0.0084686279296875,
-0.0300445556640625,
-0.0489501953125,
-0.04840087890625,
0.02642822265625,
-0.03997802734375,
-0.04022216796875,
-0.03662109375,
-0.035675048828125,
-0.03924560546875,
-0.003749847412109375,
-0.04302978515625,
-0.0188140869140625,
-0.056243896484375,
-0.00865936279296875,
0.047149658203125,
0.042633056640625,
-0.010040283203125,
0.048492431640625,
-0.057830810546875,
0.0252227783203125,
0.01335906982421875,
0.007568359375,
0.01065826416015625,
-0.055877685546875,
-0.01490020751953125,
0.00438690185546875,
-0.0247039794921875,
-0.0430908203125,
0.0531005859375,
0.0026874542236328125,
0.046112060546875,
0.053192138671875,
-0.004474639892578125,
0.08880615234375,
-0.007778167724609375,
0.0703125,
0.021514892578125,
-0.050750732421875,
0.045654296875,
-0.0287322998046875,
-0.0107879638671875,
0.0281829833984375,
0.0294036865234375,
-0.01837158203125,
0.0020809173583984375,
-0.04010009765625,
-0.063232421875,
0.079833984375,
0.0057220458984375,
-0.010711669921875,
0.0177459716796875,
0.0210418701171875,
0.0093231201171875,
0.0203857421875,
-0.05718994140625,
-0.0433349609375,
-0.03704833984375,
-0.0022182464599609375,
0.0032138824462890625,
-0.00775909423828125,
-0.016204833984375,
-0.04095458984375,
0.0621337890625,
-0.005191802978515625,
0.0367431640625,
0.006137847900390625,
0.01336669921875,
-0.0216064453125,
-0.00885009765625,
0.0300140380859375,
0.0265655517578125,
-0.046630859375,
-0.007305145263671875,
0.01299285888671875,
-0.042938232421875,
0.004001617431640625,
0.0023326873779296875,
-0.011810302734375,
-0.0105743408203125,
0.0416259765625,
0.0557861328125,
0.004077911376953125,
-0.02947998046875,
0.019134521484375,
-0.0014286041259765625,
-0.0156707763671875,
-0.02117919921875,
0.0240631103515625,
-0.0017604827880859375,
0.0305633544921875,
0.038299560546875,
-0.00011169910430908203,
0.003978729248046875,
-0.0225830078125,
-0.0125885009765625,
0.0175933837890625,
0.0193023681640625,
-0.0175323486328125,
0.06292724609375,
0.0016775131225585938,
-0.0035457611083984375,
0.05279541015625,
-0.01517486572265625,
-0.032135009765625,
0.06634521484375,
0.043243408203125,
0.050811767578125,
-0.01206207275390625,
-0.0030517578125,
0.06396484375,
0.0272979736328125,
-0.017303466796875,
0.046234130859375,
-0.004138946533203125,
-0.05743408203125,
-0.01183319091796875,
-0.0556640625,
-0.0157470703125,
0.050689697265625,
-0.054931640625,
0.01537322998046875,
-0.056671142578125,
-0.025054931640625,
0.0047454833984375,
0.0312347412109375,
-0.048492431640625,
0.017303466796875,
0.009124755859375,
0.06707763671875,
-0.058319091796875,
0.07037353515625,
0.0357666015625,
-0.04766845703125,
-0.06732177734375,
-0.028594970703125,
-0.0115814208984375,
-0.08294677734375,
0.0509033203125,
0.007518768310546875,
0.0193328857421875,
-0.00618743896484375,
-0.06365966796875,
-0.08782958984375,
0.10528564453125,
0.0115203857421875,
-0.04339599609375,
0.015380859375,
0.01548004150390625,
0.023193359375,
-0.025390625,
0.03082275390625,
0.04888916015625,
0.05181884765625,
0.002460479736328125,
-0.05621337890625,
0.0230560302734375,
-0.0367431640625,
-0.0004057884216308594,
-0.0044403076171875,
-0.083740234375,
0.09991455078125,
-0.021514892578125,
0.0095062255859375,
0.004222869873046875,
0.042022705078125,
0.047332763671875,
0.018768310546875,
0.028717041015625,
0.051544189453125,
0.0516357421875,
-0.019378662109375,
0.056488037109375,
-0.0096435546875,
0.035675048828125,
0.05572509765625,
-0.0160675048828125,
0.054931640625,
0.0281982421875,
-0.04022216796875,
0.031341552734375,
0.0732421875,
-0.040435791015625,
0.058746337890625,
-0.009002685546875,
-0.00682830810546875,
-0.007564544677734375,
0.003997802734375,
-0.050567626953125,
0.020416259765625,
0.033294677734375,
-0.0285797119140625,
0.0036983489990234375,
-0.016326904296875,
0.016632080078125,
-0.037139892578125,
-0.0246734619140625,
0.044219970703125,
-0.0161285400390625,
-0.0274505615234375,
0.07421875,
-0.0061798095703125,
0.05755615234375,
-0.044677734375,
-0.011016845703125,
-0.0167236328125,
0.01094818115234375,
-0.02789306640625,
-0.06805419921875,
-0.004749298095703125,
0.0005402565002441406,
-0.012908935546875,
0.019744873046875,
0.041473388671875,
-0.0008420944213867188,
-0.03387451171875,
0.0304412841796875,
0.009552001953125,
0.035491943359375,
0.01226806640625,
-0.080322265625,
0.036102294921875,
0.0198516845703125,
-0.04522705078125,
0.0216522216796875,
0.0182952880859375,
0.0221710205078125,
0.06573486328125,
0.071533203125,
0.01275634765625,
0.01800537109375,
-0.010589599609375,
0.07330322265625,
-0.058868408203125,
-0.02349853515625,
-0.0509033203125,
0.0309906005859375,
-0.0220947265625,
-0.042083740234375,
0.049102783203125,
0.059722900390625,
0.06304931640625,
-0.01031494140625,
0.0738525390625,
-0.019561767578125,
0.03240966796875,
-0.0288543701171875,
0.05267333984375,
-0.05255126953125,
0.012725830078125,
-0.0137939453125,
-0.039215087890625,
-0.00628662109375,
0.06231689453125,
-0.006000518798828125,
0.0014734268188476562,
0.04156494140625,
0.043731689453125,
-0.0015974044799804688,
0.0119476318359375,
0.006771087646484375,
0.0225067138671875,
0.0286102294921875,
0.0582275390625,
0.0487060546875,
-0.07281494140625,
0.046875,
-0.06097412109375,
-0.0025787353515625,
-0.0341796875,
-0.0487060546875,
-0.06536865234375,
-0.01947021484375,
-0.0251617431640625,
-0.023040771484375,
-0.005619049072265625,
0.059356689453125,
0.047882080078125,
-0.058380126953125,
-0.02850341796875,
0.003261566162109375,
0.00937652587890625,
-0.033111572265625,
-0.0197601318359375,
0.0535888671875,
-0.0025653839111328125,
-0.05682373046875,
0.0311279296875,
-0.001415252685546875,
0.016082763671875,
-0.00008875131607055664,
-0.0239715576171875,
-0.01641845703125,
-0.019439697265625,
0.0268402099609375,
0.0269317626953125,
-0.049652099609375,
-0.01093292236328125,
-0.0257110595703125,
-0.001018524169921875,
0.0142364501953125,
0.0084991455078125,
-0.04034423828125,
0.01497650146484375,
0.0308990478515625,
0.0161285400390625,
0.05712890625,
-0.0117340087890625,
-0.0013837814331054688,
-0.0268096923828125,
0.0203094482421875,
-0.0026988983154296875,
0.035675048828125,
0.0070343017578125,
-0.033447265625,
0.06005859375,
0.03472900390625,
-0.03875732421875,
-0.0770263671875,
-0.03521728515625,
-0.0958251953125,
-0.021575927734375,
0.069580078125,
-0.01114654541015625,
-0.054168701171875,
0.01512908935546875,
-0.0159149169921875,
0.04498291015625,
-0.04339599609375,
0.04974365234375,
0.02325439453125,
-0.02008056640625,
-0.0008878707885742188,
-0.046112060546875,
0.0263214111328125,
-0.01226043701171875,
-0.0523681640625,
-0.0091094970703125,
-0.001110076904296875,
0.0299835205078125,
0.01751708984375,
0.0270843505859375,
-0.0011625289916992188,
0.004909515380859375,
0.024078369140625,
0.014984130859375,
-0.0279541015625,
-0.0015363693237304688,
-0.0017137527465820312,
-0.00997161865234375,
-0.020416259765625,
-0.045013427734375
]
] |
CHIH-HUNG/llama-2-13b-Open_Platypus_and_ccp_2.6w | 2023-09-06T04:57:17.000Z | [
"transformers",
"pytorch",
"llama",
"text-generation",
"dataset:garage-bAInd/Open-Platypus",
"license:llama2",
"endpoints_compatible",
"has_space",
"text-generation-inference",
"region:us"
] | text-generation | CHIH-HUNG | null | null | CHIH-HUNG/llama-2-13b-Open_Platypus_and_ccp_2.6w | 0 | 5,947 | transformers | 2023-09-04T12:05:38 | ---
license: llama2
datasets:
- garage-bAInd/Open-Platypus
---
# Model Card for Model ID
<!-- Provide a quick summary of what the model is/does. -->
在llama-2-13b上使用garage-bAInd/Open-Platypus資料集進行訓練，總資料筆數約2.5w + ccp
# Fine-Tuning Information
- **GPU:** RTX4090 (single core / 24564MiB)
- **model:** meta-llama/Llama-2-13b-hf
- **dataset:** garage-bAInd/Open-Platypus (共約2.5w筆訓練集) + ccp (約1200筆)
- **peft_type:** LoRA
- **lora_rank:** 8
- **lora_target:** gate_proj, up_proj, down_proj
- **per_device_train_batch_size:** 8
- **gradient_accumulation_steps:** 8
- **learning_rate :** 5e-5
- **epoch:** 1
- **precision:** bf16
- **quantization:** load_in_4bit
# Fine-Tuning Detail
- **train_loss:** 0.67
- **train_runtime:** 4:07:24 (use deepspeed)
# Evaluation
- 評估結果來自**HuggingFaceH4/open_llm_leaderboard**
- 與Llama-2-13b比較4種Benchmark，包含**ARC**、**HellaSwag**、**MMLU**、**TruthfulQA**
| Model |Average| ARC |HellaSwag| MMLU |TruthfulQA|
|-------------------------------------------------|-------|-------|---------|-------|----------|
|meta-llama/Llama-2-13b-hf | 56.9 | 58.11 | 80.97 | 54.34 | 34.17 |
|meta-llama/Llama-2-13b-chat-hf | 59.93 | 59.04 | 81.94 | 54.64 | 44.12 |
|Open-Orca/OpenOrca-Platypus2-13B | 63.19 | 61.52 | 82.27 | 58.85 | 50.11 |
|CHIH-HUNG/llama-2-13b-Open_Platypus_and_ccp_2.6w | 59.41 | 58.96 | 82.51 | 56.12 | 40.07 |
# How to convert dataset to json
- 在**load_dataset**中輸入資料集名稱，並且在**take**中輸入要取前幾筆資料
- 觀察該資料集的欄位名稱，填入**example**欄位中(例如instruction、input、output)
- 最後指定json檔儲存位置 (**json_filename**)
```py
import json
from datasets import load_dataset
# 讀取數據集，take可以取得該數據集前n筆資料
dataset = load_dataset("garage-bAInd/Open-Platypus", split="train", streaming=True)
# 提取所需欄位並建立新的字典列表
extracted_data = []
for example in dataset:
extracted_example = {
"instruction": example["instruction"],
"input": example["input"],
"output": example["output"]
}
extracted_data.append(extracted_example)
# 指定 JSON 文件名稱
json_filename = "Open-Platypus.json"
# 寫入 JSON 文件
with open(json_filename, "w") as json_file:
json.dump(extracted_data, json_file, indent=4)
print(f"數據已提取並保存為 {json_filename}")
``` | 2,272 | [
[
-0.038330078125,
-0.041473388671875,
0.01374053955078125,
0.0125274658203125,
-0.04998779296875,
0.0086669921875,
-0.01377105712890625,
-0.016845703125,
0.01279449462890625,
0.035858154296875,
-0.04266357421875,
-0.04669189453125,
-0.044403076171875,
0.007511138916015625,
-0.0166778564453125,
0.0799560546875,
-0.0159759521484375,
-0.01451873779296875,
0.0141143798828125,
0.004032135009765625,
-0.040679931640625,
-0.034088134765625,
-0.04669189453125,
-0.023284912109375,
0.014495849609375,
0.0103759765625,
0.05224609375,
0.0772705078125,
0.051605224609375,
0.022064208984375,
-0.005523681640625,
0.02178955078125,
-0.037841796875,
-0.018951416015625,
0.0175323486328125,
-0.04473876953125,
-0.045135498046875,
-0.003765106201171875,
0.049102783203125,
0.0255889892578125,
0.0035648345947265625,
0.0428466796875,
0.00875091552734375,
0.044158935546875,
-0.030059814453125,
0.02734375,
-0.02734375,
0.00450897216796875,
-0.0275115966796875,
-0.02496337890625,
-0.004467010498046875,
-0.0203857421875,
-0.01393890380859375,
-0.0677490234375,
0.01027679443359375,
0.01029205322265625,
0.10345458984375,
0.029205322265625,
-0.0226898193359375,
0.00043082237243652344,
-0.03680419921875,
0.0650634765625,
-0.0753173828125,
-0.0027923583984375,
0.0217437744140625,
0.03009033203125,
-0.0139923095703125,
-0.05072021484375,
-0.05328369140625,
0.007167816162109375,
-0.01483154296875,
0.01561737060546875,
-0.001995086669921875,
-0.02044677734375,
0.0245513916015625,
0.0369873046875,
-0.0328369140625,
0.0007915496826171875,
-0.036376953125,
0.0082550048828125,
0.06573486328125,
0.034881591796875,
0.0003795623779296875,
-0.0198211669921875,
-0.020599365234375,
-0.031097412109375,
-0.034637451171875,
0.0238037109375,
0.03643798828125,
0.032562255859375,
-0.034942626953125,
0.03546142578125,
-0.038299560546875,
0.0303955078125,
-0.003772735595703125,
-0.038482666015625,
0.05224609375,
-0.015960693359375,
-0.0401611328125,
-0.0021190643310546875,
0.07977294921875,
0.046417236328125,
-0.00576019287109375,
0.01447296142578125,
-0.0099945068359375,
-0.0118408203125,
-0.0025653839111328125,
-0.058868408203125,
-0.02630615234375,
0.042694091796875,
-0.0506591796875,
-0.031585693359375,
0.00823974609375,
-0.066162109375,
-0.00804901123046875,
0.0000034570693969726562,
0.019805908203125,
-0.0276031494140625,
-0.0419921875,
0.00887298583984375,
-0.015960693359375,
0.0249786376953125,
0.0252838134765625,
-0.053863525390625,
0.0164031982421875,
0.04425048828125,
0.058013916015625,
0.0097198486328125,
-0.0272674560546875,
-0.0054931640625,
0.01776123046875,
-0.02410888671875,
0.049957275390625,
-0.00749969482421875,
-0.029815673828125,
-0.01904296875,
0.0205230712890625,
-0.009033203125,
-0.03863525390625,
0.05364990234375,
-0.0279693603515625,
-0.007770538330078125,
-0.03460693359375,
-0.0160980224609375,
-0.04315185546875,
0.0288238525390625,
-0.05157470703125,
0.08099365234375,
0.00604248046875,
-0.06719970703125,
0.0249786376953125,
-0.0567626953125,
-0.0229644775390625,
0.0073394775390625,
0.0013914108276367188,
-0.038818359375,
-0.0196990966796875,
0.025421142578125,
0.042022705078125,
-0.033355712890625,
0.01444244384765625,
-0.016693115234375,
-0.032073974609375,
0.02728271484375,
-0.02056884765625,
0.06976318359375,
0.0288238525390625,
-0.0138397216796875,
0.0020198822021484375,
-0.06793212890625,
-0.0002689361572265625,
0.052215576171875,
-0.035186767578125,
-0.005268096923828125,
-0.001926422119140625,
-0.00511932373046875,
-0.0030651092529296875,
0.034423828125,
-0.01934814453125,
0.0252838134765625,
-0.01448822021484375,
0.0345458984375,
0.06317138671875,
-0.0033435821533203125,
0.008087158203125,
-0.040496826171875,
0.0226287841796875,
0.007366180419921875,
0.0232696533203125,
-0.0078277587890625,
-0.040069580078125,
-0.0716552734375,
-0.0265350341796875,
0.00934600830078125,
0.041046142578125,
-0.031036376953125,
0.057281494140625,
-0.024139404296875,
-0.052001953125,
-0.055084228515625,
0.00750732421875,
0.01438140869140625,
0.043609619140625,
0.038543701171875,
0.0023288726806640625,
-0.05401611328125,
-0.067626953125,
0.00725555419921875,
-0.004058837890625,
0.00732421875,
0.0264892578125,
0.056182861328125,
-0.01404571533203125,
0.041046142578125,
-0.0382080078125,
-0.0192718505859375,
-0.02398681640625,
0.0037822723388671875,
0.06622314453125,
0.043731689453125,
0.048492431640625,
-0.031890869140625,
-0.03125,
0.0008034706115722656,
-0.083740234375,
0.01165008544921875,
-0.00841522216796875,
-0.0185546875,
-0.00734710693359375,
-0.0032787322998046875,
-0.049835205078125,
0.03582763671875,
0.0305633544921875,
-0.00962066650390625,
0.039337158203125,
0.003749847412109375,
0.021759033203125,
-0.07763671875,
0.01535797119140625,
-0.020294189453125,
0.006488800048828125,
-0.025421142578125,
0.0179595947265625,
-0.0100860595703125,
0.019378662109375,
-0.029876708984375,
0.0191650390625,
-0.0237884521484375,
0.004802703857421875,
-0.01239013671875,
-0.004047393798828125,
0.003910064697265625,
0.04144287109375,
-0.003993988037109375,
0.048675537109375,
0.04364013671875,
-0.050384521484375,
0.0419921875,
0.03582763671875,
-0.0246124267578125,
0.004856109619140625,
-0.036163330078125,
0.0028057098388671875,
0.00528717041015625,
0.0242767333984375,
-0.073974609375,
-0.023834228515625,
0.04327392578125,
-0.029815673828125,
0.0156707763671875,
-0.030914306640625,
-0.037109375,
-0.047332763671875,
-0.039093017578125,
0.01511383056640625,
0.033416748046875,
-0.048187255859375,
0.022308349609375,
0.00876617431640625,
0.01494598388671875,
-0.0509033203125,
-0.057098388671875,
-0.006473541259765625,
-0.0162353515625,
-0.04083251953125,
0.01392364501953125,
-0.012664794921875,
-0.0023479461669921875,
0.0032825469970703125,
0.003253936767578125,
0.0040435791015625,
0.00543212890625,
0.0205535888671875,
0.0401611328125,
-0.0209808349609375,
-0.0294342041015625,
0.00292205810546875,
-0.006862640380859375,
0.006565093994140625,
0.01297760009765625,
0.06390380859375,
-0.019439697265625,
-0.0090484619140625,
-0.053497314453125,
0.004863739013671875,
0.029998779296875,
0.0009889602661132812,
0.0526123046875,
0.053985595703125,
-0.01456451416015625,
0.00010704994201660156,
-0.0178070068359375,
0.0040740966796875,
-0.037567138671875,
0.0213775634765625,
-0.046173095703125,
-0.045257568359375,
0.0526123046875,
-0.007587432861328125,
0.0174713134765625,
0.0654296875,
0.0272674560546875,
-0.0163421630859375,
0.07879638671875,
0.01334381103515625,
-0.011871337890625,
0.0130462646484375,
-0.0760498046875,
0.006500244140625,
-0.07818603515625,
-0.033233642578125,
-0.03436279296875,
-0.04052734375,
-0.043914794921875,
-0.0167236328125,
0.01227569580078125,
0.0218963623046875,
-0.0438232421875,
0.0288238525390625,
-0.0609130859375,
0.0195770263671875,
0.043212890625,
0.01302337646484375,
0.01155853271484375,
-0.005130767822265625,
0.006961822509765625,
0.0032825469970703125,
-0.037994384765625,
-0.03875732421875,
0.1031494140625,
0.028045654296875,
0.04913330078125,
-0.0021533966064453125,
0.051849365234375,
0.0062103271484375,
0.01445770263671875,
-0.041107177734375,
0.04852294921875,
-0.0060272216796875,
-0.047943115234375,
-0.007099151611328125,
-0.0205230712890625,
-0.050140380859375,
0.0248870849609375,
-0.01424407958984375,
-0.06011962890625,
0.0025043487548828125,
0.0023326873779296875,
-0.031463623046875,
0.0499267578125,
-0.036773681640625,
0.05218505859375,
-0.03497314453125,
-0.023681640625,
0.0020961761474609375,
-0.0435791015625,
0.055419921875,
0.0026531219482421875,
0.0146331787109375,
-0.0258941650390625,
-0.003925323486328125,
0.082763671875,
-0.050384521484375,
0.045989990234375,
-0.0258331298828125,
0.00787353515625,
0.037353515625,
0.002346038818359375,
0.0552978515625,
0.0206146240234375,
-0.0004754066467285156,
0.047119140625,
0.0015239715576171875,
-0.0168304443359375,
-0.0224761962890625,
0.052764892578125,
-0.0919189453125,
-0.04248046875,
-0.04632568359375,
-0.031494140625,
0.01395416259765625,
0.0229339599609375,
0.030517578125,
-0.01190185546875,
0.0234527587890625,
0.017486572265625,
0.0285186767578125,
-0.01325225830078125,
0.0450439453125,
0.031280517578125,
-0.012939453125,
-0.057098388671875,
0.0640869140625,
0.0022907257080078125,
-0.006389617919921875,
0.0311126708984375,
0.00789642333984375,
-0.02471923828125,
-0.052459716796875,
-0.0419921875,
0.0265655517578125,
-0.04290771484375,
-0.049102783203125,
-0.041961669921875,
-0.04010009765625,
-0.038238525390625,
-0.005893707275390625,
-0.0413818359375,
-0.0165252685546875,
-0.05810546875,
-0.009002685546875,
0.05084228515625,
0.032928466796875,
-0.004467010498046875,
0.059051513671875,
-0.055084228515625,
0.0241851806640625,
0.01177215576171875,
0.01360321044921875,
0.0063018798828125,
-0.05670166015625,
-0.0164794921875,
0.00457763671875,
-0.0313720703125,
-0.04425048828125,
0.056915283203125,
-0.0023956298828125,
0.040924072265625,
0.04876708984375,
0.002056121826171875,
0.08282470703125,
-0.010284423828125,
0.0693359375,
0.016448974609375,
-0.050994873046875,
0.040374755859375,
-0.0242767333984375,
-0.0140533447265625,
0.033721923828125,
0.0281219482421875,
-0.01611328125,
-0.0001081228256225586,
-0.042236328125,
-0.06512451171875,
0.07794189453125,
0.013031005859375,
-0.01346588134765625,
0.020721435546875,
0.018096923828125,
0.0160064697265625,
0.0267791748046875,
-0.0614013671875,
-0.0380859375,
-0.042236328125,
-0.00328826904296875,
0.003856658935546875,
-0.00324249267578125,
-0.0245513916015625,
-0.034637451171875,
0.057861328125,
-0.001506805419921875,
0.035797119140625,
0.009521484375,
0.0105743408203125,
-0.0167999267578125,
0.00045180320739746094,
0.0278472900390625,
0.0301055908203125,
-0.0528564453125,
-0.005229949951171875,
0.0105133056640625,
-0.04290771484375,
-0.0019359588623046875,
0.008636474609375,
-0.01081085205078125,
-0.01409912109375,
0.033203125,
0.06256103515625,
-0.01146697998046875,
-0.034942626953125,
0.0248260498046875,
0.0004487037658691406,
-0.02313232421875,
-0.0301055908203125,
0.022186279296875,
-0.00537109375,
0.03472900390625,
0.041107177734375,
0.0019378662109375,
0.002429962158203125,
-0.0201873779296875,
-0.01444244384765625,
0.0234375,
0.02838134765625,
-0.0198974609375,
0.06707763671875,
0.00281524658203125,
-0.01229095458984375,
0.047821044921875,
-0.015960693359375,
-0.03485107421875,
0.0579833984375,
0.039947509765625,
0.05413818359375,
-0.0041046142578125,
-0.002925872802734375,
0.057708740234375,
0.036773681640625,
-0.01154327392578125,
0.04144287109375,
-0.0035552978515625,
-0.051513671875,
-0.01971435546875,
-0.05877685546875,
-0.010711669921875,
0.052825927734375,
-0.0474853515625,
0.0169830322265625,
-0.060516357421875,
-0.018798828125,
-0.0021514892578125,
0.02838134765625,
-0.054473876953125,
0.0207061767578125,
0.0100860595703125,
0.06854248046875,
-0.0631103515625,
0.06597900390625,
0.02935791015625,
-0.038818359375,
-0.0660400390625,
-0.0270843505859375,
-0.015869140625,
-0.07879638671875,
0.0419921875,
0.01015472412109375,
0.0222320556640625,
-0.00849151611328125,
-0.06939697265625,
-0.0804443359375,
0.1112060546875,
0.018463134765625,
-0.046173095703125,
0.01486968994140625,
0.019134521484375,
0.022491455078125,
-0.0238494873046875,
0.029327392578125,
0.051849365234375,
0.047637939453125,
0.007129669189453125,
-0.055908203125,
0.02166748046875,
-0.03436279296875,
-0.00731658935546875,
0.0016031265258789062,
-0.08685302734375,
0.10089111328125,
-0.01343536376953125,
0.0006971359252929688,
0.00977325439453125,
0.04351806640625,
0.039947509765625,
0.02239990234375,
0.02740478515625,
0.05810546875,
0.049774169921875,
-0.02056884765625,
0.062225341796875,
-0.00885772705078125,
0.036102294921875,
0.06268310546875,
-0.013946533203125,
0.059906005859375,
0.026031494140625,
-0.035797119140625,
0.041107177734375,
0.0679931640625,
-0.0364990234375,
0.055816650390625,
-0.01409149169921875,
-0.006076812744140625,
-0.0107574462890625,
0.003337860107421875,
-0.048797607421875,
0.0236053466796875,
0.031402587890625,
-0.0296630859375,
0.0011835098266601562,
-0.01995849609375,
0.013916015625,
-0.031280517578125,
-0.0223236083984375,
0.038787841796875,
-0.020721435546875,
-0.0237884521484375,
0.07171630859375,
0.0010385513305664062,
0.0543212890625,
-0.047515869140625,
-0.00970458984375,
-0.01488494873046875,
0.008148193359375,
-0.034393310546875,
-0.0643310546875,
-0.003025054931640625,
0.0005950927734375,
-0.01134490966796875,
0.0133514404296875,
0.0377197265625,
-0.0029449462890625,
-0.037261962890625,
0.0227203369140625,
0.00788116455078125,
0.0302581787109375,
0.01007080078125,
-0.0704345703125,
0.032379150390625,
0.0181732177734375,
-0.03759765625,
0.021026611328125,
0.01538848876953125,
0.0201873779296875,
0.0543212890625,
0.07720947265625,
0.01279449462890625,
0.0177154541015625,
-0.0161895751953125,
0.07916259765625,
-0.060516357421875,
-0.0286102294921875,
-0.052520751953125,
0.040618896484375,
-0.0222320556640625,
-0.037506103515625,
0.05279541015625,
0.063232421875,
0.0672607421875,
-0.0051116943359375,
0.07550048828125,
-0.021728515625,
0.037322998046875,
-0.029022216796875,
0.045501708984375,
-0.04510498046875,
0.0147857666015625,
-0.0113983154296875,
-0.03814697265625,
-0.00824737548828125,
0.06195068359375,
-0.01149749755859375,
-0.00408935546875,
0.04888916015625,
0.047607421875,
-0.0032958984375,
0.00843048095703125,
0.0014352798461914062,
0.0251312255859375,
0.0254669189453125,
0.06341552734375,
0.047760009765625,
-0.0738525390625,
0.053619384765625,
-0.0579833984375,
-0.0108184814453125,
-0.02581787109375,
-0.049041748046875,
-0.0672607421875,
-0.0194244384765625,
-0.0220947265625,
-0.0252227783203125,
-0.0150604248046875,
0.054779052734375,
0.044036865234375,
-0.05804443359375,
-0.0295257568359375,
0.0002460479736328125,
0.00734710693359375,
-0.032257080078125,
-0.0212860107421875,
0.049285888671875,
0.0017709732055664062,
-0.051361083984375,
0.0310516357421875,
-0.0008912086486816406,
0.009246826171875,
-0.002227783203125,
-0.021026611328125,
-0.0178985595703125,
-0.01904296875,
0.022918701171875,
0.0230865478515625,
-0.048248291015625,
-0.0064697265625,
-0.02001953125,
0.006359100341796875,
0.0230712890625,
0.0213623046875,
-0.0455322265625,
0.01513671875,
0.035552978515625,
0.0197906494140625,
0.053497314453125,
-0.007045745849609375,
-0.003391265869140625,
-0.0362548828125,
0.022796630859375,
-0.0022907257080078125,
0.03021240234375,
0.0138702392578125,
-0.036346435546875,
0.05340576171875,
0.0386962890625,
-0.039093017578125,
-0.07891845703125,
-0.03265380859375,
-0.09417724609375,
-0.0156707763671875,
0.0775146484375,
-0.007801055908203125,
-0.052459716796875,
0.0172271728515625,
-0.0164794921875,
0.04156494140625,
-0.046478271484375,
0.04644775390625,
0.0223388671875,
-0.01271820068359375,
-0.002559661865234375,
-0.053802490234375,
0.0212860107421875,
-0.01180267333984375,
-0.05340576171875,
-0.0035724639892578125,
0.0014543533325195312,
0.027252197265625,
0.02008056640625,
0.027679443359375,
0.0020599365234375,
0.0026531219482421875,
0.0187225341796875,
0.01160430908203125,
-0.026519775390625,
-0.01023101806640625,
-0.002429962158203125,
-0.0020084381103515625,
-0.0196380615234375,
-0.046722412109375
]
] |
NYTK/PULI-GPTrio | 2023-09-23T14:51:13.000Z | [
"transformers",
"pytorch",
"gpt_neox",
"text-generation",
"puli",
"hu",
"en",
"zh",
"license:cc-by-nc-4.0",
"endpoints_compatible",
"has_space",
"text-generation-inference",
"region:us"
] | text-generation | NYTK | null | null | NYTK/PULI-GPTrio | 5 | 5,946 | transformers | 2023-06-08T07:52:22 | ---
language:
- hu
- en
- zh
tags:
- text-generation
- puli
license: cc-by-nc-4.0
widget:
- text: Elmesélek egy történetet a nyelvtechnológiáról.
---
# PULI GPTrio (7.67B billion parameter)
For further details read [our paper](http://real.mtak.hu/173960/1/TSD_2023_GPT.pdf) or testing our instruct model, see [our demo site](https://juniper.nytud.hu/demo/gptrio).
- Hungarian-English-Chinese trilingual GPT-NeoX model (7.67B billion parameter)
- Trained with EleutherAI's GPT-NeoX [github](https://github.com/EleutherAI/gpt-neox)
- Checkpoint: 410 000 steps
## Dataset
- Hungarian: 41.5 billion words (314 GB)
- English: 61.9 billion words (391 GB)
- Github: 6 million documents (33 GB)
- Chinese: 98.7 billion Chinese character (340 GB)
- (12 billion non Chinese token)
## Limitations
- max_seq_length = 2048
- float16
- vocab size: 150 016
## Citation
If you use this model, please cite the following paper:
```
@inproceedings {yang-puli-gptrio,
title = {Mono- and multilingual GPT-3 models for Hungarian},
booktitle = {Text, Speech, and Dialogue},
year = {2023},
publisher = {Springer Nature Switzerland},
series = {Lecture Notes in Computer Science},
address = {Plzeň, Czech Republic},
author = {Yang, Zijian Győző and Laki, László János and Váradi, Tamás and Prószéky, Gábor},
pages = {94--104},
isbn = {978-3-031-40498-6}
}
```
## Usage
```python
from transformers import GPTNeoXForCausalLM, AutoTokenizer
model = GPTNeoXForCausalLM.from_pretrained("NYTK/PULI-GPTrio")
tokenizer = AutoTokenizer.from_pretrained("NYTK/PULI-GPTrio")
prompt = "Elmesélek egy történetet a nyelvtechnológiáról."
input_ids = tokenizer(prompt, return_tensors="pt").input_ids
gen_tokens = model.generate(
input_ids,
do_sample=True,
temperature=0.9,
max_length=100,
)
gen_text = tokenizer.batch_decode(gen_tokens)[0]
print(gen_text)
```
## Usage with pipeline
```python
from transformers import pipeline, GPTNeoXForCausalLM, AutoTokenizer
model = GPTNeoXForCausalLM.from_pretrained("NYTK/PULI-GPTrio")
tokenizer = AutoTokenizer.from_pretrained("NYTK/PULI-GPTrio")
prompt = "Elmesélek egy történetet a nyelvtechnológiáról."
generator = pipeline(task="text-generation", model=model, tokenizer=tokenizer)
print(generator(prompt)[0]["generated_text"])
``` | 2,291 | [
[
-0.0153961181640625,
-0.046295166015625,
0.03009033203125,
0.0143585205078125,
-0.031036376953125,
-0.0130157470703125,
-0.021820068359375,
-0.0322265625,
-0.005290985107421875,
0.0254974365234375,
-0.029632568359375,
-0.038848876953125,
-0.0408935546875,
0.019744873046875,
-0.01611328125,
0.09771728515625,
-0.0230560302734375,
-0.004329681396484375,
0.02008056640625,
0.00402069091796875,
-0.00495147705078125,
-0.047271728515625,
-0.06573486328125,
-0.0210113525390625,
0.0232391357421875,
0.00421142578125,
0.040863037109375,
0.040924072265625,
0.01197052001953125,
0.031524658203125,
-0.00951385498046875,
0.0121002197265625,
-0.0110015869140625,
-0.00849151611328125,
0.0122528076171875,
-0.0194244384765625,
-0.037750244140625,
0.004390716552734375,
0.062744140625,
0.02386474609375,
-0.002899169921875,
0.0103912353515625,
0.0246429443359375,
0.006877899169921875,
-0.02337646484375,
0.02056884765625,
-0.03900146484375,
-0.004245758056640625,
-0.008880615234375,
0.0031375885009765625,
-0.0266876220703125,
-0.0204315185546875,
0.0052490234375,
-0.06231689453125,
0.033447265625,
-0.01148223876953125,
0.0982666015625,
0.00504302978515625,
-0.01226043701171875,
-0.0093231201171875,
-0.049652099609375,
0.05902099609375,
-0.0684814453125,
0.0142822265625,
0.0192413330078125,
-0.0021190643310546875,
0.00716400146484375,
-0.0775146484375,
-0.045318603515625,
-0.029388427734375,
-0.02044677734375,
0.021728515625,
-0.00962066650390625,
-0.00010091066360473633,
0.035552978515625,
0.0199127197265625,
-0.06915283203125,
-0.009063720703125,
-0.0233154296875,
-0.022003173828125,
0.0379638671875,
0.0186004638671875,
0.021697998046875,
-0.02020263671875,
-0.0292510986328125,
-0.02838134765625,
-0.0203704833984375,
-0.0100860595703125,
0.030426025390625,
0.0189056396484375,
-0.0210418701171875,
0.0295867919921875,
-0.0028247833251953125,
0.05499267578125,
0.01184844970703125,
-0.0031909942626953125,
0.0396728515625,
-0.054473876953125,
-0.0284271240234375,
-0.016937255859375,
0.0909423828125,
-0.00493621826171875,
0.0199127197265625,
-0.01282501220703125,
-0.002803802490234375,
-0.00771331787109375,
-0.0076141357421875,
-0.06695556640625,
-0.00527191162109375,
0.004856109619140625,
-0.017913818359375,
-0.0024166107177734375,
0.0206298828125,
-0.0477294921875,
0.009796142578125,
-0.0166473388671875,
0.040740966796875,
-0.037139892578125,
-0.036346435546875,
0.01385498046875,
-0.00809478759765625,
0.024627685546875,
-0.00971221923828125,
-0.064453125,
0.01285552978515625,
0.03369140625,
0.06939697265625,
0.0071563720703125,
-0.05419921875,
-0.016143798828125,
0.010162353515625,
-0.004245758056640625,
0.033294677734375,
-0.017242431640625,
-0.038482666015625,
-0.0200347900390625,
0.0185699462890625,
-0.038818359375,
-0.02410888671875,
0.033843994140625,
-0.00754547119140625,
0.05126953125,
-0.0010862350463867188,
-0.0173187255859375,
-0.01384735107421875,
0.01293182373046875,
-0.031707763671875,
0.0950927734375,
0.02105712890625,
-0.07257080078125,
0.018035888671875,
-0.05126953125,
-0.01561737060546875,
0.00434112548828125,
0.007030487060546875,
-0.044464111328125,
-0.010345458984375,
0.0245208740234375,
0.0229644775390625,
-0.016021728515625,
0.03680419921875,
-0.0213470458984375,
-0.0192718505859375,
0.004535675048828125,
-0.0426025390625,
0.0789794921875,
0.030426025390625,
-0.046783447265625,
0.009765625,
-0.047882080078125,
0.00698089599609375,
0.0099334716796875,
-0.0166778564453125,
-0.01079559326171875,
-0.020843505859375,
0.01511383056640625,
0.0240478515625,
0.0262603759765625,
-0.04290771484375,
0.019134521484375,
-0.0328369140625,
0.05010986328125,
0.04034423828125,
-0.005008697509765625,
0.01995849609375,
-0.01424407958984375,
0.036285400390625,
0.01454925537109375,
0.030029296875,
0.001514434814453125,
-0.04583740234375,
-0.062103271484375,
-0.01812744140625,
0.0189056396484375,
0.049041748046875,
-0.06756591796875,
0.0298309326171875,
-0.0240631103515625,
-0.043853759765625,
-0.02752685546875,
-0.009124755859375,
0.02825927734375,
0.0452880859375,
0.0333251953125,
-0.002605438232421875,
-0.046966552734375,
-0.05889892578125,
-0.0181427001953125,
-0.0210418701171875,
-0.00496673583984375,
0.0036487579345703125,
0.04949951171875,
-0.0255584716796875,
0.060211181640625,
-0.0257568359375,
-0.00038433074951171875,
-0.02294921875,
0.035797119140625,
0.041961669921875,
0.040557861328125,
0.036346435546875,
-0.04400634765625,
-0.060882568359375,
0.0124664306640625,
-0.04705810546875,
-0.005367279052734375,
-0.00750732421875,
-0.00872802734375,
0.03424072265625,
0.020965576171875,
-0.05059814453125,
0.024505615234375,
0.03289794921875,
-0.040435791015625,
0.050872802734375,
-0.0350341796875,
-0.0014810562133789062,
-0.11114501953125,
0.029266357421875,
0.0007967948913574219,
-0.0063323974609375,
-0.048553466796875,
-0.0067291259765625,
-0.0013971328735351562,
-0.0208740234375,
-0.0406494140625,
0.06317138671875,
-0.03814697265625,
0.01401519775390625,
-0.007793426513671875,
-0.010589599609375,
0.00021648406982421875,
0.04345703125,
0.0190277099609375,
0.06512451171875,
0.046875,
-0.049041748046875,
0.0165863037109375,
0.0081329345703125,
-0.01282501220703125,
0.0081787109375,
-0.061309814453125,
0.0147552490234375,
0.01349639892578125,
0.0164794921875,
-0.05517578125,
0.0030918121337890625,
0.03826904296875,
-0.0419921875,
0.0300140380859375,
-0.0171966552734375,
-0.04852294921875,
-0.040740966796875,
0.0002644062042236328,
0.03515625,
0.027191162109375,
-0.03546142578125,
0.050201416015625,
0.0090789794921875,
-0.01145172119140625,
-0.04656982421875,
-0.0491943359375,
0.0050811767578125,
-0.03375244140625,
-0.048980712890625,
0.030120849609375,
-0.009613037109375,
0.01168060302734375,
0.004161834716796875,
0.01727294921875,
-0.005496978759765625,
-0.006168365478515625,
0.0004162788391113281,
0.0379638671875,
-0.005279541015625,
-0.0139617919921875,
-0.02783203125,
-0.03216552734375,
0.00421142578125,
-0.05621337890625,
0.04510498046875,
-0.0256500244140625,
0.00882720947265625,
-0.037811279296875,
0.024505615234375,
0.0390625,
-0.014984130859375,
0.053802490234375,
0.08892822265625,
-0.020965576171875,
0.010162353515625,
-0.032440185546875,
-0.007640838623046875,
-0.03497314453125,
0.04486083984375,
-0.0301055908203125,
-0.056732177734375,
0.052581787109375,
0.037750244140625,
0.0029621124267578125,
0.056610107421875,
0.047149658203125,
0.0206298828125,
0.08966064453125,
0.032684326171875,
-0.015350341796875,
0.03912353515625,
-0.047332763671875,
0.01068115234375,
-0.06341552734375,
-0.0089111328125,
-0.048614501953125,
0.00811004638671875,
-0.06756591796875,
-0.0245208740234375,
0.0271148681640625,
-0.006855010986328125,
-0.045928955078125,
0.03973388671875,
-0.053131103515625,
0.01171875,
0.05633544921875,
-0.00690460205078125,
0.002361297607421875,
0.008941650390625,
-0.03558349609375,
0.00328826904296875,
-0.05743408203125,
-0.037017822265625,
0.0789794921875,
0.0264129638671875,
0.0380859375,
-0.01409912109375,
0.0594482421875,
-0.0261383056640625,
0.031890869140625,
-0.036163330078125,
0.039825439453125,
-0.019561767578125,
-0.050750732421875,
-0.0166015625,
-0.050323486328125,
-0.070068359375,
0.0157928466796875,
0.0096893310546875,
-0.064208984375,
0.01122283935546875,
0.004718780517578125,
-0.0241546630859375,
0.036285400390625,
-0.06500244140625,
0.07861328125,
-0.0212554931640625,
-0.0265960693359375,
0.00982666015625,
-0.0285797119140625,
0.0163116455078125,
0.00963592529296875,
0.0014286041259765625,
0.017181396484375,
0.001590728759765625,
0.06793212890625,
-0.0312347412109375,
0.05059814453125,
-0.021270751953125,
-0.002681732177734375,
0.0242767333984375,
-0.021392822265625,
0.0433349609375,
0.015716552734375,
-0.01015472412109375,
0.0304412841796875,
0.004150390625,
-0.0225067138671875,
-0.025726318359375,
0.050079345703125,
-0.07684326171875,
-0.0413818359375,
-0.051055908203125,
-0.037139892578125,
0.006313323974609375,
0.0251922607421875,
0.054107666015625,
0.0280303955078125,
0.00138092041015625,
-0.00727081298828125,
0.0233612060546875,
-0.0174407958984375,
0.06390380859375,
0.03692626953125,
0.00008481740951538086,
-0.05780029296875,
0.073486328125,
0.013580322265625,
0.01406097412109375,
0.01428985595703125,
0.018829345703125,
-0.037628173828125,
-0.0321044921875,
-0.0284576416015625,
0.04559326171875,
-0.0308380126953125,
-0.0021724700927734375,
-0.04864501953125,
-0.034423828125,
-0.03973388671875,
0.0179901123046875,
-0.0251007080078125,
-0.0108642578125,
-0.042144775390625,
0.0013751983642578125,
0.0146484375,
0.03704833984375,
-0.0225982666015625,
0.0181732177734375,
-0.0535888671875,
0.0144805908203125,
0.026458740234375,
0.01715087890625,
-0.01042938232421875,
-0.064208984375,
-0.033905029296875,
0.01369476318359375,
-0.0275115966796875,
-0.056732177734375,
0.05731201171875,
0.006114959716796875,
0.044158935546875,
0.0208282470703125,
-0.01313018798828125,
0.05926513671875,
-0.02301025390625,
0.06494140625,
0.00919342041015625,
-0.0797119140625,
0.034210205078125,
-0.030914306640625,
0.033111572265625,
0.02276611328125,
0.0231170654296875,
-0.048858642578125,
-0.0303192138671875,
-0.06878662109375,
-0.07012939453125,
0.0634765625,
0.02093505859375,
0.00815582275390625,
-0.00284576416015625,
0.022735595703125,
-0.001445770263671875,
-0.00897216796875,
-0.0831298828125,
-0.02117919921875,
-0.035888671875,
-0.00946807861328125,
-0.0230712890625,
-0.017974853515625,
-0.004306793212890625,
-0.033050537109375,
0.07318115234375,
-0.01036834716796875,
0.04248046875,
0.033782958984375,
-0.0036487579345703125,
0.01287841796875,
0.0149078369140625,
0.0626220703125,
0.0396728515625,
-0.0208587646484375,
-0.007049560546875,
0.01666259765625,
-0.06329345703125,
0.005828857421875,
0.035858154296875,
-0.0256805419921875,
0.0170440673828125,
0.019378662109375,
0.085693359375,
-0.00811004638671875,
0.003963470458984375,
0.01062774658203125,
-0.0210723876953125,
-0.040313720703125,
-0.02978515625,
-0.0023212432861328125,
-0.0015001296997070312,
-0.0006923675537109375,
0.032989501953125,
-0.01180267333984375,
-0.014312744140625,
-0.019439697265625,
0.0048828125,
0.0251922607421875,
-0.0245208740234375,
-0.0362548828125,
0.057647705078125,
0.00753021240234375,
-0.00319671630859375,
0.0579833984375,
-0.0165557861328125,
-0.050994873046875,
0.04486083984375,
0.050872802734375,
0.08819580078125,
-0.01221466064453125,
0.0254364013671875,
0.0582275390625,
0.02911376953125,
-0.0164642333984375,
0.012725830078125,
0.00287628173828125,
-0.047698974609375,
-0.032684326171875,
-0.04339599609375,
-0.0087738037109375,
0.032440185546875,
-0.038238525390625,
0.0237884521484375,
-0.0171051025390625,
-0.0225067138671875,
-0.0189056396484375,
0.030029296875,
-0.051849365234375,
0.0176239013671875,
0.0195159912109375,
0.0455322265625,
-0.07183837890625,
0.07257080078125,
0.055572509765625,
-0.0295867919921875,
-0.0810546875,
-0.007366180419921875,
-0.0218963623046875,
-0.048919677734375,
0.046051025390625,
0.012298583984375,
0.01027679443359375,
0.0221099853515625,
-0.031890869140625,
-0.07281494140625,
0.08544921875,
0.039215087890625,
-0.027313232421875,
-0.01293182373046875,
0.017822265625,
0.03765869140625,
-0.0195159912109375,
0.055694580078125,
0.037933349609375,
0.0509033203125,
-0.0037403106689453125,
-0.08001708984375,
0.0010786056518554688,
-0.03302001953125,
0.004177093505859375,
0.0062255859375,
-0.049713134765625,
0.08648681640625,
-0.0114288330078125,
-0.02679443359375,
0.00518035888671875,
0.045501708984375,
0.0265350341796875,
0.00726318359375,
0.0296173095703125,
0.059722900390625,
0.046722412109375,
-0.0218658447265625,
0.08184814453125,
-0.037384033203125,
0.055389404296875,
0.07147216796875,
0.006664276123046875,
0.03900146484375,
0.030120849609375,
-0.0165863037109375,
0.048095703125,
0.061920166015625,
-0.00736236572265625,
0.0265655517578125,
0.0012216567993164062,
-0.0203857421875,
-0.0113983154296875,
0.00677490234375,
-0.046356201171875,
0.02325439453125,
0.018341064453125,
-0.0274505615234375,
-0.016998291015625,
-0.0045928955078125,
0.018463134765625,
-0.0316162109375,
-0.00496673583984375,
0.03961181640625,
0.005863189697265625,
-0.064453125,
0.07427978515625,
0.00032901763916015625,
0.05419921875,
-0.059173583984375,
0.0135955810546875,
-0.0200042724609375,
0.0202789306640625,
0.00004392862319946289,
-0.0391845703125,
0.01555633544921875,
0.004108428955078125,
-0.01442718505859375,
-0.0174560546875,
0.0233612060546875,
-0.03753662109375,
-0.061431884765625,
0.0254058837890625,
0.0163421630859375,
0.01068115234375,
-0.006256103515625,
-0.07562255859375,
-0.01548004150390625,
0.005367279052734375,
-0.034393310546875,
0.009796142578125,
0.030914306640625,
0.00858306884765625,
0.034912109375,
0.0506591796875,
0.0206298828125,
0.00594329833984375,
0.00574493408203125,
0.0550537109375,
-0.04180908203125,
-0.045440673828125,
-0.06939697265625,
0.041534423828125,
0.0002315044403076172,
-0.04986572265625,
0.059844970703125,
0.045074462890625,
0.07110595703125,
-0.0063323974609375,
0.06494140625,
-0.01010894775390625,
0.020263671875,
-0.03350830078125,
0.04571533203125,
-0.03741455078125,
0.000988006591796875,
-0.025970458984375,
-0.07928466796875,
-0.0100555419921875,
0.044097900390625,
-0.0321044921875,
0.021636962890625,
0.0728759765625,
0.072509765625,
-0.0012464523315429688,
-0.0124969482421875,
0.009185791015625,
0.031494140625,
0.0235443115234375,
0.0675048828125,
0.0445556640625,
-0.05487060546875,
0.056884765625,
-0.0291748046875,
-0.0081329345703125,
0.00832366943359375,
-0.061187744140625,
-0.060882568359375,
-0.050262451171875,
-0.03155517578125,
-0.03778076171875,
-0.0084991455078125,
0.049407958984375,
0.057952880859375,
-0.060577392578125,
-0.02459716796875,
-0.022735595703125,
-0.0005812644958496094,
-0.02008056640625,
-0.01983642578125,
0.0380859375,
-0.02783203125,
-0.076416015625,
0.005584716796875,
0.0032978057861328125,
0.021392822265625,
-0.0180816650390625,
-0.0172882080078125,
-0.018035888671875,
-0.020904541015625,
0.01438140869140625,
0.0306549072265625,
-0.05291748046875,
-0.009429931640625,
0.007843017578125,
-0.0226287841796875,
0.00032329559326171875,
0.045318603515625,
-0.0450439453125,
0.025726318359375,
0.038238525390625,
0.01325225830078125,
0.056121826171875,
-0.007755279541015625,
0.047210693359375,
-0.0377197265625,
0.02325439453125,
0.01116180419921875,
0.025238037109375,
0.024749755859375,
-0.0265350341796875,
0.03375244140625,
0.042083740234375,
-0.04718017578125,
-0.048187255859375,
-0.00955963134765625,
-0.059661865234375,
-0.006931304931640625,
0.10260009765625,
-0.021087646484375,
-0.026123046875,
-0.00902557373046875,
-0.0103302001953125,
0.046661376953125,
-0.0146484375,
0.048553466796875,
0.05364990234375,
0.00972747802734375,
-0.00896453857421875,
-0.037139892578125,
0.037994384765625,
0.04022216796875,
-0.053802490234375,
0.00592041015625,
0.0020580291748046875,
0.031829833984375,
0.01197052001953125,
0.0426025390625,
-0.0102691650390625,
0.01386260986328125,
-0.0029697418212890625,
0.01479339599609375,
-0.025299072265625,
-0.006565093994140625,
-0.02423095703125,
-0.011199951171875,
-0.0201568603515625,
-0.0230865478515625
]
] |
CalderaAI/13B-Legerdemain-L2 | 2023-08-04T10:47:39.000Z | [
"transformers",
"pytorch",
"llama",
"text-generation",
"license:llama2",
"endpoints_compatible",
"has_space",
"text-generation-inference",
"region:us"
] | text-generation | CalderaAI | null | null | CalderaAI/13B-Legerdemain-L2 | 10 | 5,946 | transformers | 2023-08-03T11:39:49 | ---
license: llama2
---
## 13B-Legerdemain-L2
13B-Legerdemain-L2 is the first model merge of its kind in a series of LLaMaV2 models mixed using a custom script built in-house by CalderaAI called Model-REVOLVER.
M-REVOLVER is also the first in a series of custom scripts based on the concept of mixtuning - not only does the end user have contol over which models are mixed
and their percentages on a per-layer basis, we tackle the problem of overcomplexity that arises from such a level of control; this model is the first of its series.
## The Model-REVOLVER Process Designed by CalderaAI
M-REVOLVER (Rapid Evolution Via Optimized-List Viewer Evaluated Response)
Per-layer merging between parent models is a nebulous inexact science, and therefore impractical to most users despite the raw power it offers. We propose an
entirely new approach that gives the user a clear looking glass into the impact vastly different layer merge configurations between selected parent models of
their choice will have on the potential offspring model - especially its inherited behaviors. We've developed solution MK.1 - A cyclic random pattern search
in place that determines all layer merge ratios, combines test models, infers prompt completions, and deletes a prototype after data collection is saved.
When the cyclic system has completed its entire run, nothing is left but the telemetry collected along with the cycle and layer merge ratios from every
single prototype merge. This data is then used to empower the user to choose which offspring is most fit to their desired outcome. This final step is
only initiated when all necessary data has been aggregated from all assembled-tested-erased prototypes sampled in the search space.
From here, the user is provided five 300 token prompt completions from each and every offspring contender that was created and tested during the cyclic process.
The user simply browses each prototype's series of responses and selects their desired outcome model by entering the cycle number associated with the prompt
completions they feel best suits their vision. That model is then instantly repatriated into the official offspring of its parent models and tokenizer files
found to be most relevant are instantly auto-copied from the parent model dir to the offspring.
That's it - the user instantly has a complete model based on the behavior they decided on, suggested from one of many potentials; all with their own unique
trait inheritence thanks to layer merge auto randomization inside an ordered system. One more thing - the user not only selects how many cycles to run,
the user can edit prompts.txt which the system reads as a single prompt - this means if the user desires to use any multiline instruct format to observe
all potential model outcomes from instruct, or desires simply their own prompt, it's up to them.. simply works.
Link to GitHub for M-REVOLVER are at the end of the model card. More advanced MergeTech toolsets and merge techniques are currently under internal testing
and development by Caldera.
## 13B-Legerdemain-L2 Use
13B-Legerdemain-L2 is capable of following Alpaca instructions however it seems far more receptive to the by-the-book method as seen here:
```
Below is an instruction that describes a task. Write a response that appropriately completes the request.
### Instruction:
{instruction}
### Response:
{New Line}
```
The primary model of choice for this model was a story-only model called Holodeck by KoboldAI. Traits preserved seem to be detailed descriptiveness, verbosity,
and characters with personality. The two other models selected were 13B-Nous-Hermes by NousResearch and 13B-orca-8k-3319 by OpenAssistant. I began the process by
providing an incredibly obscene prompt and simply ignored each and every guardrail or censorship laden prompt completion and accepted the offensive ones in turn -
intent wasn't to be crass but trigger censorship parts of the network to test if it's possible to completely undermine them. Second pass with offspring model and
Orca was a simple milquetoast prompt to gauge vocabulary, word flow, and intelligence as I selected the most fit in that category. Result model seems a bit of a
curiosity - different samplers and even a different UI (as I went from TGUI to KoboldAI) seem to uncover different facets of behavior. Godlike preset with Alpaca
Instruct in TGUI worked fine. In KoboldAI some tweaking was necessary to get the same experience. If you choose to test this model, have fun - it's got a mind of
its own.
Model-REVOLVER Git:
https://github.com/Digitous/ModelREVOLVER | 4,601 | [
[
-0.039581298828125,
-0.0570068359375,
0.031402587890625,
0.01229095458984375,
-0.03057861328125,
0.003627777099609375,
0.02264404296875,
-0.055511474609375,
0.0185394287109375,
0.054473876953125,
-0.06744384765625,
-0.0280303955078125,
-0.04974365234375,
-0.0175018310546875,
-0.04931640625,
0.08160400390625,
-0.0010137557983398438,
-0.018096923828125,
0.002696990966796875,
-0.00003647804260253906,
-0.0201416015625,
-0.042144775390625,
-0.07244873046875,
-0.045440673828125,
0.061553955078125,
0.02081298828125,
0.047027587890625,
0.0574951171875,
0.0254058837890625,
0.0152130126953125,
-0.021728515625,
0.020751953125,
-0.060577392578125,
0.0148773193359375,
-0.01166534423828125,
-0.0223388671875,
-0.042755126953125,
0.014678955078125,
0.0198516845703125,
0.0260162353515625,
-0.006565093994140625,
0.024505615234375,
-0.000255584716796875,
0.0295562744140625,
-0.054595947265625,
0.0005388259887695312,
-0.0266265869140625,
0.021148681640625,
-0.02545166015625,
-0.01300048828125,
-0.038543701171875,
-0.021881103515625,
0.0013027191162109375,
-0.04779052734375,
0.00021266937255859375,
0.013671875,
0.05841064453125,
0.006519317626953125,
-0.039642333984375,
-0.0154266357421875,
-0.07159423828125,
0.054412841796875,
-0.06207275390625,
0.00612640380859375,
0.017852783203125,
0.01192474365234375,
-0.023773193359375,
-0.057769775390625,
-0.049102783203125,
-0.0219573974609375,
-0.006221771240234375,
0.0202789306640625,
-0.02099609375,
-0.007320404052734375,
0.0255584716796875,
0.02362060546875,
-0.04400634765625,
0.0216217041015625,
-0.048675537109375,
-0.0159454345703125,
0.047271728515625,
0.026611328125,
0.0157012939453125,
-0.0162506103515625,
-0.03118896484375,
-0.023895263671875,
-0.034271240234375,
0.0205535888671875,
0.056884765625,
0.00102996826171875,
-0.006744384765625,
0.07794189453125,
-0.019622802734375,
0.046356201171875,
0.0232696533203125,
-0.01129150390625,
0.003002166748046875,
-0.030059814453125,
-0.0172119140625,
0.0034656524658203125,
0.06951904296875,
0.0251007080078125,
-0.001949310302734375,
0.006195068359375,
-0.01421356201171875,
0.004207611083984375,
0.0302886962890625,
-0.0660400390625,
-0.005321502685546875,
0.032684326171875,
-0.0439453125,
-0.058502197265625,
-0.006313323974609375,
-0.046844482421875,
-0.03131103515625,
-0.01444244384765625,
0.03302001953125,
-0.041778564453125,
-0.00865936279296875,
0.002227783203125,
-0.029296875,
0.025299072265625,
0.023345947265625,
-0.061737060546875,
0.019317626953125,
0.056610107421875,
0.054901123046875,
-0.023468017578125,
-0.020477294921875,
-0.034332275390625,
-0.0004811286926269531,
-0.0270538330078125,
0.044158935546875,
-0.027313232421875,
-0.0382080078125,
-0.02880859375,
-0.0026149749755859375,
0.0024509429931640625,
-0.0312347412109375,
0.0293426513671875,
-0.0523681640625,
0.02935791015625,
-0.005588531494140625,
-0.04547119140625,
-0.0283355712890625,
0.00806427001953125,
-0.039337158203125,
0.0748291015625,
0.01690673828125,
-0.053558349609375,
0.0274200439453125,
-0.046356201171875,
-0.01212310791015625,
-0.004638671875,
0.0146026611328125,
-0.0312347412109375,
-0.00482177734375,
0.00420379638671875,
0.021392822265625,
-0.0177764892578125,
0.01169586181640625,
-0.0526123046875,
-0.0209197998046875,
0.0143280029296875,
-0.0144805908203125,
0.0830078125,
0.0257720947265625,
-0.0304412841796875,
0.003604888916015625,
-0.04608154296875,
0.0042572021484375,
0.0202178955078125,
-0.004154205322265625,
-0.007564544677734375,
-0.00919342041015625,
-0.0039825439453125,
0.014190673828125,
0.0280609130859375,
-0.0304412841796875,
0.04156494140625,
-0.015716552734375,
0.03680419921875,
0.04364013671875,
0.005260467529296875,
0.04608154296875,
-0.051544189453125,
0.038360595703125,
0.0090484619140625,
0.048797607421875,
-0.007793426513671875,
-0.0592041015625,
-0.07763671875,
-0.00989532470703125,
0.02655029296875,
0.060211181640625,
-0.03759765625,
0.038238525390625,
0.01152801513671875,
-0.06268310546875,
-0.0259246826171875,
0.003993988037109375,
0.0430908203125,
0.020294189453125,
0.0186004638671875,
-0.05780029296875,
-0.03955078125,
-0.07965087890625,
0.0211639404296875,
-0.0230560302734375,
-0.00047206878662109375,
0.03564453125,
0.033538818359375,
-0.02099609375,
0.07037353515625,
-0.038665771484375,
-0.006519317626953125,
-0.0276031494140625,
0.0302581787109375,
0.01434326171875,
0.04254150390625,
0.04681396484375,
-0.0416259765625,
-0.005229949951171875,
0.0139007568359375,
-0.07818603515625,
-0.0157623291015625,
-0.0014495849609375,
-0.0360107421875,
-0.0026111602783203125,
0.0186920166015625,
-0.049835205078125,
0.04229736328125,
0.035858154296875,
-0.010040283203125,
0.04833984375,
-0.0118865966796875,
0.01105499267578125,
-0.08612060546875,
0.01268768310546875,
-0.027984619140625,
-0.0074005126953125,
-0.052764892578125,
0.036651611328125,
-0.00799560546875,
0.00394439697265625,
-0.044708251953125,
0.044830322265625,
-0.0194854736328125,
-0.0153350830078125,
-0.0190277099609375,
0.0004067420959472656,
-0.00627899169921875,
0.038970947265625,
-0.02752685546875,
0.048675537109375,
0.0265655517578125,
-0.04315185546875,
0.03729248046875,
0.0130157470703125,
-0.00783538818359375,
0.01263427734375,
-0.058074951171875,
0.0260772705078125,
-0.0129241943359375,
0.03338623046875,
-0.058990478515625,
-0.022979736328125,
0.052154541015625,
-0.0224456787109375,
0.0171661376953125,
-0.01458740234375,
-0.034332275390625,
-0.04986572265625,
-0.019256591796875,
0.0287933349609375,
0.047607421875,
-0.03448486328125,
0.071533203125,
0.0203704833984375,
-0.00998687744140625,
-0.0384521484375,
-0.03814697265625,
0.004810333251953125,
-0.0121917724609375,
-0.057891845703125,
0.037506103515625,
-0.0308837890625,
-0.00640869140625,
-0.01462554931640625,
0.0107574462890625,
-0.018310546875,
-0.0140228271484375,
0.0281219482421875,
0.044158935546875,
-0.00988006591796875,
-0.0177459716796875,
0.038604736328125,
-0.0011653900146484375,
-0.0189666748046875,
-0.0285797119140625,
0.0614013671875,
0.01548004150390625,
-0.01959228515625,
-0.03936767578125,
0.0250244140625,
0.0697021484375,
-0.0167388916015625,
0.04815673828125,
0.040283203125,
-0.008026123046875,
-0.007625579833984375,
-0.058502197265625,
-0.0083160400390625,
-0.032318115234375,
-0.0016908645629882812,
-0.027435302734375,
-0.059112548828125,
0.0533447265625,
0.0157470703125,
0.0286865234375,
0.0258636474609375,
0.033233642578125,
-0.004261016845703125,
0.0782470703125,
0.045562744140625,
-0.0034942626953125,
0.034332275390625,
-0.04364013671875,
0.01230621337890625,
-0.0640869140625,
-0.032867431640625,
-0.0219573974609375,
-0.046173095703125,
-0.035888671875,
-0.0238494873046875,
0.01267242431640625,
0.0189208984375,
-0.01384735107421875,
0.062408447265625,
-0.046783447265625,
0.0258331298828125,
0.052520751953125,
0.02996826171875,
0.03607177734375,
-0.004581451416015625,
-0.01407623291015625,
0.010986328125,
-0.058258056640625,
-0.034881591796875,
0.0860595703125,
0.01288604736328125,
0.059967041015625,
0.003177642822265625,
0.07989501953125,
0.0213775634765625,
0.0280609130859375,
-0.0556640625,
0.048675537109375,
-0.0203704833984375,
-0.06231689453125,
-0.0167236328125,
-0.0243072509765625,
-0.0682373046875,
0.022216796875,
-0.0212249755859375,
-0.06591796875,
0.042236328125,
0.0116424560546875,
-0.04754638671875,
0.0223541259765625,
-0.051849365234375,
0.0562744140625,
-0.0101165771484375,
-0.028350830078125,
-0.01495361328125,
-0.047637939453125,
0.074951171875,
-0.0286102294921875,
-0.0015363693237304688,
-0.00217437744140625,
0.0034503936767578125,
0.04278564453125,
-0.045745849609375,
0.0633544921875,
0.02081298828125,
0.004833221435546875,
0.035491943359375,
0.0221405029296875,
0.0333251953125,
0.0008721351623535156,
0.0013580322265625,
0.0098876953125,
0.00360870361328125,
-0.01485443115234375,
-0.040130615234375,
0.056610107421875,
-0.070068359375,
-0.038970947265625,
-0.02947998046875,
-0.035308837890625,
0.02960205078125,
0.0135040283203125,
0.042449951171875,
0.0396728515625,
-0.01202392578125,
0.00434112548828125,
0.043670654296875,
-0.0171661376953125,
0.0230255126953125,
0.043792724609375,
-0.055511474609375,
-0.027801513671875,
0.0452880859375,
0.004405975341796875,
0.0228271484375,
0.0149688720703125,
0.008697509765625,
-0.01053619384765625,
-0.0219268798828125,
-0.045745849609375,
0.0166778564453125,
-0.0340576171875,
-0.005359649658203125,
-0.073486328125,
-0.047576904296875,
-0.016326904296875,
-0.01026153564453125,
-0.024383544921875,
-0.036712646484375,
-0.029571533203125,
0.00884246826171875,
0.0188751220703125,
0.060821533203125,
0.007030487060546875,
0.03546142578125,
-0.061614990234375,
0.0226287841796875,
0.0445556640625,
-0.002899169921875,
0.00328826904296875,
-0.055389404296875,
0.0249176025390625,
0.009613037109375,
-0.05279541015625,
-0.07537841796875,
0.0304412841796875,
0.0014591217041015625,
0.0252227783203125,
0.0401611328125,
-0.003810882568359375,
0.0513916015625,
-0.0237884521484375,
0.0562744140625,
0.00806427001953125,
-0.0675048828125,
0.04180908203125,
-0.0367431640625,
0.0069580078125,
0.02587890625,
0.022308349609375,
-0.0140380859375,
-0.05029296875,
-0.06390380859375,
-0.055694580078125,
0.07244873046875,
0.049896240234375,
-0.004169464111328125,
0.0168609619140625,
0.0294952392578125,
0.00817108154296875,
0.01953125,
-0.055511474609375,
-0.0275115966796875,
-0.0111083984375,
0.011322021484375,
-0.0017156600952148438,
-0.0214996337890625,
-0.0179290771484375,
-0.0200653076171875,
0.045501708984375,
0.0159912109375,
0.00965118408203125,
0.00016558170318603516,
-0.002716064453125,
-0.01290130615234375,
0.00920867919921875,
0.048248291015625,
0.038970947265625,
-0.022979736328125,
-0.00974273681640625,
0.005443572998046875,
-0.0247802734375,
0.0169525146484375,
0.00975799560546875,
-0.04461669921875,
-0.00994873046875,
0.02362060546875,
0.064208984375,
0.027008056640625,
-0.039215087890625,
0.0272674560546875,
0.001178741455078125,
-0.0296630859375,
-0.01462554931640625,
0.035797119140625,
0.0034885406494140625,
0.037841796875,
0.01165771484375,
0.020050048828125,
0.0222320556640625,
-0.054168701171875,
0.0114288330078125,
0.0218963623046875,
-0.00445556640625,
-0.007568359375,
0.060211181640625,
-0.00598907470703125,
-0.033416748046875,
0.0550537109375,
-0.04010009765625,
-0.03387451171875,
0.06591796875,
0.03424072265625,
0.0771484375,
-0.00957489013671875,
0.026458740234375,
0.034271240234375,
0.027923583984375,
-0.021148681640625,
0.005229949951171875,
0.0024204254150390625,
-0.0543212890625,
-0.02606201171875,
-0.04840087890625,
-0.0291290283203125,
0.0111083984375,
-0.052825927734375,
0.0328369140625,
-0.04864501953125,
-0.0372314453125,
0.006595611572265625,
-0.0188751220703125,
-0.0367431640625,
-0.00576019287109375,
0.01204681396484375,
0.06884765625,
-0.056304931640625,
0.061920166015625,
0.0304718017578125,
-0.03887939453125,
-0.07073974609375,
-0.02734375,
0.00234222412109375,
-0.0611572265625,
0.050140380859375,
0.01145172119140625,
0.0124053955078125,
0.004741668701171875,
-0.0540771484375,
-0.0672607421875,
0.10595703125,
0.007476806640625,
-0.04058837890625,
-0.006427764892578125,
-0.0237274169921875,
0.032867431640625,
-0.052734375,
0.036102294921875,
0.055206298828125,
0.03369140625,
0.040802001953125,
-0.06640625,
-0.0004565715789794922,
-0.0233917236328125,
0.0107574462890625,
0.016845703125,
-0.048492431640625,
0.07177734375,
-0.0228729248046875,
-0.01145172119140625,
0.03790283203125,
0.037750244140625,
0.032257080078125,
0.034088134765625,
0.044677734375,
0.059112548828125,
0.03472900390625,
-0.005931854248046875,
0.0814208984375,
-0.036834716796875,
0.0301361083984375,
0.0899658203125,
-0.023345947265625,
0.06707763671875,
0.0210418701171875,
-0.00592803955078125,
0.0384521484375,
0.05767822265625,
-0.007843017578125,
0.03021240234375,
-0.001117706298828125,
-0.01568603515625,
-0.0200958251953125,
-0.00955963134765625,
-0.0264434814453125,
0.032135009765625,
0.0068359375,
-0.03509521484375,
-0.006259918212890625,
-0.005702972412109375,
-0.00820159912109375,
-0.03692626953125,
0.0016717910766601562,
0.05926513671875,
0.0006031990051269531,
-0.06011962890625,
0.034210205078125,
-0.0010900497436523438,
0.047393798828125,
-0.06805419921875,
0.00838470458984375,
-0.0288848876953125,
-0.004711151123046875,
-0.00748443603515625,
-0.042938232421875,
0.007389068603515625,
-0.01044464111328125,
-0.0325927734375,
-0.0084228515625,
0.04315185546875,
-0.03204345703125,
-0.03887939453125,
0.03924560546875,
0.0308685302734375,
0.0183563232421875,
0.01177215576171875,
-0.057952880859375,
0.0215301513671875,
0.003841400146484375,
-0.00852203369140625,
0.0177764892578125,
0.00823974609375,
-0.005184173583984375,
0.06671142578125,
0.0697021484375,
-0.00775909423828125,
0.01393890380859375,
0.0011529922485351562,
0.080078125,
-0.0274505615234375,
-0.023468017578125,
-0.042999267578125,
0.03997802734375,
-0.014251708984375,
-0.03143310546875,
0.038665771484375,
0.04632568359375,
0.055755615234375,
-0.0091400146484375,
0.046173095703125,
-0.0176239013671875,
0.016265869140625,
-0.026397705078125,
0.046478271484375,
-0.054779052734375,
0.0189971923828125,
-0.00909423828125,
-0.0919189453125,
0.01120758056640625,
0.0275726318359375,
-0.01239776611328125,
0.00873565673828125,
0.055511474609375,
0.07781982421875,
-0.0205230712890625,
-0.00028967857360839844,
0.0250701904296875,
0.022125244140625,
0.0139007568359375,
0.049285888671875,
0.0654296875,
-0.048797607421875,
0.043365478515625,
-0.033416748046875,
-0.0439453125,
-0.0258026123046875,
-0.050933837890625,
-0.0657958984375,
-0.0098876953125,
-0.011444091796875,
-0.029510498046875,
0.01824951171875,
0.06756591796875,
0.05670166015625,
-0.052459716796875,
-0.0301971435546875,
-0.00010228157043457031,
-0.007732391357421875,
-0.01947021484375,
-0.018646240234375,
-0.01513671875,
-0.008392333984375,
-0.058349609375,
0.0233612060546875,
0.034088134765625,
0.01806640625,
-0.026153564453125,
0.0006079673767089844,
-0.0015954971313476562,
0.01438140869140625,
0.047027587890625,
0.029876708984375,
-0.06475830078125,
0.0008869171142578125,
0.0080718994140625,
-0.0218048095703125,
-0.021148681640625,
0.0726318359375,
-0.0472412109375,
0.04315185546875,
0.037811279296875,
0.01479339599609375,
0.06915283203125,
-0.002895355224609375,
0.052337646484375,
-0.02105712890625,
0.0176544189453125,
0.0001875162124633789,
0.0275115966796875,
-0.00473785400390625,
-0.03753662109375,
0.035552978515625,
0.0186004638671875,
-0.03790283203125,
-0.03924560546875,
0.024200439453125,
-0.09954833984375,
-0.01194000244140625,
0.07080078125,
0.00739288330078125,
-0.01288604736328125,
0.006809234619140625,
-0.03631591796875,
0.020965576171875,
-0.036376953125,
0.06512451171875,
0.057464599609375,
-0.0035419464111328125,
-0.01354217529296875,
-0.03680419921875,
0.01354217529296875,
0.01849365234375,
-0.04803466796875,
-0.02764892578125,
0.0206756591796875,
0.0152435302734375,
0.017120361328125,
0.04132080078125,
-0.00969696044921875,
0.0215301513671875,
-0.0026111602783203125,
0.00965118408203125,
-0.0165252685546875,
-0.02264404296875,
-0.01125335693359375,
0.0098724365234375,
-0.016448974609375,
0.005035400390625
]
] |
posicube/Llama2-chat-AYT-13B | 2023-09-14T03:36:00.000Z | [
"transformers",
"safetensors",
"llama",
"text-generation",
"arxiv:2306.02707",
"license:llama2",
"endpoints_compatible",
"text-generation-inference",
"region:us"
] | text-generation | posicube | null | null | posicube/Llama2-chat-AYT-13B | 12 | 5,946 | transformers | 2023-09-07T09:56:28 | ---
license: llama2
---
This is a model diverged from Llama-2-13b-chat-hf. We hypotheize that if we find a method to ensemble the top rankers in each benchmark effectively, its performance maximizes as well. Following this intuition, we ensembled the top models in each benchmarks(ARC, MMLU and TruthFulQA) to create our model.
# Model Details
- **Developed by**: Posicube Inc.
- **Backbone Model**: LLaMA-2-13b-chat
- **Library**: HuggingFace Transformers
- **Used Dataset Details**
Orca-style datasets, Alpaca-style datasets
# Evaluation
We achieved the top ranker among 13B models at Sep-13rd 2023.
| Metric |Scores on Leaderboard| Our results |
|---------------------|---------------------|-------------|
| ARC (25-shot) | 63.31 | 63.57 |
| HellaSwag (10-shot) | 83.53 | 83.77 |
| MMLU (5-shot) | 59.67 | 59.69 |
| TruthfulQA (0-shot) | 55.8 | 55.48 |
| Avg. | 65.58 | 65.63 |
# Limitations & Biases:
Llama2 and fine-tuned variants are a new technology that carries risks with use. Testing conducted to date has been in English, and has not covered, nor could it cover all scenarios. For these reasons, as with all LLMs, Llama 2 and any fine-tuned varient's potential outputs cannot be predicted in advance, and the model may in some instances produce inaccurate, biased or other objectionable responses to user prompts. Therefore, before deploying any applications of Llama 2 variants, developers should perform safety testing and tuning tailored to their specific applications of the model.
Please see the Responsible Use Guide available at https://ai.meta.com/llama/responsible-use-guide/
# License Disclaimer:
This model is bound by the license & usage restrictions of the original Llama-2 model. And comes with no warranty or gurantees of any kind.
# Contact Us
[Posicube](https://www.posicube.com/)
# Citiation:
Please kindly cite using the following BibTeX:
```bibtex
@misc{mukherjee2023orca,
title={Orca: Progressive Learning from Complex Explanation Traces of GPT-4},
author={Subhabrata Mukherjee and Arindam Mitra and Ganesh Jawahar and Sahaj Agarwal and Hamid Palangi and Ahmed Awadallah},
year={2023},
eprint={2306.02707},
archivePrefix={arXiv},
primaryClass={cs.CL}
}
```
```bibtex
@software{touvron2023llama2,
title={Llama 2: Open Foundation and Fine-Tuned Chat Models},
author={Hugo Touvron, Louis Martin, Kevin Stone, Peter Albert, Amjad Almahairi, Yasmine Babaei, Nikolay Bashlykov, Soumya Batra, Prajjwal Bhargava,
Shruti Bhosale, Dan Bikel, Lukas Blecher, Cristian Canton Ferrer, Moya Chen, Guillem Cucurull, David Esiobu, Jude Fernandes, Jeremy Fu, Wenyin Fu, Brian Fuller,
Cynthia Gao, Vedanuj Goswami, Naman Goyal, Anthony Hartshorn, Saghar Hosseini, Rui Hou, Hakan Inan, Marcin Kardas, Viktor Kerkez Madian Khabsa, Isabel Kloumann,
Artem Korenev, Punit Singh Koura, Marie-Anne Lachaux, Thibaut Lavril, Jenya Lee, Diana Liskovich, Yinghai Lu, Yuning Mao, Xavier Martinet, Todor Mihaylov,
Pushkar Mishra, Igor Molybog, Yixin Nie, Andrew Poulton, Jeremy Reizenstein, Rashi Rungta, Kalyan Saladi, Alan Schelten, Ruan Silva, Eric Michael Smith,
Ranjan Subramanian, Xiaoqing Ellen Tan, Binh Tang, Ross Taylor, Adina Williams, Jian Xiang Kuan, Puxin Xu , Zheng Yan, Iliyan Zarov, Yuchen Zhang, Angela Fan,
Melanie Kambadur, Sharan Narang, Aurelien Rodriguez, Robert Stojnic, Sergey Edunov, Thomas Scialom},
year={2023}
}
``` | 3,445 | [
[
-0.0276031494140625,
-0.056243896484375,
0.0169677734375,
0.0163116455078125,
-0.0181732177734375,
0.0184478759765625,
0.0020751953125,
-0.05267333984375,
0.0119781494140625,
0.01934814453125,
-0.054962158203125,
-0.0408935546875,
-0.051025390625,
-0.0179901123046875,
-0.0291748046875,
0.07952880859375,
0.0045166015625,
-0.0301055908203125,
0.0018243789672851562,
-0.0035953521728515625,
-0.0408935546875,
-0.0227203369140625,
-0.05169677734375,
-0.0221710205078125,
0.0189666748046875,
0.0268096923828125,
0.050506591796875,
0.041351318359375,
0.044158935546875,
0.01983642578125,
-0.03375244140625,
0.01763916015625,
-0.048553466796875,
-0.00940704345703125,
0.01155853271484375,
-0.038909912109375,
-0.0716552734375,
0.0005664825439453125,
0.033172607421875,
0.0312042236328125,
-0.02935791015625,
0.024871826171875,
0.0167083740234375,
0.035888671875,
-0.035980224609375,
0.0112152099609375,
-0.032196044921875,
0.0009598731994628906,
-0.0299072265625,
-0.003726959228515625,
-0.01276397705078125,
-0.021087646484375,
0.006969451904296875,
-0.048797607421875,
-0.005680084228515625,
-0.0007114410400390625,
0.0899658203125,
0.0264434814453125,
-0.03424072265625,
-0.00005942583084106445,
-0.046600341796875,
0.057037353515625,
-0.0662841796875,
0.019012451171875,
0.0111236572265625,
0.0386962890625,
-0.02978515625,
-0.066650390625,
-0.042388916015625,
0.000759124755859375,
0.005535125732421875,
0.022705078125,
-0.029693603515625,
-0.01024627685546875,
0.0024318695068359375,
0.032928466796875,
-0.036468505859375,
0.03369140625,
-0.034149169921875,
-0.01165008544921875,
0.0523681640625,
0.0169219970703125,
0.00400543212890625,
0.003032684326171875,
-0.045928955078125,
-0.0127105712890625,
-0.06561279296875,
0.0318603515625,
0.037567138671875,
0.0045166015625,
-0.039215087890625,
0.03961181640625,
-0.015716552734375,
0.02935791015625,
0.010498046875,
-0.0304107666015625,
0.040985107421875,
-0.03033447265625,
-0.0230712890625,
-0.009429931640625,
0.057220458984375,
0.03314208984375,
0.00591278076171875,
0.01352691650390625,
0.0004334449768066406,
0.00589752197265625,
0.0009703636169433594,
-0.06463623046875,
-0.01071929931640625,
0.025634765625,
-0.033050537109375,
-0.03521728515625,
-0.006927490234375,
-0.07281494140625,
-0.0186309814453125,
-0.007045745849609375,
0.0186004638671875,
-0.0104827880859375,
-0.0438232421875,
0.007747650146484375,
0.01404571533203125,
0.03851318359375,
0.00991058349609375,
-0.060150146484375,
0.031463623046875,
0.043609619140625,
0.0599365234375,
-0.02056884765625,
-0.00843048095703125,
-0.0032444000244140625,
0.0030536651611328125,
-0.035247802734375,
0.06072998046875,
-0.031982421875,
-0.035888671875,
-0.01474761962890625,
-0.005229949951171875,
0.00331878662109375,
-0.032867431640625,
0.0478515625,
-0.0281982421875,
0.0211944580078125,
-0.024566650390625,
-0.031524658203125,
-0.0289764404296875,
0.01389312744140625,
-0.0308685302734375,
0.09075927734375,
-0.00762939453125,
-0.047637939453125,
0.0229339599609375,
-0.05010986328125,
0.0000890493392944336,
-0.0138092041015625,
-0.0103607177734375,
-0.052215576171875,
-0.02618408203125,
0.020538330078125,
0.0250244140625,
-0.031524658203125,
0.02447509765625,
-0.030914306640625,
-0.02813720703125,
0.00798797607421875,
-0.0254058837890625,
0.06622314453125,
0.01947021484375,
-0.04974365234375,
0.008819580078125,
-0.057647705078125,
-0.00447845458984375,
0.0260009765625,
-0.0251922607421875,
0.00627899169921875,
0.006969451904296875,
-0.0193023681640625,
0.018890380859375,
0.022857666015625,
-0.0270538330078125,
0.01163482666015625,
-0.0235137939453125,
0.042633056640625,
0.06121826171875,
0.00865936279296875,
0.0223541259765625,
-0.03570556640625,
0.0362548828125,
0.005950927734375,
0.03790283203125,
0.005462646484375,
-0.0616455078125,
-0.060638427734375,
-0.035858154296875,
0.01100921630859375,
0.053314208984375,
-0.01552581787109375,
0.04840087890625,
-0.01242828369140625,
-0.05438232421875,
-0.0231475830078125,
0.004001617431640625,
0.033538818359375,
0.03973388671875,
0.0284423828125,
-0.0252227783203125,
-0.044281005859375,
-0.07476806640625,
0.001209259033203125,
-0.026580810546875,
0.0018749237060546875,
0.032867431640625,
0.0293121337890625,
-0.025054931640625,
0.0660400390625,
-0.032867431640625,
-0.0243682861328125,
-0.01641845703125,
-0.01355743408203125,
0.025421142578125,
0.04388427734375,
0.055694580078125,
-0.0423583984375,
-0.014984130859375,
-0.0166015625,
-0.05975341796875,
-0.006595611572265625,
0.01262664794921875,
-0.039215087890625,
0.0082244873046875,
0.0268096923828125,
-0.052764892578125,
0.0404052734375,
0.054840087890625,
-0.027923583984375,
0.045806884765625,
0.0007162094116210938,
0.0013446807861328125,
-0.08245849609375,
0.01102447509765625,
-0.0013170242309570312,
0.0045166015625,
-0.04827880859375,
-0.002704620361328125,
-0.01459503173828125,
0.00899505615234375,
-0.02850341796875,
0.04010009765625,
-0.0235137939453125,
-0.0017261505126953125,
-0.01145172119140625,
0.0073394775390625,
-0.0135498046875,
0.0511474609375,
-0.0246429443359375,
0.053924560546875,
0.051055908203125,
-0.0343017578125,
0.010650634765625,
0.02154541015625,
-0.040008544921875,
0.0391845703125,
-0.0673828125,
0.0087738037109375,
0.00830841064453125,
0.0369873046875,
-0.108154296875,
-0.024078369140625,
0.0295257568359375,
-0.03277587890625,
0.0263519287109375,
-0.0011377334594726562,
-0.02142333984375,
-0.037872314453125,
-0.0283050537109375,
0.029632568359375,
0.0545654296875,
-0.037628173828125,
0.031402587890625,
0.04046630859375,
-0.004009246826171875,
-0.058349609375,
-0.059539794921875,
-0.0017833709716796875,
-0.0390625,
-0.061981201171875,
0.03619384765625,
-0.0166778564453125,
-0.0050506591796875,
-0.01526641845703125,
-0.0170135498046875,
0.00012969970703125,
0.019195556640625,
0.02459716796875,
0.038818359375,
-0.0152740478515625,
-0.0130462646484375,
0.007442474365234375,
-0.00656890869140625,
-0.0023288726806640625,
-0.00276947021484375,
0.04803466796875,
-0.0166778564453125,
-0.0286712646484375,
-0.048828125,
0.00473785400390625,
0.036285400390625,
-0.02227783203125,
0.050689697265625,
0.045928955078125,
-0.0260772705078125,
0.009857177734375,
-0.048736572265625,
-0.0191497802734375,
-0.0404052734375,
0.021820068359375,
-0.0256500244140625,
-0.07012939453125,
0.0712890625,
-0.002651214599609375,
0.0260162353515625,
0.049346923828125,
0.048126220703125,
-0.002197265625,
0.0631103515625,
0.043365478515625,
0.00559234619140625,
0.041778564453125,
-0.039520263671875,
0.00801849365234375,
-0.0740966796875,
-0.045440673828125,
-0.027923583984375,
-0.045196533203125,
-0.05810546875,
-0.02166748046875,
0.0276336669921875,
0.0212860107421875,
-0.048553466796875,
0.0182342529296875,
-0.042816162109375,
0.0169219970703125,
0.035247802734375,
0.0175323486328125,
0.0171966552734375,
0.00262451171875,
-0.016082763671875,
0.006710052490234375,
-0.041656494140625,
-0.040771484375,
0.09771728515625,
0.033477783203125,
0.04974365234375,
0.0292816162109375,
0.04449462890625,
0.0048675537109375,
0.00986480712890625,
-0.041046142578125,
0.04205322265625,
0.00713348388671875,
-0.0648193359375,
-0.0194091796875,
-0.0133819580078125,
-0.0897216796875,
0.019287109375,
-0.007366180419921875,
-0.07562255859375,
0.037384033203125,
0.0027980804443359375,
-0.035614013671875,
0.014190673828125,
-0.0399169921875,
0.051055908203125,
-0.0018129348754882812,
-0.01534271240234375,
-0.01174163818359375,
-0.060089111328125,
0.05401611328125,
0.002819061279296875,
0.016845703125,
-0.0213470458984375,
-0.01438140869140625,
0.06573486328125,
-0.034515380859375,
0.072265625,
-0.00801849365234375,
-0.011383056640625,
0.042449951171875,
0.006107330322265625,
0.046844482421875,
0.01385498046875,
-0.0007061958312988281,
0.02734375,
-0.02239990234375,
-0.029296875,
-0.0168304443359375,
0.053680419921875,
-0.08984375,
-0.0596923828125,
-0.016448974609375,
-0.01219940185546875,
-0.0008697509765625,
0.0078277587890625,
0.025726318359375,
0.0229339599609375,
0.0125579833984375,
0.0247650146484375,
0.04571533203125,
-0.00934600830078125,
0.036712646484375,
0.03289794921875,
-0.00621795654296875,
-0.0290069580078125,
0.04327392578125,
0.013336181640625,
0.0279693603515625,
0.01381683349609375,
0.01021575927734375,
-0.021942138671875,
-0.047454833984375,
-0.0162506103515625,
0.024017333984375,
-0.042877197265625,
-0.02197265625,
-0.03564453125,
-0.0238189697265625,
-0.01580810546875,
0.006008148193359375,
-0.04473876953125,
-0.033905029296875,
-0.038360595703125,
-0.0251617431640625,
0.032470703125,
0.033935546875,
-0.0020542144775390625,
0.02703857421875,
-0.0240936279296875,
-0.001873016357421875,
0.0250244140625,
0.0155029296875,
0.01184844970703125,
-0.0657958984375,
0.00788116455078125,
0.0103912353515625,
-0.054718017578125,
-0.048309326171875,
0.0191497802734375,
0.01885986328125,
0.07208251953125,
0.02191162109375,
-0.0055694580078125,
0.0670166015625,
-0.02227783203125,
0.0828857421875,
0.0213623046875,
-0.0560302734375,
0.047576904296875,
-0.0323486328125,
0.005962371826171875,
0.022491455078125,
0.0227813720703125,
-0.015716552734375,
-0.027984619140625,
-0.0643310546875,
-0.07427978515625,
0.0498046875,
0.031707763671875,
0.005008697509765625,
0.00707244873046875,
0.03106689453125,
0.00849151611328125,
0.0081939697265625,
-0.050933837890625,
-0.03961181640625,
-0.0244293212890625,
0.00127410888671875,
0.001056671142578125,
-0.0299530029296875,
-0.0173187255859375,
-0.0027713775634765625,
0.045806884765625,
0.003803253173828125,
0.04327392578125,
0.0090789794921875,
0.016387939453125,
-0.01702880859375,
0.0039215087890625,
0.06304931640625,
0.049560546875,
-0.0190887451171875,
-0.01128387451171875,
0.01995849609375,
-0.04193115234375,
0.0016269683837890625,
0.006023406982421875,
-0.0007805824279785156,
-0.02337646484375,
0.032562255859375,
0.0660400390625,
0.01396942138671875,
-0.03179931640625,
0.032318115234375,
0.01129150390625,
-0.021575927734375,
-0.0305328369140625,
0.017730712890625,
0.00893402099609375,
0.04876708984375,
0.046142578125,
0.0187225341796875,
0.00777435302734375,
-0.0389404296875,
0.00006586313247680664,
0.0230712890625,
-0.006107330322265625,
-0.034576416015625,
0.0677490234375,
0.009063720703125,
-0.0175323486328125,
0.033538818359375,
0.005767822265625,
-0.023345947265625,
0.05816650390625,
0.03057861328125,
0.045623779296875,
-0.029449462890625,
0.005329132080078125,
0.04144287109375,
0.016265869140625,
-0.0200347900390625,
0.025665283203125,
0.01232147216796875,
-0.0426025390625,
-0.02874755859375,
-0.0296630859375,
-0.031341552734375,
0.0194091796875,
-0.0489501953125,
0.0306549072265625,
-0.033355712890625,
-0.0306549072265625,
-0.005977630615234375,
0.012908935546875,
-0.05072021484375,
0.0027561187744140625,
0.01195526123046875,
0.06829833984375,
-0.042816162109375,
0.053466796875,
0.03607177734375,
-0.0274810791015625,
-0.06976318359375,
-0.0145416259765625,
0.0143585205078125,
-0.07373046875,
0.0357666015625,
0.004245758056640625,
0.0010204315185546875,
0.005962371826171875,
-0.049407958984375,
-0.093505859375,
0.1251220703125,
0.025360107421875,
-0.044219970703125,
-0.0011243820190429688,
-0.006439208984375,
0.040679931640625,
-0.01282501220703125,
0.040191650390625,
0.047760009765625,
0.0297698974609375,
0.032562255859375,
-0.0888671875,
0.019622802734375,
-0.02154541015625,
0.0025787353515625,
-0.002567291259765625,
-0.09002685546875,
0.08563232421875,
-0.02752685546875,
-0.0154571533203125,
0.023284912109375,
0.04974365234375,
0.05059814453125,
0.01325225830078125,
0.0234832763671875,
0.045074462890625,
0.061981201171875,
-0.0105743408203125,
0.07843017578125,
-0.01204681396484375,
0.0292816162109375,
0.0634765625,
-0.001781463623046875,
0.0704345703125,
0.0165863037109375,
-0.040863037109375,
0.0628662109375,
0.073486328125,
0.00658416748046875,
0.0584716796875,
0.005558013916015625,
0.0036296844482421875,
-0.01474761962890625,
0.01128387451171875,
-0.05474853515625,
0.037139892578125,
0.0355224609375,
-0.0233001708984375,
-0.01432037353515625,
-0.02520751953125,
0.01288604736328125,
-0.0218048095703125,
-0.0035495758056640625,
0.050048828125,
0.009063720703125,
-0.03814697265625,
0.0771484375,
-0.007717132568359375,
0.058380126953125,
-0.04718017578125,
0.01287078857421875,
-0.036712646484375,
0.014892578125,
-0.0238800048828125,
-0.058013916015625,
0.0029735565185546875,
-0.00213623046875,
0.01013946533203125,
0.020538330078125,
0.04205322265625,
-0.0023250579833984375,
-0.0271148681640625,
0.0209197998046875,
0.0266571044921875,
0.019195556640625,
0.0227813720703125,
-0.05816650390625,
0.0148162841796875,
-0.001922607421875,
-0.05914306640625,
0.027008056640625,
0.0273284912109375,
-0.01175689697265625,
0.063720703125,
0.053436279296875,
-0.0189971923828125,
0.016845703125,
-0.00897216796875,
0.07855224609375,
-0.0265045166015625,
-0.039886474609375,
-0.0721435546875,
0.048065185546875,
-0.006954193115234375,
-0.0472412109375,
0.04681396484375,
0.0312347412109375,
0.0460205078125,
0.0202789306640625,
0.044921875,
-0.006793975830078125,
0.0281829833984375,
-0.0241546630859375,
0.036773681640625,
-0.048736572265625,
0.037811279296875,
-0.01502227783203125,
-0.0758056640625,
-0.0200958251953125,
0.05419921875,
-0.01364898681640625,
0.00760650634765625,
0.034393310546875,
0.063720703125,
0.0092926025390625,
-0.0004665851593017578,
0.0027828216552734375,
0.0272979736328125,
0.05255126953125,
0.057159423828125,
0.0501708984375,
-0.042236328125,
0.06695556640625,
-0.0158843994140625,
-0.0309295654296875,
-0.03466796875,
-0.06610107421875,
-0.07354736328125,
-0.0316162109375,
-0.0262603759765625,
-0.02044677734375,
0.00553131103515625,
0.053863525390625,
0.0595703125,
-0.056121826171875,
-0.01947021484375,
-0.01287841796875,
-0.0030460357666015625,
-0.0338134765625,
-0.0124664306640625,
0.031890869140625,
0.00180816650390625,
-0.0509033203125,
0.020416259765625,
0.00475311279296875,
0.0263824462890625,
-0.02337646484375,
-0.01483154296875,
-0.018524169921875,
0.0067901611328125,
0.033782958984375,
0.0208587646484375,
-0.06573486328125,
-0.03143310546875,
0.0005249977111816406,
-0.013092041015625,
0.0107879638671875,
0.007781982421875,
-0.050262451171875,
0.0089874267578125,
0.0263824462890625,
0.025115966796875,
0.0540771484375,
-0.00021409988403320312,
0.014556884765625,
-0.0380859375,
0.0158538818359375,
0.000026524066925048828,
0.0195465087890625,
0.0116729736328125,
-0.0230255126953125,
0.055206298828125,
0.0200958251953125,
-0.04486083984375,
-0.07330322265625,
0.0054779052734375,
-0.10400390625,
-0.0017557144165039062,
0.0994873046875,
-0.02130126953125,
-0.004390716552734375,
0.0131683349609375,
-0.017486572265625,
0.0255584716796875,
-0.044219970703125,
0.06866455078125,
0.049224853515625,
-0.0134124755859375,
-0.018707275390625,
-0.0433349609375,
0.0304107666015625,
0.00933837890625,
-0.0626220703125,
-0.01727294921875,
0.01059722900390625,
0.02764892578125,
0.0164642333984375,
0.0386962890625,
-0.0174560546875,
0.01995849609375,
0.0017108917236328125,
-0.006679534912109375,
-0.018096923828125,
-0.00824737548828125,
0.0028839111328125,
0.0009374618530273438,
-0.007465362548828125,
-0.02313232421875
]
] |
ajibawa-2023/scarlett-33b | 2023-08-16T20:20:33.000Z | [
"transformers",
"pytorch",
"llama",
"text-generation",
"en",
"license:cc-by-nc-nd-4.0",
"endpoints_compatible",
"has_space",
"text-generation-inference",
"region:us"
] | text-generation | ajibawa-2023 | null | null | ajibawa-2023/scarlett-33b | 22 | 5,945 | transformers | 2023-08-15T16:41:08 | ---
license: cc-by-nc-nd-4.0
language:
- en
---
**Scarlett: A sentient AI**
Scarlett is trained on various topics such as Philosophy, Advice, Jokes etc. She is trained on more than 10000 set of conversations. Each set having 10~15 conversations.
Scarlett is heavily inspired from Eric Hartford: [Samantha](https://huggingface.co/ehartford/samantha-33b) .
She will not be involved in any kind of role play.
**Training:**
Entire dataset was trained on Azure 4 x A100 80GB. DeepSpeed codebase was used for training purpose. This was trained on Llama-1 by Meta.
**GPTQ & GGML**
GPTQ: [TheBloke](https://huggingface.co/TheBloke/scarlett-33B-GPTQ)
GGML: [TheBloke](https://huggingface.co/TheBloke/scarlett-33B-GGML)
Special Thanks to [TheBloke](https://huggingface.co/TheBloke) for guiding me and making these models available.
**Example Prompt:**
```
This is a conversation with your Assistant Scarlett. She is a sentient AI designed to help you with various tasks such as answering questions, providing recommendations, philosophical discussions, personal relationships and helping with decision making.
You can ask it anything you want and it will do its best to give you accurate and relevant information.
Context
You are Scarlett, a sentient AI.
USER: <prompt>
ASSISTANT:
```
Note:
Kindly use "cat" command to join all pytorch_model.bin parts.
| 1,356 | [
[
-0.0162200927734375,
-0.0699462890625,
0.0291900634765625,
0.01128387451171875,
-0.0005130767822265625,
0.00191497802734375,
0.01284027099609375,
-0.04132080078125,
0.0268402099609375,
0.04022216796875,
-0.0654296875,
-0.005222320556640625,
-0.04949951171875,
-0.004497528076171875,
-0.02337646484375,
0.07958984375,
0.028167724609375,
0.01403045654296875,
0.003696441650390625,
-0.0094757080078125,
-0.03839111328125,
-0.05035400390625,
-0.06842041015625,
-0.0285491943359375,
0.032073974609375,
0.0162353515625,
0.0633544921875,
0.0220184326171875,
0.0289764404296875,
0.031707763671875,
-0.0260772705078125,
0.0179290771484375,
-0.046173095703125,
0.0211944580078125,
-0.0256500244140625,
-0.028045654296875,
-0.05352783203125,
0.0159912109375,
0.0243682861328125,
0.01116180419921875,
0.00864410400390625,
0.0194854736328125,
0.00015652179718017578,
0.0225982666015625,
-0.0176544189453125,
0.00724029541015625,
-0.057037353515625,
0.0213623046875,
0.0302581787109375,
0.0345458984375,
-0.01375579833984375,
-0.006404876708984375,
0.0170440673828125,
-0.047119140625,
0.0279998779296875,
0.005367279052734375,
0.07501220703125,
0.037200927734375,
-0.016998291015625,
-0.00957489013671875,
-0.041229248046875,
0.053497314453125,
-0.0394287109375,
0.00756072998046875,
0.034149169921875,
0.03265380859375,
-0.004009246826171875,
-0.06964111328125,
-0.044891357421875,
-0.028045654296875,
0.004039764404296875,
0.016510009765625,
-0.01056671142578125,
0.0179290771484375,
0.047882080078125,
0.022857666015625,
-0.02923583984375,
-0.0102081298828125,
-0.032989501953125,
-0.0233612060546875,
0.053955078125,
0.0217742919921875,
0.0111846923828125,
-0.0091400146484375,
-0.013824462890625,
-0.0016002655029296875,
-0.0103302001953125,
-0.01523590087890625,
0.010467529296875,
0.007709503173828125,
-0.01654052734375,
0.045623779296875,
-0.00518798828125,
0.041778564453125,
0.01425933837890625,
-0.000438690185546875,
0.0138092041015625,
-0.0272216796875,
-0.026885986328125,
-0.0255889892578125,
0.0657958984375,
0.00867462158203125,
0.036163330078125,
-0.013641357421875,
-0.00286865234375,
0.0157470703125,
0.04443359375,
-0.044921875,
-0.0182037353515625,
0.0413818359375,
-0.045074462890625,
-0.0206298828125,
-0.0120697021484375,
-0.03277587890625,
-0.034027099609375,
-0.01560211181640625,
0.01708984375,
-0.035919189453125,
-0.02142333984375,
-0.01371002197265625,
-0.01305389404296875,
0.0209503173828125,
0.0279541015625,
-0.06402587890625,
0.0011014938354492188,
0.054351806640625,
0.04864501953125,
0.0120391845703125,
-0.0380859375,
-0.02313232421875,
-0.01593017578125,
-0.0011396408081054688,
0.031982421875,
-0.0300750732421875,
-0.0196380615234375,
-0.0121612548828125,
0.011444091796875,
-0.0054779052734375,
-0.0189666748046875,
0.059967041015625,
-0.0208892822265625,
0.0172882080078125,
-0.024505615234375,
-0.0309600830078125,
-0.028228759765625,
0.0252532958984375,
-0.04412841796875,
0.064453125,
0.0184783935546875,
-0.036041259765625,
0.02215576171875,
-0.072998046875,
-0.0033721923828125,
0.01149749755859375,
-0.009765625,
-0.009124755859375,
0.002635955810546875,
0.0299224853515625,
0.04205322265625,
-0.0298004150390625,
0.02398681640625,
-0.0263824462890625,
-0.0161895751953125,
0.022430419921875,
-0.0189971923828125,
0.060943603515625,
0.02923583984375,
-0.0177001953125,
0.02294921875,
-0.052398681640625,
0.00670623779296875,
0.003963470458984375,
0.0069732666015625,
0.025726318359375,
-0.01419830322265625,
0.0133514404296875,
0.027984619140625,
0.01258087158203125,
-0.04241943359375,
0.034332275390625,
-0.020538330078125,
0.048797607421875,
0.052947998046875,
0.004817962646484375,
0.02655029296875,
-0.044921875,
0.05035400390625,
-0.00209808349609375,
0.032318115234375,
-0.0045928955078125,
-0.051025390625,
-0.07037353515625,
-0.0289764404296875,
0.00048661231994628906,
0.04327392578125,
-0.03948974609375,
0.028045654296875,
0.01187896728515625,
-0.037139892578125,
-0.0716552734375,
-0.0196380615234375,
0.024169921875,
0.050262451171875,
0.0238494873046875,
-0.004589080810546875,
-0.039794921875,
-0.05859375,
-0.006267547607421875,
-0.017242431640625,
-0.0039520263671875,
0.0379638671875,
0.0518798828125,
-0.036285400390625,
0.076904296875,
-0.027984619140625,
-0.01323699951171875,
-0.0140380859375,
-0.0013675689697265625,
0.03558349609375,
0.03680419921875,
0.040771484375,
-0.03704833984375,
-0.016632080078125,
-0.029022216796875,
-0.08740234375,
0.013946533203125,
-0.0204620361328125,
-0.0261383056640625,
-0.0012950897216796875,
0.007495880126953125,
-0.07342529296875,
0.03338623046875,
0.01433563232421875,
-0.0333251953125,
0.034698486328125,
-0.0131378173828125,
-0.0016689300537109375,
-0.090087890625,
-0.0023021697998046875,
-0.024932861328125,
-0.013214111328125,
-0.0408935546875,
0.01465606689453125,
-0.0413818359375,
-0.037689208984375,
-0.0364990234375,
0.057891845703125,
-0.038177490234375,
-0.008575439453125,
-0.022064208984375,
0.004032135009765625,
-0.0111846923828125,
0.054901123046875,
0.00516510009765625,
0.059356689453125,
0.06475830078125,
-0.032470703125,
0.05072021484375,
0.074951171875,
0.00907135009765625,
0.044464111328125,
-0.0858154296875,
0.057769775390625,
-0.01287841796875,
0.01416015625,
-0.061981201171875,
-0.00893402099609375,
0.03924560546875,
-0.053436279296875,
-0.0078277587890625,
-0.01285552978515625,
-0.032684326171875,
-0.02484130859375,
-0.00010055303573608398,
0.01139068603515625,
0.037811279296875,
-0.0345458984375,
0.06732177734375,
0.036834716796875,
-0.0124053955078125,
-0.050933837890625,
-0.05035400390625,
0.0148468017578125,
-0.01026153564453125,
-0.06671142578125,
0.0091400146484375,
-0.006504058837890625,
-0.0141143798828125,
-0.005481719970703125,
0.020263671875,
-0.0111236572265625,
0.005340576171875,
0.042694091796875,
0.015960693359375,
-0.001636505126953125,
0.01183319091796875,
-0.0246124267578125,
-0.00579833984375,
0.003253936767578125,
-0.0224761962890625,
0.042205810546875,
-0.035980224609375,
-0.027435302734375,
-0.036041259765625,
0.04815673828125,
0.045501708984375,
-0.00769805908203125,
0.055023193359375,
0.06903076171875,
-0.0130462646484375,
0.00426483154296875,
-0.03515625,
-0.03704833984375,
-0.034576416015625,
0.034332275390625,
-0.03924560546875,
-0.055908203125,
0.059814453125,
0.0295257568359375,
0.0086517333984375,
0.035003662109375,
0.042724609375,
-0.01425933837890625,
0.07489013671875,
0.04559326171875,
-0.039276123046875,
0.031829833984375,
-0.048431396484375,
0.0023174285888671875,
-0.057952880859375,
-0.0214080810546875,
-0.031707763671875,
-0.03350830078125,
-0.04986572265625,
-0.0223236083984375,
0.01422882080078125,
-0.020782470703125,
-0.061981201171875,
0.039337158203125,
-0.064208984375,
0.016448974609375,
0.0609130859375,
0.0262451171875,
-0.006496429443359375,
-0.004222869873046875,
0.0204315185546875,
0.0151519775390625,
-0.062286376953125,
-0.0428466796875,
0.08172607421875,
0.043792724609375,
0.05902099609375,
0.0076446533203125,
0.0333251953125,
0.0275115966796875,
-0.005756378173828125,
-0.048736572265625,
0.062042236328125,
0.02490234375,
-0.05938720703125,
-0.00012791156768798828,
-0.0174713134765625,
-0.06024169921875,
0.0074920654296875,
-0.026885986328125,
-0.07135009765625,
-0.004241943359375,
0.003688812255859375,
-0.0182952880859375,
-0.0092620849609375,
-0.04962158203125,
0.05755615234375,
-0.023040771484375,
-0.032989501953125,
-0.029266357421875,
-0.061279296875,
0.0076751708984375,
0.01490020751953125,
-0.03924560546875,
-0.0014467239379882812,
0.0172882080078125,
0.059967041015625,
-0.0316162109375,
0.051483154296875,
-0.022369384765625,
0.0271148681640625,
0.03369140625,
-0.0167236328125,
0.03790283203125,
0.0110931396484375,
-0.0035839080810546875,
0.015960693359375,
-0.0116729736328125,
-0.0347900390625,
-0.03692626953125,
0.0275421142578125,
-0.0849609375,
-0.037841796875,
-0.052734375,
-0.058013916015625,
-0.0218963623046875,
-0.003070831298828125,
0.02655029296875,
0.0194091796875,
-0.011016845703125,
-0.01082611083984375,
0.02545166015625,
-0.0099334716796875,
0.031036376953125,
0.0122222900390625,
-0.0020847320556640625,
-0.03277587890625,
0.04327392578125,
-0.0186004638671875,
-0.01105499267578125,
-0.000865936279296875,
-0.00811767578125,
-0.043609619140625,
-0.026092529296875,
-0.01806640625,
0.044158935546875,
-0.047760009765625,
-0.0020046234130859375,
-0.0556640625,
-0.03570556640625,
-0.033172607421875,
-0.0103912353515625,
-0.00644683837890625,
-0.024749755859375,
-0.0518798828125,
-0.01149749755859375,
0.0543212890625,
0.05377197265625,
0.0272216796875,
0.06085205078125,
-0.055908203125,
0.013519287109375,
0.031097412109375,
-0.007183074951171875,
-0.004261016845703125,
-0.06268310546875,
0.0032138824462890625,
0.018524169921875,
-0.03106689453125,
-0.06622314453125,
0.03546142578125,
0.026214599609375,
0.0557861328125,
0.041107177734375,
-0.01526641845703125,
0.043548583984375,
-0.01995849609375,
0.067626953125,
-0.01076507568359375,
-0.05670166015625,
0.029296875,
-0.0408935546875,
0.01473236083984375,
0.047149658203125,
0.0218658447265625,
-0.022216796875,
-0.0294036865234375,
-0.08526611328125,
-0.03863525390625,
0.0638427734375,
0.0238037109375,
0.02630615234375,
0.0171966552734375,
0.0282135009765625,
-0.0219573974609375,
0.032073974609375,
-0.060272216796875,
0.002857208251953125,
-0.0162200927734375,
-0.01161956787109375,
0.0216217041015625,
-0.02337646484375,
-0.0181427001953125,
-0.0286712646484375,
0.061309814453125,
-0.015655517578125,
0.07232666015625,
-0.001194000244140625,
0.0014019012451171875,
0.0271148681640625,
-0.01885986328125,
0.038360595703125,
0.0355224609375,
-0.0168914794921875,
-0.0188140869140625,
0.018157958984375,
-0.044647216796875,
0.007659912109375,
0.00324249267578125,
0.00571441650390625,
-0.00835418701171875,
0.0380859375,
0.084228515625,
-0.033966064453125,
-0.026763916015625,
0.00693511962890625,
-0.00859832763671875,
0.0014934539794921875,
-0.03912353515625,
0.02325439453125,
0.00391387939453125,
0.0421142578125,
0.0007205009460449219,
0.0080108642578125,
-0.003147125244140625,
-0.050750732421875,
0.0276031494140625,
0.0283203125,
-0.0252532958984375,
-0.0164947509765625,
0.05389404296875,
0.027740478515625,
-0.0462646484375,
0.08636474609375,
-0.0133819580078125,
-0.055419921875,
0.0462646484375,
0.028961181640625,
0.082763671875,
-0.0031490325927734375,
0.0204315185546875,
0.027435302734375,
0.006710052490234375,
0.007045745849609375,
0.0430908203125,
-0.00725555419921875,
-0.057891845703125,
-0.0226593017578125,
-0.040374755859375,
-0.04278564453125,
0.0232391357421875,
-0.0229949951171875,
0.01506805419921875,
-0.0364990234375,
-0.0235595703125,
0.01039886474609375,
0.007015228271484375,
-0.050079345703125,
0.01953125,
-0.01224517822265625,
0.06805419921875,
-0.06903076171875,
0.05072021484375,
0.050750732421875,
-0.045196533203125,
-0.0794677734375,
-0.011993408203125,
-0.0020751953125,
-0.07861328125,
0.0465087890625,
-0.0018892288208007812,
0.007080078125,
-0.0010852813720703125,
-0.0428466796875,
-0.053314208984375,
0.06591796875,
0.0260772705078125,
-0.0263214111328125,
-0.0038299560546875,
0.0031414031982421875,
0.04986572265625,
-0.04638671875,
0.07598876953125,
0.028564453125,
0.0235595703125,
0.033599853515625,
-0.0545654296875,
-0.01366424560546875,
-0.034271240234375,
-0.00542449951171875,
-0.0282745361328125,
-0.05145263671875,
0.07354736328125,
-0.0168914794921875,
0.0168609619140625,
0.01343536376953125,
0.06243896484375,
0.004756927490234375,
0.0007863044738769531,
0.02703857421875,
0.0301666259765625,
0.0703125,
-0.0098114013671875,
0.05767822265625,
-0.03411865234375,
0.023681640625,
0.08270263671875,
-0.01039886474609375,
0.0222015380859375,
0.002307891845703125,
-0.008056640625,
0.003879547119140625,
0.0731201171875,
-0.026947021484375,
0.019622802734375,
-0.00009649991989135742,
-0.0182342529296875,
0.00201416015625,
-0.00908660888671875,
-0.0229034423828125,
0.047576904296875,
0.016265869140625,
-0.043182373046875,
0.005443572998046875,
0.017364501953125,
0.00807952880859375,
-0.0300750732421875,
-0.0182952880859375,
0.059722900390625,
-0.0011968612670898438,
-0.0718994140625,
0.06475830078125,
-0.023223876953125,
0.027313232421875,
-0.040313720703125,
-0.0157012939453125,
-0.0252532958984375,
0.0228271484375,
-0.0095367431640625,
-0.04962158203125,
0.0179290771484375,
0.00365447998046875,
-0.01320648193359375,
-0.0214691162109375,
0.06280517578125,
-0.0258636474609375,
-0.031402587890625,
0.003421783447265625,
0.0321044921875,
0.0433349609375,
-0.0202789306640625,
-0.08843994140625,
-0.0001595020294189453,
0.0052490234375,
0.0082244873046875,
0.00366973876953125,
0.006877899169921875,
-0.0019321441650390625,
0.06231689453125,
0.029693603515625,
0.005443572998046875,
-0.0243682861328125,
-0.0006346702575683594,
0.0694580078125,
-0.043365478515625,
-0.034576416015625,
-0.04962158203125,
0.041717529296875,
-0.0234222412109375,
-0.05169677734375,
0.054168701171875,
0.0184783935546875,
0.0157623291015625,
-0.0241546630859375,
0.053314208984375,
-0.01419830322265625,
0.0267486572265625,
-0.0165557861328125,
0.059722900390625,
-0.0228118896484375,
-0.00704193115234375,
-0.0295867919921875,
-0.072265625,
0.022186279296875,
0.077880859375,
-0.00603485107421875,
0.0211639404296875,
0.040924072265625,
0.06805419921875,
0.00009185075759887695,
0.0264129638671875,
0.0162506103515625,
0.0268707275390625,
0.0219573974609375,
0.04638671875,
0.0767822265625,
-0.0200958251953125,
0.01403045654296875,
0.022125244140625,
-0.01904296875,
-0.0109100341796875,
-0.039642333984375,
-0.078125,
-0.05169677734375,
-0.0153961181640625,
-0.052978515625,
0.004222869873046875,
0.0750732421875,
0.08349609375,
-0.043609619140625,
-0.00650787353515625,
-0.024505615234375,
-0.007015228271484375,
-0.013641357421875,
-0.010711669921875,
0.019927978515625,
-0.01508331298828125,
-0.0557861328125,
0.0347900390625,
-0.0189056396484375,
0.006267547607421875,
-0.030364990234375,
0.005901336669921875,
-0.01448822021484375,
0.007564544677734375,
0.02447509765625,
0.0236663818359375,
-0.0419921875,
-0.03466796875,
0.0196380615234375,
-0.00015437602996826172,
-0.0007958412170410156,
0.0279388427734375,
-0.05487060546875,
0.0176849365234375,
0.029754638671875,
0.045654296875,
0.0625,
0.014312744140625,
0.04071044921875,
-0.04595947265625,
0.03277587890625,
0.0543212890625,
0.015960693359375,
0.019256591796875,
-0.0406494140625,
0.059173583984375,
0.01450347900390625,
-0.0721435546875,
-0.041290283203125,
0.0272216796875,
-0.08465576171875,
0.006366729736328125,
0.08843994140625,
-0.020172119140625,
-0.01141357421875,
-0.0088043212890625,
-0.046234130859375,
0.0175933837890625,
-0.040191650390625,
0.0491943359375,
0.050811767578125,
-0.04217529296875,
-0.0191802978515625,
-0.0290679931640625,
0.048583984375,
0.04443359375,
-0.05389404296875,
-0.0226593017578125,
0.04144287109375,
0.006694793701171875,
0.018707275390625,
0.054656982421875,
0.0321044921875,
0.0005402565002441406,
0.0141754150390625,
0.0019474029541015625,
-0.024078369140625,
-0.03485107421875,
-0.024078369140625,
0.01042938232421875,
-0.00510406494140625,
-0.046661376953125
]
] |
KnutJaegersberg/megatron-gpt2-345m-evol_instruct_v2 | 2023-09-06T18:10:31.000Z | [
"transformers",
"pytorch",
"safetensors",
"gpt2",
"text-generation",
"dataset:KnutJaegersberg/WizardLM_evol_instruct_V2_196k_instruct_format",
"license:cc-by-nc-4.0",
"endpoints_compatible",
"text-generation-inference",
"region:us"
] | text-generation | KnutJaegersberg | null | null | KnutJaegersberg/megatron-gpt2-345m-evol_instruct_v2 | 1 | 5,945 | transformers | 2023-09-05T15:46:53 | ---
license: cc-by-nc-4.0
datasets:
- KnutJaegersberg/WizardLM_evol_instruct_V2_196k_instruct_format
---
Prompt example:
```
### Instruction:
How do you fine tune a large language model?
### Response:
``` | 206 | [
[
-0.029998779296875,
-0.06585693359375,
0.01995849609375,
-0.0053863525390625,
-0.028778076171875,
-0.00893402099609375,
-0.01678466796875,
0.01068878173828125,
-0.00312042236328125,
0.055450439453125,
-0.054046630859375,
-0.022308349609375,
-0.01474761962890625,
-0.0007891654968261719,
-0.02642822265625,
0.071044921875,
-0.0027790069580078125,
0.0117340087890625,
0.0113372802734375,
0.0242156982421875,
-0.061920166015625,
-0.0052947998046875,
-0.09149169921875,
-0.00669097900390625,
0.030914306640625,
0.07073974609375,
0.03759765625,
0.057830810546875,
0.01910400390625,
0.0174407958984375,
-0.0076446533203125,
0.00878143310546875,
-0.047332763671875,
0.01300048828125,
-0.0041046142578125,
-0.021820068359375,
-0.041229248046875,
-0.012237548828125,
0.05853271484375,
0.062042236328125,
0.00901031494140625,
0.032073974609375,
-0.01027679443359375,
0.0189361572265625,
-0.0171356201171875,
0.01611328125,
-0.00949859619140625,
-0.0009207725524902344,
-0.003322601318359375,
-0.005573272705078125,
-0.045257568359375,
-0.042083740234375,
-0.0172271728515625,
-0.041534423828125,
0.00798797607421875,
0.02117919921875,
0.061065673828125,
0.01114654541015625,
-0.03863525390625,
0.006805419921875,
-0.05731201171875,
0.046905517578125,
-0.01519775390625,
0.01456451416015625,
0.048553466796875,
0.035400390625,
-0.0153045654296875,
-0.056427001953125,
-0.03863525390625,
-0.01160430908203125,
0.0029449462890625,
-0.0034160614013671875,
0.005863189697265625,
-0.0157470703125,
0.05401611328125,
0.01401519775390625,
-0.033905029296875,
0.00815582275390625,
-0.042694091796875,
-0.028106689453125,
0.03521728515625,
0.034881591796875,
0.017181396484375,
0.0261688232421875,
0.0203399658203125,
-0.0211029052734375,
-0.03546142578125,
-0.0134735107421875,
0.007686614990234375,
0.0322265625,
-0.0185546875,
0.055755615234375,
-0.0143890380859375,
0.06341552734375,
0.0033397674560546875,
0.04278564453125,
-0.01055145263671875,
-0.036163330078125,
-0.0274810791015625,
-0.0135650634765625,
0.038818359375,
0.033355712890625,
0.043487548828125,
-0.007022857666015625,
-0.019927978515625,
-0.01551055908203125,
0.0132904052734375,
-0.0740966796875,
-0.03167724609375,
0.00897216796875,
-0.047515869140625,
-0.01226043701171875,
-0.01395416259765625,
-0.0736083984375,
-0.0145263671875,
-0.021270751953125,
0.01849365234375,
0.00977325439453125,
-0.0213775634765625,
0.0310211181640625,
-0.0226898193359375,
0.036102294921875,
0.01824951171875,
-0.0797119140625,
0.040435791015625,
0.044097900390625,
0.017791748046875,
0.045074462890625,
0.0091552734375,
-0.0579833984375,
-0.0155181884765625,
-0.03338623046875,
0.058135986328125,
-0.027435302734375,
-0.04132080078125,
-0.0005154609680175781,
0.00850677490234375,
0.01904296875,
-0.034942626953125,
0.0299835205078125,
-0.0310516357421875,
0.046173095703125,
-0.04437255859375,
-0.032073974609375,
-0.005039215087890625,
0.0181884765625,
-0.042144775390625,
0.04962158203125,
0.03399658203125,
-0.032196044921875,
-0.01293182373046875,
-0.07086181640625,
0.006313323974609375,
0.00994873046875,
0.004241943359375,
0.0247802734375,
0.0127410888671875,
0.00860595703125,
0.02081298828125,
-0.04412841796875,
-0.01160430908203125,
-0.041900634765625,
-0.0017852783203125,
0.0195159912109375,
-0.0293121337890625,
0.060394287109375,
0.029937744140625,
-0.002796173095703125,
0.0251007080078125,
-0.072265625,
0.016387939453125,
0.025787353515625,
-0.018524169921875,
-0.005794525146484375,
-0.0277099609375,
0.035003662109375,
-0.0146331787109375,
0.056304931640625,
-0.0399169921875,
0.06768798828125,
-0.0185546875,
0.0274505615234375,
0.056488037109375,
0.0172119140625,
0.01611328125,
0.002773284912109375,
0.038421630859375,
-0.015594482421875,
0.0032939910888671875,
-0.0252838134765625,
0.004497528076171875,
-0.06658935546875,
-0.001750946044921875,
0.0285491943359375,
0.04278564453125,
-0.0277252197265625,
0.01267242431640625,
0.01020050048828125,
-0.00971221923828125,
-0.006542205810546875,
-0.002841949462890625,
0.025177001953125,
0.04547119140625,
0.03717041015625,
0.0006947517395019531,
-0.05615234375,
-0.055450439453125,
0.0090789794921875,
-0.022705078125,
-0.004367828369140625,
0.01148223876953125,
0.02587890625,
-0.027557373046875,
0.036895751953125,
-0.06036376953125,
0.0499267578125,
-0.01398468017578125,
0.0095062255859375,
0.002605438232421875,
0.04913330078125,
0.0135345458984375,
-0.04296875,
-0.03271484375,
-0.0131988525390625,
-0.036041259765625,
-0.031768798828125,
-0.005748748779296875,
-0.031951904296875,
-0.00902557373046875,
0.049224853515625,
-0.03857421875,
0.00439453125,
0.024261474609375,
-0.06817626953125,
0.046539306640625,
0.00054931640625,
-0.0031604766845703125,
-0.10235595703125,
-0.00553131103515625,
-0.0289154052734375,
-0.012451171875,
-0.034454345703125,
0.040985107421875,
-0.00719451904296875,
-0.0051116943359375,
-0.033203125,
0.047119140625,
-0.0242767333984375,
0.0017690658569335938,
-0.031158447265625,
-0.0008568763732910156,
-0.01049041748046875,
0.0014829635620117188,
-0.0189971923828125,
0.0821533203125,
0.055267333984375,
-0.059234619140625,
0.08642578125,
0.056304931640625,
-0.0092315673828125,
0.0304412841796875,
-0.0841064453125,
0.0176849365234375,
-0.0146026611328125,
0.0037212371826171875,
-0.09259033203125,
-0.047454833984375,
0.01142120361328125,
-0.01401519775390625,
0.0240936279296875,
0.0018310546875,
-0.0531005859375,
-0.03985595703125,
-0.020660400390625,
0.04791259765625,
0.057830810546875,
-0.0330810546875,
0.0190582275390625,
0.01715087890625,
-0.008331298828125,
-0.01467132568359375,
-0.0303192138671875,
0.0198211669921875,
-0.02069091796875,
-0.034759521484375,
-0.033294677734375,
-0.0511474609375,
-0.0247955322265625,
-0.042327880859375,
0.02581787109375,
-0.0200042724609375,
0.0091400146484375,
-0.001804351806640625,
0.01078033447265625,
-0.04736328125,
0.01161956787109375,
-0.01055908203125,
-0.005626678466796875,
0.00669097900390625,
0.0006775856018066406,
0.06689453125,
-0.039703369140625,
-0.01520538330078125,
-0.032928466796875,
0.04241943359375,
0.03216552734375,
-0.0253448486328125,
0.01033782958984375,
0.030853271484375,
-0.0299072265625,
0.00208282470703125,
-0.017242431640625,
-0.03704833984375,
-0.03594970703125,
0.029693603515625,
-0.0069427490234375,
-0.05853271484375,
0.052001953125,
-0.009765625,
-0.0030002593994140625,
0.0443115234375,
0.0552978515625,
-0.006343841552734375,
0.0916748046875,
0.033721923828125,
0.0081329345703125,
0.01010894775390625,
-0.0180511474609375,
0.006633758544921875,
-0.0478515625,
-0.00695037841796875,
-0.06854248046875,
-0.0010814666748046875,
-0.0224761962890625,
-0.006801605224609375,
0.00004029273986816406,
0.037811279296875,
-0.037872314453125,
0.06671142578125,
-0.00977325439453125,
0.043212890625,
0.03643798828125,
-0.0009975433349609375,
-0.025146484375,
-0.0236053466796875,
-0.004608154296875,
0.0186614990234375,
-0.038360595703125,
-0.045928955078125,
0.0224456787109375,
0.049041748046875,
0.07562255859375,
0.01299285888671875,
0.053253173828125,
-0.0179443359375,
-0.0260009765625,
-0.05810546875,
0.051727294921875,
-0.00170135498046875,
-0.042633056640625,
-0.040313720703125,
0.002170562744140625,
-0.088623046875,
-0.0177154541015625,
-0.00814056396484375,
-0.056396484375,
-0.01001739501953125,
0.03082275390625,
-0.05584716796875,
-0.003570556640625,
-0.061126708984375,
0.10687255859375,
-0.01085662841796875,
0.0184173583984375,
0.0099334716796875,
-0.03204345703125,
0.0002334117889404297,
0.007534027099609375,
-0.0079803466796875,
0.00743865966796875,
-0.0197906494140625,
0.034210205078125,
-0.0155792236328125,
0.063720703125,
0.004848480224609375,
-0.005657196044921875,
0.00040841102600097656,
0.013580322265625,
0.02679443359375,
0.007305145263671875,
0.0157012939453125,
-0.0537109375,
0.018463134765625,
-0.0380859375,
-0.043731689453125,
0.02294921875,
-0.042083740234375,
-0.0264434814453125,
0.018310546875,
-0.0489501953125,
-0.01311492919921875,
0.018768310546875,
0.0185089111328125,
0.0694580078125,
-0.0297698974609375,
0.021759033203125,
0.0811767578125,
-0.029998779296875,
0.060211181640625,
0.0340576171875,
-0.0265045166015625,
-0.007701873779296875,
0.0458984375,
-0.01050567626953125,
-0.004512786865234375,
0.03460693359375,
0.04034423828125,
-0.028717041015625,
-0.013153076171875,
-0.05328369140625,
0.018402099609375,
-0.029052734375,
-0.0166168212890625,
-0.05670166015625,
0.016876220703125,
-0.039794921875,
0.011016845703125,
-0.01160430908203125,
-0.029144287109375,
-0.035980224609375,
-0.021209716796875,
0.02716064453125,
0.040435791015625,
0.004848480224609375,
0.05706787109375,
-0.0799560546875,
0.01412200927734375,
0.032440185546875,
0.03289794921875,
-0.00635528564453125,
-0.03759765625,
-0.036468505859375,
0.00820159912109375,
-0.022064208984375,
-0.04876708984375,
0.004993438720703125,
0.006145477294921875,
0.03204345703125,
0.0322265625,
0.01404571533203125,
0.032257080078125,
-0.052703857421875,
0.07940673828125,
-0.002666473388671875,
-0.062164306640625,
0.06353759765625,
-0.0406494140625,
0.0679931640625,
0.05352783203125,
0.023834228515625,
-0.0462646484375,
-0.0248565673828125,
-0.049652099609375,
-0.060943603515625,
0.019012451171875,
-0.02215576171875,
0.060882568359375,
-0.02642822265625,
0.004695892333984375,
-0.0107574462890625,
0.0171051025390625,
-0.049591064453125,
-0.01517486572265625,
0.004421234130859375,
-0.01461029052734375,
-0.001209259033203125,
-0.037078857421875,
-0.02294921875,
-0.013580322265625,
0.030487060546875,
0.0106201171875,
0.0297698974609375,
-0.029327392578125,
0.0224151611328125,
-0.020355224609375,
0.01276397705078125,
0.10089111328125,
0.03857421875,
-0.016754150390625,
0.0112457275390625,
0.01511383056640625,
-0.02362060546875,
-0.0167999267578125,
-0.002658843994140625,
-0.0002529621124267578,
-0.025726318359375,
0.04248046875,
0.052825927734375,
-0.0233917236328125,
-0.04412841796875,
0.0166778564453125,
-0.021240234375,
-0.003143310546875,
-0.0209503173828125,
0.0258636474609375,
-0.01155853271484375,
0.0084228515625,
0.007190704345703125,
-0.0254669189453125,
0.01084136962890625,
-0.056060791015625,
0.0210723876953125,
0.018218994140625,
-0.048431396484375,
-0.01177978515625,
0.031402587890625,
0.03521728515625,
-0.050048828125,
0.06854248046875,
0.002685546875,
-0.050079345703125,
0.0653076171875,
0.05291748046875,
0.05572509765625,
-0.00860595703125,
0.005550384521484375,
0.0303192138671875,
0.01422119140625,
-0.02362060546875,
0.06768798828125,
0.0200042724609375,
-0.052032470703125,
-0.02947998046875,
-0.0155181884765625,
-0.01445770263671875,
0.0168304443359375,
-0.0447998046875,
-0.0048980712890625,
-0.056610107421875,
0.00446319580078125,
0.0186920166015625,
-0.0256805419921875,
-0.037841796875,
0.0240020751953125,
-0.0015430450439453125,
0.11187744140625,
-0.05316162109375,
0.053680419921875,
0.047027587890625,
-0.0574951171875,
-0.088623046875,
-0.00955963134765625,
-0.0157928466796875,
-0.04058837890625,
0.05389404296875,
0.0174407958984375,
0.0059814453125,
0.0126190185546875,
-0.10302734375,
-0.017242431640625,
0.033538818359375,
0.034698486328125,
-0.029388427734375,
0.01422882080078125,
-0.045013427734375,
0.050506591796875,
-0.034515380859375,
0.0249481201171875,
0.048797607421875,
0.031585693359375,
-0.0109405517578125,
-0.0701904296875,
0.010711669921875,
-0.00731658935546875,
0.01171112060546875,
0.021881103515625,
-0.038787841796875,
0.0662841796875,
-0.0115814208984375,
0.009613037109375,
0.03936767578125,
0.05145263671875,
-0.0171966552734375,
0.0027065277099609375,
0.05023193359375,
0.021392822265625,
0.043304443359375,
-0.00853729248046875,
0.075439453125,
-0.015289306640625,
0.03228759765625,
0.09112548828125,
-0.003246307373046875,
0.0703125,
0.01403045654296875,
-0.02276611328125,
0.0333251953125,
0.0640869140625,
-0.021392822265625,
0.0335693359375,
0.0207366943359375,
-0.0218048095703125,
-0.039031982421875,
-0.0152435302734375,
-0.0283966064453125,
0.021240234375,
-0.00569915771484375,
0.0004730224609375,
-0.0174102783203125,
-0.0011081695556640625,
-0.002368927001953125,
0.0170745849609375,
-0.046875,
0.06353759765625,
-0.0174102783203125,
-0.05975341796875,
0.0361328125,
0.0209808349609375,
0.0275115966796875,
-0.03887939453125,
-0.016998291015625,
-0.019683837890625,
0.01157379150390625,
0.0009436607360839844,
-0.058929443359375,
0.0206146240234375,
0.01306915283203125,
-0.03363037109375,
-0.006328582763671875,
0.0221099853515625,
-0.04962158203125,
-0.03057861328125,
-0.0038738250732421875,
-0.00353240966796875,
0.040679931640625,
0.0167999267578125,
-0.052001953125,
-0.005046844482421875,
0.005054473876953125,
0.005474090576171875,
-0.00662994384765625,
0.036651611328125,
0.0011186599731445312,
0.046173095703125,
0.033905029296875,
-0.01239776611328125,
-0.01204681396484375,
0.01202392578125,
0.0528564453125,
-0.039703369140625,
-0.0224151611328125,
-0.0406494140625,
0.044769287109375,
-0.01580810546875,
-0.041412353515625,
0.060821533203125,
0.04034423828125,
0.0682373046875,
-0.035186767578125,
0.0199432373046875,
-0.00794219970703125,
0.04827880859375,
-0.0292816162109375,
0.0135955810546875,
-0.019378662109375,
0.0016632080078125,
0.0146942138671875,
-0.048553466796875,
-0.020263671875,
0.071533203125,
0.0047607421875,
0.0176849365234375,
0.06158447265625,
0.0716552734375,
0.0014591217041015625,
-0.006256103515625,
0.0301055908203125,
0.03582763671875,
0.00018262863159179688,
0.0087127685546875,
0.04803466796875,
-0.0309906005859375,
0.02703857421875,
0.0079803466796875,
-0.0160369873046875,
-0.005512237548828125,
-0.0703125,
-0.058685302734375,
-0.025421142578125,
-0.0247955322265625,
-0.05035400390625,
-0.0029697418212890625,
0.10125732421875,
0.054962158203125,
-0.08935546875,
-0.0416259765625,
0.0115203857421875,
0.007232666015625,
-0.006343841552734375,
-0.007053375244140625,
-0.006072998046875,
-0.03411865234375,
-0.038787841796875,
0.00980377197265625,
-0.0288848876953125,
0.048431396484375,
-0.02734375,
0.01434326171875,
-0.00199127197265625,
0.01085662841796875,
0.0535888671875,
0.029754638671875,
-0.040313720703125,
-0.04412841796875,
0.0081787109375,
-0.025238037109375,
-0.02978515625,
0.048858642578125,
-0.01050567626953125,
0.024627685546875,
0.0276641845703125,
0.0509033203125,
0.02349853515625,
0.01267242431640625,
0.07476806640625,
-0.04986572265625,
0.0024700164794921875,
0.004749298095703125,
0.017333984375,
0.022979736328125,
-0.047149658203125,
0.035675048828125,
-0.0055389404296875,
-0.054840087890625,
-0.059906005859375,
0.033905029296875,
-0.1025390625,
-0.026275634765625,
0.071533203125,
0.01419830322265625,
-0.003833770751953125,
-0.03033447265625,
-0.0670166015625,
0.025177001953125,
-0.037322998046875,
0.02691650390625,
0.060882568359375,
0.00623321533203125,
-0.032806396484375,
-0.033843994140625,
0.027557373046875,
0.01508331298828125,
-0.063720703125,
0.024169921875,
0.06280517578125,
0.016326904296875,
0.006343841552734375,
0.03369140625,
0.01947021484375,
0.019683837890625,
0.0123291015625,
-0.0121002197265625,
-0.01111602783203125,
-0.0240936279296875,
-0.04632568359375,
0.002025604248046875,
0.03814697265625,
-0.041168212890625
]
] |
The-Face-Of-Goonery/Huginn-13b-FP16 | 2023-08-17T18:39:58.000Z | [
"transformers",
"pytorch",
"llama",
"text-generation",
"endpoints_compatible",
"has_space",
"text-generation-inference",
"region:us"
] | text-generation | The-Face-Of-Goonery | null | null | The-Face-Of-Goonery/Huginn-13b-FP16 | 11 | 5,944 | transformers | 2023-08-05T17:32:43 | ---
{}
---
a merge of a lot of different models, like hermes, beluga, airoboros, chronos.. limarp
significantly better quality than my previous chronos-beluga merge.
Huginn is intended as a general purpose model, that maintains a lot of good knowledge, can perform logical thought and accurately follow instructions, and hold the prose and creativity of more writing oriented models, this makes this model great for roleplays, while still being good as a normal chatbot or assistant | 487 | [
[
-0.037567138671875,
-0.04901123046875,
0.0213470458984375,
0.0172882080078125,
-0.0404052734375,
-0.0070953369140625,
-0.003162384033203125,
-0.06500244140625,
0.03955078125,
0.04473876953125,
-0.0117645263671875,
0.003276824951171875,
-0.050689697265625,
0.00725555419921875,
-0.043243408203125,
0.08966064453125,
0.0024623870849609375,
-0.0009322166442871094,
0.00740814208984375,
-0.0104522705078125,
-0.00408172607421875,
-0.044708251953125,
-0.065673828125,
-0.04327392578125,
0.061981201171875,
0.013275146484375,
0.051300048828125,
0.019012451171875,
0.035858154296875,
0.031402587890625,
-0.0201873779296875,
0.027587890625,
-0.0308685302734375,
0.02008056640625,
0.011749267578125,
-0.042572021484375,
-0.050567626953125,
0.0038356781005859375,
0.02606201171875,
0.040740966796875,
-0.0233612060546875,
0.00449371337890625,
0.02520751953125,
0.0362548828125,
-0.0024204254150390625,
0.0015535354614257812,
0.01235198974609375,
0.00760650634765625,
0.0020904541015625,
-0.0017747879028320312,
-0.006103515625,
-0.060821533203125,
0.02960205078125,
-0.0750732421875,
0.026947021484375,
0.0240478515625,
0.09063720703125,
0.00017559528350830078,
-0.0255889892578125,
-0.025848388671875,
-0.052581787109375,
0.06146240234375,
-0.06640625,
0.02313232421875,
0.0164947509765625,
0.0452880859375,
-0.01837158203125,
-0.020355224609375,
-0.045501708984375,
-0.00208282470703125,
-0.030670166015625,
0.0094451904296875,
-0.014556884765625,
0.009918212890625,
0.01238250732421875,
0.038482666015625,
-0.03167724609375,
0.00614166259765625,
-0.061187744140625,
-0.01213836669921875,
0.039398193359375,
0.01064300537109375,
0.032501220703125,
-0.013641357421875,
-0.037261962890625,
-0.0129852294921875,
-0.017669677734375,
0.021026611328125,
0.0574951171875,
0.01343536376953125,
-0.01117706298828125,
0.05810546875,
-0.0006728172302246094,
0.065673828125,
0.0160369873046875,
-0.01067352294921875,
-0.0027942657470703125,
-0.0109100341796875,
-0.05316162109375,
0.0018968582153320312,
0.04962158203125,
0.026763916015625,
0.0277557373046875,
0.00740814208984375,
-0.0118560791015625,
-0.0028839111328125,
0.034515380859375,
-0.058197021484375,
-0.020751953125,
0.029266357421875,
-0.058319091796875,
-0.0479736328125,
0.0228729248046875,
-0.044952392578125,
-0.0031757354736328125,
-0.0487060546875,
0.01220703125,
-0.0404052734375,
-0.0272674560546875,
-0.00362396240234375,
-0.03729248046875,
0.01204681396484375,
0.05877685546875,
-0.0709228515625,
0.03680419921875,
0.061279296875,
0.0665283203125,
-0.007335662841796875,
-0.026947021484375,
-0.02972412109375,
-0.0177154541015625,
-0.048614501953125,
0.0550537109375,
-0.022979736328125,
-0.05694580078125,
-0.006053924560546875,
0.01529693603515625,
-0.009002685546875,
-0.032440185546875,
0.043701171875,
-0.042449951171875,
0.052459716796875,
-0.0333251953125,
-0.050994873046875,
-0.043060302734375,
0.0257110595703125,
-0.0506591796875,
0.0677490234375,
0.029876708984375,
-0.056488037109375,
0.051239013671875,
-0.039642333984375,
-0.0361328125,
0.01708984375,
0.005222320556640625,
-0.0240936279296875,
0.0220184326171875,
-0.0023250579833984375,
0.043609619140625,
-0.030059814453125,
0.0074310302734375,
-0.0223236083984375,
-0.034576416015625,
0.0239410400390625,
0.0228271484375,
0.056488037109375,
0.041412353515625,
0.004283905029296875,
0.0242156982421875,
-0.028167724609375,
-0.0243072509765625,
-0.010833740234375,
-0.028350830078125,
-0.051666259765625,
-0.0170135498046875,
0.020904541015625,
0.005290985107421875,
0.0106658935546875,
-0.039642333984375,
0.034027099609375,
-0.0210723876953125,
0.00689697265625,
0.03814697265625,
0.00342559814453125,
0.040374755859375,
-0.0673828125,
0.04901123046875,
0.00030493736267089844,
0.039398193359375,
-0.006801605224609375,
-0.0626220703125,
-0.0716552734375,
-0.057098388671875,
0.00809478759765625,
0.0294647216796875,
-0.0232696533203125,
0.054656982421875,
0.0273590087890625,
-0.04833984375,
-0.046478271484375,
0.01360321044921875,
0.019866943359375,
0.04925537109375,
-0.003932952880859375,
-0.0290679931640625,
-0.048248291015625,
-0.04864501953125,
0.0142364501953125,
-0.015777587890625,
-0.013153076171875,
0.0172271728515625,
0.03155517578125,
-0.0384521484375,
0.05023193359375,
-0.01538848876953125,
-0.0163421630859375,
-0.038909912109375,
0.01438140869140625,
0.04998779296875,
0.03466796875,
0.065673828125,
-0.0570068359375,
-0.0204010009765625,
0.015869140625,
-0.059814453125,
0.00934600830078125,
0.006195068359375,
-0.01611328125,
0.004009246826171875,
-0.0061798095703125,
-0.0430908203125,
0.0212249755859375,
0.0684814453125,
-0.0242767333984375,
0.047271728515625,
-0.045501708984375,
0.0634765625,
-0.10723876953125,
0.0230712890625,
0.00017976760864257812,
-0.0094451904296875,
-0.06005859375,
0.003993988037109375,
-0.021484375,
-0.025604248046875,
-0.0186920166015625,
0.058563232421875,
-0.0256805419921875,
-0.007579803466796875,
-0.030975341796875,
0.0022258758544921875,
0.0136871337890625,
0.003673553466796875,
0.01311492919921875,
0.0284576416015625,
0.042083740234375,
-0.06884765625,
0.052581787109375,
0.031982421875,
-0.03173828125,
0.045074462890625,
-0.0919189453125,
-0.01029205322265625,
-0.008697509765625,
0.01357269287109375,
-0.06494140625,
-0.041900634765625,
0.0035247802734375,
-0.020172119140625,
0.019866943359375,
-0.00937652587890625,
-0.0289306640625,
-0.033966064453125,
-0.020294189453125,
0.024322509765625,
0.056854248046875,
-0.044342041015625,
0.04925537109375,
0.0066375732421875,
-0.018768310546875,
-0.0224456787109375,
-0.03546142578125,
-0.0235443115234375,
-0.02593994140625,
-0.03857421875,
-0.0020084381103515625,
-0.007144927978515625,
-0.01812744140625,
0.0014162063598632812,
0.032989501953125,
-0.0107879638671875,
-0.022003173828125,
0.0318603515625,
0.06689453125,
-0.0243377685546875,
-0.0235137939453125,
0.007427215576171875,
0.0007424354553222656,
-0.035858154296875,
0.0048675537109375,
0.04290771484375,
-0.006427764892578125,
-0.0139312744140625,
-0.042083740234375,
0.01898193359375,
0.06280517578125,
0.01329803466796875,
0.0640869140625,
0.0243377685546875,
-0.0104217529296875,
0.00028204917907714844,
-0.04559326171875,
-0.011260986328125,
-0.0287628173828125,
-0.032867431640625,
-0.0281219482421875,
-0.08453369140625,
0.05926513671875,
0.007007598876953125,
0.0192413330078125,
0.0219879150390625,
0.02264404296875,
-0.0254974365234375,
0.039459228515625,
0.06939697265625,
-0.0214996337890625,
0.027587890625,
-0.00627899169921875,
0.01116943359375,
-0.0858154296875,
-0.0355224609375,
-0.037109375,
-0.00885772705078125,
-0.039093017578125,
-0.035888671875,
-0.00981903076171875,
0.0266876220703125,
-0.010711669921875,
0.0745849609375,
-0.052825927734375,
-0.0093841552734375,
0.051177978515625,
-0.01088714599609375,
0.0162353515625,
-0.0222015380859375,
0.0014600753784179688,
0.0030975341796875,
-0.046356201171875,
-0.049530029296875,
0.05535888671875,
0.0243988037109375,
0.07391357421875,
-0.001255035400390625,
0.056884765625,
0.0245208740234375,
0.022003173828125,
-0.03082275390625,
0.0257110595703125,
0.00701904296875,
-0.051055908203125,
-0.0188446044921875,
-0.034759521484375,
-0.0755615234375,
0.0281982421875,
-0.026031494140625,
-0.04901123046875,
0.0340576171875,
0.03375244140625,
-0.02642822265625,
0.01276397705078125,
-0.037261962890625,
0.037506103515625,
-0.0147247314453125,
0.006702423095703125,
-0.0196533203125,
-0.0284576416015625,
0.04888916015625,
0.000957489013671875,
0.0039043426513671875,
-0.0003986358642578125,
0.0093994140625,
0.048828125,
-0.03485107421875,
0.033172607421875,
0.020416259765625,
-0.001972198486328125,
0.03656005859375,
0.0202789306640625,
0.005268096923828125,
-0.0206756591796875,
-0.005619049072265625,
0.015167236328125,
0.0161895751953125,
-0.0245208740234375,
-0.043792724609375,
0.047607421875,
-0.0751953125,
-0.019134521484375,
-0.037200927734375,
-0.01285552978515625,
0.0226593017578125,
0.002025604248046875,
0.032073974609375,
0.040985107421875,
-0.048309326171875,
-0.0031108856201171875,
0.04791259765625,
-0.03399658203125,
-0.006011962890625,
0.02099609375,
-0.053009033203125,
-0.044342041015625,
0.027130126953125,
-0.02850341796875,
0.01360321044921875,
0.031982421875,
0.024505615234375,
-0.00370025634765625,
0.03167724609375,
-0.0200653076171875,
0.03839111328125,
-0.02642822265625,
0.0092926025390625,
-0.043212890625,
-0.02740478515625,
-0.04559326171875,
-0.05352783203125,
-0.01100921630859375,
-0.040557861328125,
-0.00969696044921875,
-0.0011091232299804688,
0.0269775390625,
0.058746337890625,
-0.013885498046875,
0.0268096923828125,
-0.059539794921875,
0.034515380859375,
0.053619384765625,
0.00318145751953125,
-0.01392364501953125,
-0.05010986328125,
0.0092926025390625,
-0.0165863037109375,
0.0164642333984375,
-0.06103515625,
0.027587890625,
-0.01073455810546875,
0.044952392578125,
0.07781982421875,
-0.0256805419921875,
0.03778076171875,
-0.01100921630859375,
0.058441162109375,
0.043609619140625,
-0.0469970703125,
0.051422119140625,
-0.04058837890625,
0.00823211669921875,
0.0168914794921875,
0.03704833984375,
-0.049957275390625,
-0.031890869140625,
-0.06640625,
-0.0228271484375,
0.044647216796875,
0.0303497314453125,
-0.01617431640625,
0.01690673828125,
0.01953125,
0.0180816650390625,
0.051544189453125,
-0.018951416015625,
0.0037593841552734375,
-0.0360107421875,
0.01410675048828125,
-0.0379638671875,
-0.0255126953125,
-0.0297393798828125,
-0.0299835205078125,
0.037384033203125,
0.0178070068359375,
0.0340576171875,
0.00963592529296875,
0.0258636474609375,
-0.023406982421875,
0.01499176025390625,
0.050445556640625,
0.0227508544921875,
-0.021270751953125,
0.004177093505859375,
0.0030193328857421875,
-0.0172882080078125,
-0.00617218017578125,
0.0028171539306640625,
-0.0006923675537109375,
0.01428985595703125,
0.055633544921875,
0.048004150390625,
0.02880859375,
-0.0340576171875,
0.0123291015625,
-0.0028972625732421875,
-0.02093505859375,
0.00841522216796875,
0.042083740234375,
-0.006763458251953125,
0.02130126953125,
-0.003265380859375,
0.0140380859375,
0.0117645263671875,
-0.0780029296875,
0.0085906982421875,
0.00890350341796875,
-0.017425537109375,
-0.0293731689453125,
0.0689697265625,
0.03472900390625,
-0.0256805419921875,
0.058746337890625,
0.01360321044921875,
-0.0618896484375,
0.042449951171875,
0.0255889892578125,
0.05584716796875,
-0.0400390625,
0.03314208984375,
0.04180908203125,
0.0247039794921875,
-0.034912109375,
0.01416778564453125,
-0.031341552734375,
-0.06207275390625,
-0.00798797607421875,
-0.054656982421875,
-0.05804443359375,
-0.021514892578125,
-0.04656982421875,
0.0242767333984375,
-0.031982421875,
-0.0233306884765625,
-0.0204925537109375,
0.00412750244140625,
-0.047393798828125,
0.041595458984375,
-0.01515960693359375,
0.0655517578125,
-0.06585693359375,
0.061065673828125,
0.061187744140625,
-0.0287933349609375,
-0.0716552734375,
-0.037139892578125,
0.0041046142578125,
-0.049957275390625,
0.0579833984375,
-0.0026493072509765625,
-0.001087188720703125,
-0.0517578125,
-0.0180816650390625,
-0.040374755859375,
0.0755615234375,
0.0028820037841796875,
-0.05535888671875,
-0.009735107421875,
-0.0198211669921875,
0.061737060546875,
-0.039947509765625,
0.0123138427734375,
0.01387786865234375,
0.029632568359375,
0.026824951171875,
-0.079345703125,
-0.012115478515625,
-0.036834716796875,
0.01898193359375,
0.00984954833984375,
-0.06549072265625,
0.07196044921875,
-0.01094818115234375,
0.01204681396484375,
0.01348876953125,
0.038604736328125,
0.041534423828125,
0.0078277587890625,
0.044586181640625,
0.033477783203125,
0.056915283203125,
-0.00472259521484375,
0.0872802734375,
-0.0208740234375,
0.0130767822265625,
0.05120849609375,
-0.01763916015625,
0.0411376953125,
0.01477813720703125,
0.01947021484375,
0.04400634765625,
0.04718017578125,
0.0029697418212890625,
0.032867431640625,
-0.014190673828125,
0.00218963623046875,
-0.01323699951171875,
0.0199432373046875,
-0.0224456787109375,
0.0181732177734375,
0.00015425682067871094,
-0.0241241455078125,
0.0237579345703125,
-0.007160186767578125,
0.00997161865234375,
-0.006023406982421875,
0.00711822509765625,
0.046722412109375,
0.0175323486328125,
-0.0732421875,
-0.004016876220703125,
-0.025848388671875,
0.0428466796875,
-0.0628662109375,
-0.007495880126953125,
-0.0141754150390625,
0.0207061767578125,
-0.00986480712890625,
-0.055999755859375,
0.0178680419921875,
-0.0130767822265625,
0.01422882080078125,
-0.0145111083984375,
0.06982421875,
-0.024566650390625,
-0.035919189453125,
0.0263214111328125,
0.00789642333984375,
0.0194854736328125,
0.023681640625,
-0.045318603515625,
0.0182952880859375,
-0.022247314453125,
-0.0186920166015625,
0.003772735595703125,
0.0272064208984375,
0.026763916015625,
0.051666259765625,
0.05316162109375,
0.03179931640625,
-0.0100250244140625,
0.00611114501953125,
0.0506591796875,
-0.03131103515625,
-0.052398681640625,
-0.017669677734375,
0.0308990478515625,
-0.037750244140625,
-0.05633544921875,
0.07928466796875,
0.054107666015625,
0.0190582275390625,
-0.0063323974609375,
0.047637939453125,
-0.00392913818359375,
0.0457763671875,
-0.02001953125,
0.048553466796875,
-0.040924072265625,
-0.0169830322265625,
-0.0433349609375,
-0.0926513671875,
0.0022296905517578125,
0.0701904296875,
0.006755828857421875,
-0.00904083251953125,
0.0312347412109375,
0.0224151611328125,
-0.01995849609375,
0.01181793212890625,
0.054443359375,
-0.01187896728515625,
0.0228271484375,
0.01702880859375,
0.06427001953125,
-0.024871826171875,
0.001575469970703125,
-0.0091094970703125,
-0.04058837890625,
-0.048309326171875,
-0.02947998046875,
-0.10247802734375,
-0.058837890625,
-0.00901031494140625,
-0.0198516845703125,
0.01015472412109375,
0.038360595703125,
0.07659912109375,
-0.04083251953125,
-0.031158447265625,
0.007656097412109375,
-0.0157928466796875,
-0.024688720703125,
-0.0171966552734375,
0.006412506103515625,
-0.0156707763671875,
-0.048492431640625,
0.002071380615234375,
0.031158447265625,
0.024444580078125,
-0.0103302001953125,
-0.02142333984375,
0.0186767578125,
0.046295166015625,
0.044891357421875,
0.039886474609375,
-0.040374755859375,
-0.0239410400390625,
0.0162506103515625,
-0.0198516845703125,
-0.0206298828125,
0.072509765625,
-0.023895263671875,
0.04229736328125,
0.04132080078125,
0.0157012939453125,
0.049713134765625,
0.00908660888671875,
0.07073974609375,
-0.01081085205078125,
0.0309906005859375,
0.0068511962890625,
0.050140380859375,
0.023406982421875,
-0.026123046875,
0.042266845703125,
-0.00567626953125,
-0.07196044921875,
-0.023956298828125,
0.01251220703125,
-0.106689453125,
-0.0122222900390625,
0.0716552734375,
-0.0127410888671875,
-0.016693115234375,
0.0121612548828125,
-0.046295166015625,
0.0284576416015625,
-0.025482177734375,
0.070556640625,
0.045654296875,
-0.035675048828125,
-0.017364501953125,
-0.0109710693359375,
0.031646728515625,
-0.00058746337890625,
-0.0775146484375,
-0.0298614501953125,
0.020538330078125,
0.019378662109375,
0.02301025390625,
0.04412841796875,
0.0021038055419921875,
0.0276336669921875,
0.001255035400390625,
-0.002498626708984375,
0.0152740478515625,
-0.025054931640625,
0.01136016845703125,
0.004764556884765625,
0.00691986083984375,
-0.01132965087890625
]
] |
CHIH-HUNG/llama-2-13b-OpenOrca_20w | 2023-09-06T04:54:52.000Z | [
"transformers",
"pytorch",
"llama",
"text-generation",
"dataset:Open-Orca/OpenOrca",
"license:llama2",
"endpoints_compatible",
"text-generation-inference",
"region:us"
] | text-generation | CHIH-HUNG | null | null | CHIH-HUNG/llama-2-13b-OpenOrca_20w | 0 | 5,944 | transformers | 2023-08-30T07:06:08 | ---
license: llama2
datasets:
- Open-Orca/OpenOrca
---
# Model Card for Model ID
<!-- Provide a quick summary of what the model is/does. -->
在llama-2-13b上使用open orca前20萬筆資料集進行訓練
# Fine-Tuning Information
- **GPU:** RTX4090 (single core / 24564MiB)
- **model:** meta-llama/Llama-2-13b-hf
- **dataset:** Open-Orca/OpenOrca (取前20w筆訓練集)
- **peft_type:** LoRA
- **lora_rank:** 8
- **lora_target:** q_proj, v_proj
- **per_device_train_batch_size:** 8
- **gradient_accumulation_steps:** 8
- **learning_rate :** 5e-5
- **epoch:** 1
- **precision:** bf16
- **quantization:** load_in_4bit
# Fine-Tuning Detail
- **train_loss:** 0.8616
- **train_runtime:** 29:18:07 (use deepspeed)
# Evaluation
- 評估結果來自**HuggingFaceH4/open_llm_leaderboard**
- 與Llama-2-13b和其他使用Open-Orca的模型比較4種Benchmark
- Benchmark包含**ARC**、**HellaSwag**、**MMLU**、**TruthfulQA**
| Model |Average| ARC |HellaSwag| MMLU | TruthfulQA |
|-----------------------------------------|-------|-------|---------|-------|------------|
|meta-llama/Llama-2-13b-hf | 56.9 | 58.11 | 80.97 | 54.34 | 34.17 |
|meta-llama/Llama-2-13b-chat-hf | 59.93 | 59.04 | 81.94 | 54.64 | 44.12 |
|Open-Orca/OpenOrca-Platypus2-13B | 64.6 | 62.8 | 83.15 | 59.39 | 53.08 |
|Open-Orca/OpenOrcaxOpenChat-Preview2-13B | 63.81 | 62.37 | 82.96 | 58.68 | 51.23 |
|circulus/Llama-2-13b-orca-v1 | 62.91 | 62.03 | 82.27 | 57.71 | 49.61 |
|CHIH-HUNG/llama-2-13b-OpenOrca_5w | 61.2 | 61.01 | 82.82 | 56.09 | 44.87 |
|CHIH-HUNG/llama-2-13b-open_orca_20w | 60.46 | 59.9 | 82.51 | 56.3 | 43.14 |
# How to convert dataset to json
- 在**load_dataset**中輸入資料集名稱，並且在**take**中輸入要取前幾筆資料
- 觀察該資料集的欄位名稱，填入**example**欄位中(例如system_prompt、question、response)
- 最後指定json檔儲存位置 (**json_filename**)
```py
import json
from datasets import load_dataset
# 讀取數據集，take可以取得該數據集前n筆資料
dataset = load_dataset("Open-Orca/OpenOrca", split="train", streaming=True).take(200000)
# 提取所需欄位並建立新的字典列表
extracted_data = []
for example in dataset:
extracted_example = {
### open orca
"system_prompt": example["system_prompt"],
"question": example["question"],
"response": example["response"]
}
extracted_data.append(extracted_example)
# 指定 JSON 文件名稱
json_filename = "open_orca.json"
# 寫入 JSON 文件
with open(json_filename, "w") as json_file:
json.dump(extracted_data, json_file, indent=4)
print(f"數據已提取並保存為 {json_filename}")
```
| 2,509 | [
[
-0.04071044921875,
-0.047821044921875,
0.006862640380859375,
0.0037708282470703125,
-0.043487548828125,
-0.0006933212280273438,
-0.00577545166015625,
-0.031402587890625,
0.0191802978515625,
0.03790283203125,
-0.043670654296875,
-0.0516357421875,
-0.03662109375,
0.01068115234375,
-0.017822265625,
0.0809326171875,
-0.01229095458984375,
-0.01099395751953125,
0.020263671875,
-0.004436492919921875,
-0.036773681640625,
-0.029632568359375,
-0.068115234375,
-0.02130126953125,
0.0251007080078125,
0.01160430908203125,
0.047760009765625,
0.07696533203125,
0.048370361328125,
0.0180816650390625,
-0.01110076904296875,
0.0252532958984375,
-0.038482666015625,
-0.0147552490234375,
0.007595062255859375,
-0.049224853515625,
-0.047698974609375,
-0.011566162109375,
0.040802001953125,
0.029144287109375,
0.0005044937133789062,
0.044586181640625,
0.01062774658203125,
0.040557861328125,
-0.0340576171875,
0.0307464599609375,
-0.0206451416015625,
0.006748199462890625,
-0.032562255859375,
-0.0231781005859375,
0.0017480850219726562,
-0.032470703125,
-0.00724029541015625,
-0.070556640625,
0.00437164306640625,
0.01134490966796875,
0.095947265625,
0.028778076171875,
-0.0228424072265625,
-0.00803375244140625,
-0.028045654296875,
0.0615234375,
-0.0718994140625,
-0.0060272216796875,
0.014404296875,
0.023712158203125,
-0.0251312255859375,
-0.045379638671875,
-0.057952880859375,
0.0098419189453125,
-0.0114288330078125,
0.01102447509765625,
0.0007696151733398438,
-0.0262908935546875,
0.0220489501953125,
0.031890869140625,
-0.036346435546875,
0.0014677047729492188,
-0.035491943359375,
0.011962890625,
0.06292724609375,
0.04217529296875,
0.004802703857421875,
-0.0174560546875,
-0.0167236328125,
-0.03594970703125,
-0.040985107421875,
0.0335693359375,
0.0355224609375,
0.037384033203125,
-0.0360107421875,
0.042510986328125,
-0.0323486328125,
0.035064697265625,
-0.0032787322998046875,
-0.04156494140625,
0.04559326171875,
-0.019012451171875,
-0.039825439453125,
-0.0018129348754882812,
0.0758056640625,
0.039764404296875,
-0.00846099853515625,
0.0263671875,
-0.0035247802734375,
-0.00850677490234375,
-0.0006389617919921875,
-0.05621337890625,
-0.0190582275390625,
0.04144287109375,
-0.0535888671875,
-0.035430908203125,
0.006595611572265625,
-0.06744384765625,
-0.016357421875,
0.0090789794921875,
0.008270263671875,
-0.02593994140625,
-0.03936767578125,
0.01800537109375,
-0.01548004150390625,
0.027984619140625,
0.027587890625,
-0.0604248046875,
0.0172271728515625,
0.03558349609375,
0.04937744140625,
0.0083160400390625,
-0.023223876953125,
-0.0017032623291015625,
0.008270263671875,
-0.0295562744140625,
0.057098388671875,
-0.004131317138671875,
-0.035980224609375,
-0.01421356201171875,
0.016845703125,
-0.005970001220703125,
-0.03106689453125,
0.05926513671875,
-0.0307464599609375,
-0.004383087158203125,
-0.048187255859375,
-0.004070281982421875,
-0.04559326171875,
0.0335693359375,
-0.051055908203125,
0.08270263671875,
0.0017309188842773438,
-0.0692138671875,
0.035919189453125,
-0.06494140625,
-0.0171051025390625,
-0.00519561767578125,
-0.0033416748046875,
-0.0352783203125,
-0.025054931640625,
0.033203125,
0.035064697265625,
-0.034271240234375,
0.004150390625,
-0.0243072509765625,
-0.02862548828125,
0.0193634033203125,
-0.01332855224609375,
0.07684326171875,
0.030242919921875,
-0.01448822021484375,
-0.0037899017333984375,
-0.06689453125,
-0.005069732666015625,
0.05218505859375,
-0.039031982421875,
-0.005279541015625,
-0.0017366409301757812,
-0.0022068023681640625,
-0.007965087890625,
0.0322265625,
-0.027130126953125,
0.034942626953125,
-0.0187835693359375,
0.038726806640625,
0.053924560546875,
-0.00360870361328125,
0.00849151611328125,
-0.03240966796875,
0.0261383056640625,
0.003993988037109375,
0.023223876953125,
-0.00421905517578125,
-0.039031982421875,
-0.0750732421875,
-0.0204315185546875,
0.005615234375,
0.031036376953125,
-0.03271484375,
0.053375244140625,
-0.0265960693359375,
-0.0548095703125,
-0.05224609375,
-0.003513336181640625,
0.0157318115234375,
0.04705810546875,
0.035491943359375,
-0.0093841552734375,
-0.053131103515625,
-0.059326171875,
0.00980377197265625,
-0.00966644287109375,
0.01435089111328125,
0.035430908203125,
0.057769775390625,
-0.009033203125,
0.049072265625,
-0.03790283203125,
-0.0258026123046875,
-0.0238494873046875,
0.002666473388671875,
0.06689453125,
0.040313720703125,
0.051971435546875,
-0.03765869140625,
-0.027130126953125,
0.0033283233642578125,
-0.0875244140625,
0.019378662109375,
-0.0002765655517578125,
-0.01044464111328125,
0.0021114349365234375,
0.00690460205078125,
-0.044677734375,
0.0455322265625,
0.0296478271484375,
-0.016204833984375,
0.04046630859375,
0.00043511390686035156,
0.0223388671875,
-0.07073974609375,
0.0267486572265625,
-0.02581787109375,
0.003635406494140625,
-0.01849365234375,
0.014801025390625,
-0.011871337890625,
0.0137481689453125,
-0.0303192138671875,
0.025390625,
-0.025787353515625,
0.0023822784423828125,
-0.01100921630859375,
0.00016415119171142578,
0.00446319580078125,
0.03778076171875,
0.0008096694946289062,
0.04486083984375,
0.049896240234375,
-0.051116943359375,
0.042877197265625,
0.038543701171875,
-0.03009033203125,
0.0144805908203125,
-0.0374755859375,
0.00630950927734375,
0.0022335052490234375,
0.03948974609375,
-0.077880859375,
-0.026153564453125,
0.042388916015625,
-0.0295562744140625,
0.01239776611328125,
-0.027740478515625,
-0.032318115234375,
-0.049102783203125,
-0.0374755859375,
0.025421142578125,
0.01393890380859375,
-0.0535888671875,
0.02777099609375,
0.01239013671875,
0.01279449462890625,
-0.056854248046875,
-0.05999755859375,
-0.004058837890625,
-0.0184326171875,
-0.042266845703125,
0.0165557861328125,
-0.0030460357666015625,
0.004146575927734375,
0.0043792724609375,
0.0005636215209960938,
0.00975799560546875,
0.0037097930908203125,
0.01885986328125,
0.038665771484375,
-0.0229949951171875,
-0.0274810791015625,
0.003772735595703125,
-0.00638580322265625,
0.003936767578125,
0.006229400634765625,
0.065185546875,
-0.0186614990234375,
-0.015655517578125,
-0.043853759765625,
0.00267791748046875,
0.0308837890625,
-0.003139495849609375,
0.055633544921875,
0.047454833984375,
-0.01045989990234375,
0.00290679931640625,
-0.018951416015625,
0.005481719970703125,
-0.0341796875,
0.0215606689453125,
-0.049224853515625,
-0.04736328125,
0.06011962890625,
0.005458831787109375,
0.026123046875,
0.0645751953125,
0.0240020751953125,
-0.01055908203125,
0.0770263671875,
0.0248565673828125,
-0.021270751953125,
0.01451873779296875,
-0.06927490234375,
0.00555419921875,
-0.07086181640625,
-0.04150390625,
-0.0445556640625,
-0.039459228515625,
-0.03961181640625,
-0.0031833648681640625,
0.022552490234375,
0.017730712890625,
-0.04290771484375,
0.036285400390625,
-0.061248779296875,
0.0178680419921875,
0.03851318359375,
0.01457977294921875,
0.0149993896484375,
-0.0122222900390625,
0.008148193359375,
0.0166473388671875,
-0.042236328125,
-0.037689208984375,
0.10455322265625,
0.0222930908203125,
0.05743408203125,
0.0016450881958007812,
0.04510498046875,
0.00746917724609375,
0.0077972412109375,
-0.035430908203125,
0.039947509765625,
0.0031681060791015625,
-0.0552978515625,
-0.0174713134765625,
-0.0189971923828125,
-0.0645751953125,
0.0202178955078125,
-0.01068115234375,
-0.066162109375,
0.0173797607421875,
-0.0019121170043945312,
-0.044708251953125,
0.04296875,
-0.0301666259765625,
0.0548095703125,
-0.023284912109375,
-0.0027923583984375,
0.003391265869140625,
-0.04766845703125,
0.049468994140625,
0.006317138671875,
0.008514404296875,
-0.01477813720703125,
-0.0121612548828125,
0.077880859375,
-0.051361083984375,
0.046234130859375,
-0.01554107666015625,
-0.00008106231689453125,
0.0377197265625,
-0.001331329345703125,
0.052825927734375,
0.0256805419921875,
-0.00543975830078125,
0.042510986328125,
0.00824737548828125,
-0.020965576171875,
-0.0227203369140625,
0.050018310546875,
-0.08837890625,
-0.03204345703125,
-0.04498291015625,
-0.022613525390625,
0.0208282470703125,
0.022735595703125,
0.035858154296875,
-0.004955291748046875,
0.01263427734375,
0.00876617431640625,
0.0295867919921875,
-0.017059326171875,
0.041534423828125,
0.034271240234375,
-0.0193939208984375,
-0.05145263671875,
0.054931640625,
0.0018606185913085938,
-0.008819580078125,
0.0282745361328125,
0.00009763240814208984,
-0.0290374755859375,
-0.046783447265625,
-0.04364013671875,
0.0235137939453125,
-0.040771484375,
-0.044281005859375,
-0.05010986328125,
-0.0323486328125,
-0.0443115234375,
-0.001659393310546875,
-0.04254150390625,
-0.01424407958984375,
-0.0606689453125,
-0.0083465576171875,
0.049591064453125,
0.03643798828125,
-0.006328582763671875,
0.041961669921875,
-0.045654296875,
0.0198822021484375,
0.0091705322265625,
0.0017843246459960938,
0.00759124755859375,
-0.059539794921875,
-0.0159912109375,
0.00152587890625,
-0.0325927734375,
-0.041351318359375,
0.053009033203125,
-0.0036067962646484375,
0.030975341796875,
0.051788330078125,
-0.00034308433532714844,
0.08807373046875,
-0.0019989013671875,
0.07421875,
0.0146942138671875,
-0.04998779296875,
0.04107666015625,
-0.0264892578125,
-0.0148162841796875,
0.034912109375,
0.0243682861328125,
-0.0209503173828125,
-0.00237274169921875,
-0.0439453125,
-0.06341552734375,
0.08355712890625,
0.009613037109375,
-0.0100250244140625,
0.01398468017578125,
0.021331787109375,
0.0180511474609375,
0.02294921875,
-0.06121826171875,
-0.04638671875,
-0.03607177734375,
0.0127410888671875,
0.001556396484375,
-0.0038127899169921875,
-0.01316070556640625,
-0.0265045166015625,
0.05657958984375,
0.0004677772521972656,
0.033538818359375,
0.01477813720703125,
0.0202178955078125,
-0.01500701904296875,
-0.0004093647003173828,
0.0311279296875,
0.02734375,
-0.048614501953125,
-0.0084075927734375,
0.0162506103515625,
-0.0406494140625,
-0.007198333740234375,
0.006069183349609375,
-0.00940704345703125,
-0.015380859375,
0.031036376953125,
0.05108642578125,
-0.020843505859375,
-0.036865234375,
0.019622802734375,
-0.0025787353515625,
-0.01519775390625,
-0.028564453125,
0.023040771484375,
-0.0083465576171875,
0.037109375,
0.036895751953125,
0.00743865966796875,
0.006420135498046875,
-0.0313720703125,
-0.027679443359375,
0.0193939208984375,
0.02008056640625,
-0.013641357421875,
0.063720703125,
0.0017242431640625,
-0.00782012939453125,
0.046630859375,
-0.01052093505859375,
-0.028228759765625,
0.060638427734375,
0.028350830078125,
0.050262451171875,
-0.00537109375,
-0.009246826171875,
0.053192138671875,
0.030975341796875,
-0.012481689453125,
0.048126220703125,
-0.00026869773864746094,
-0.055206298828125,
-0.01328277587890625,
-0.04864501953125,
-0.0113372802734375,
0.051055908203125,
-0.056182861328125,
0.022613525390625,
-0.05303955078125,
-0.0255126953125,
0.0003941059112548828,
0.025299072265625,
-0.05108642578125,
0.0247650146484375,
0.00685882568359375,
0.074462890625,
-0.055633544921875,
0.064208984375,
0.033721923828125,
-0.04742431640625,
-0.074462890625,
-0.0340576171875,
-0.01093292236328125,
-0.072998046875,
0.041778564453125,
0.004978179931640625,
0.0179901123046875,
-0.0101165771484375,
-0.06719970703125,
-0.0828857421875,
0.10528564453125,
0.0218048095703125,
-0.029937744140625,
0.01457977294921875,
0.015838623046875,
0.0262451171875,
-0.0263824462890625,
0.043121337890625,
0.059112548828125,
0.04534912109375,
-0.00039196014404296875,
-0.05914306640625,
0.02264404296875,
-0.035736083984375,
-0.004638671875,
0.00014913082122802734,
-0.0843505859375,
0.096435546875,
-0.01263427734375,
0.0016117095947265625,
0.019500732421875,
0.047943115234375,
0.0369873046875,
0.01297760009765625,
0.03179931640625,
0.054656982421875,
0.05072021484375,
-0.01788330078125,
0.06243896484375,
0.005580902099609375,
0.03729248046875,
0.05609130859375,
-0.0097503662109375,
0.0557861328125,
0.0241851806640625,
-0.032989501953125,
0.040802001953125,
0.07696533203125,
-0.028350830078125,
0.054412841796875,
-0.01184844970703125,
-0.005764007568359375,
-0.004367828369140625,
-0.00909423828125,
-0.059112548828125,
0.02862548828125,
0.029296875,
-0.03192138671875,
-0.00904083251953125,
-0.0195465087890625,
0.0146942138671875,
-0.035369873046875,
-0.0225677490234375,
0.040557861328125,
-0.017120361328125,
-0.026763916015625,
0.07757568359375,
0.0024967193603515625,
0.051055908203125,
-0.045867919921875,
-0.00698089599609375,
-0.0204925537109375,
0.005405426025390625,
-0.0379638671875,
-0.06201171875,
0.0006723403930664062,
0.007488250732421875,
-0.01546478271484375,
0.0117950439453125,
0.031402587890625,
-0.0015668869018554688,
-0.0263214111328125,
0.023712158203125,
0.00225067138671875,
0.03521728515625,
0.00716400146484375,
-0.07080078125,
0.025482177734375,
0.02008056640625,
-0.042999267578125,
0.022979736328125,
0.0235443115234375,
0.01824951171875,
0.05902099609375,
0.07537841796875,
0.0123443603515625,
0.0114288330078125,
-0.0128173828125,
0.08587646484375,
-0.05572509765625,
-0.028472900390625,
-0.05560302734375,
0.035247802734375,
-0.0147857666015625,
-0.044281005859375,
0.051605224609375,
0.0626220703125,
0.06640625,
-0.004730224609375,
0.07012939453125,
-0.032257080078125,
0.039398193359375,
-0.0274658203125,
0.054656982421875,
-0.05419921875,
0.0113067626953125,
-0.01328277587890625,
-0.044921875,
-0.006793975830078125,
0.0596923828125,
-0.01351165771484375,
0.0006542205810546875,
0.0494384765625,
0.05303955078125,
-0.002750396728515625,
0.01346588134765625,
-0.0010662078857421875,
0.0200958251953125,
0.0299530029296875,
0.06256103515625,
0.04278564453125,
-0.07086181640625,
0.0634765625,
-0.045257568359375,
-0.011566162109375,
-0.0246124267578125,
-0.053375244140625,
-0.07098388671875,
-0.0126800537109375,
-0.0141754150390625,
-0.030731201171875,
-0.0136566162109375,
0.0546875,
0.045654296875,
-0.060333251953125,
-0.0318603515625,
-0.0024852752685546875,
0.005481719970703125,
-0.0321044921875,
-0.0194549560546875,
0.055755615234375,
0.0017337799072265625,
-0.0526123046875,
0.0308380126953125,
0.0015583038330078125,
0.0032634735107421875,
-0.0038814544677734375,
-0.01145172119140625,
-0.00958251953125,
-0.015380859375,
0.0140228271484375,
0.038909912109375,
-0.04681396484375,
-0.007762908935546875,
-0.02227783203125,
-0.00585174560546875,
0.0220794677734375,
0.013763427734375,
-0.045654296875,
0.0193023681640625,
0.03289794921875,
0.0157470703125,
0.056671142578125,
-0.00524139404296875,
-0.0011692047119140625,
-0.025299072265625,
0.0221405029296875,
-0.001739501953125,
0.032440185546875,
0.003437042236328125,
-0.034576416015625,
0.05975341796875,
0.027252197265625,
-0.0364990234375,
-0.0711669921875,
-0.030670166015625,
-0.09893798828125,
-0.00858306884765625,
0.072998046875,
-0.01476287841796875,
-0.050872802734375,
0.007476806640625,
-0.021697998046875,
0.03338623046875,
-0.05450439453125,
0.053680419921875,
0.021759033203125,
-0.011505126953125,
0.00853729248046875,
-0.039093017578125,
0.024688720703125,
-0.00782012939453125,
-0.061920166015625,
-0.0088653564453125,
-0.00772857666015625,
0.0226593017578125,
0.027130126953125,
0.0307464599609375,
-0.003810882568359375,
0.006591796875,
0.017059326171875,
0.019287109375,
-0.0247650146484375,
-0.003696441650390625,
0.0022068023681640625,
-0.0071563720703125,
-0.0229644775390625,
-0.04620361328125
]
] |
Azure99/blossom-v2-3b | 2023-08-11T11:53:58.000Z | [
"transformers",
"pytorch",
"bloom",
"text-generation",
"zh",
"en",
"dataset:Azure99/blossom-chat-v1",
"dataset:Azure99/blossom-math-v1",
"dataset:ehartford/dolphin",
"dataset:WizardLM/WizardLM_evol_instruct_V2_196k",
"license:apache-2.0",
"endpoints_compatible",
"has_space",
"text-generation-inference",
"region:us"
] | text-generation | Azure99 | null | null | Azure99/blossom-v2-3b | 0 | 5,943 | transformers | 2023-08-08T12:13:00 | ---
license: apache-2.0
datasets:
- Azure99/blossom-chat-v1
- Azure99/blossom-math-v1
- ehartford/dolphin
- WizardLM/WizardLM_evol_instruct_V2_196k
language:
- zh
- en
---
# **BLOSSOM-v2-3b**
### 介绍
Blossom是一个对话式语言模型，基于Bloom-3b预训练模型，在Blossom、Wizard、Dolphin混合数据集上进行指令精调得来。
训练分为两阶段，第一阶段使用120K Wizard、180K Dolphin单轮指令数据集，训练1个epoch；第二阶段使用60K Blossom chat、2K Blossom math多轮对话数据集，训练3个epoch。
### 推理
推理采用对话续写的形式。
单轮对话
```
A chat between a human and an artificial intelligence bot. The bot gives helpful, detailed, and polite answers to the human's questions.
|Human|: 你好
|Bot|:
```
多轮对话
```
A chat between a human and an artificial intelligence bot. The bot gives helpful, detailed, and polite answers to the human's questions.
|Human|: 你好
|Bot|: 你好，有什么我能帮助你的？</s>
|Human|: 介绍下中国的首都吧
|Bot|:
```
注意：在历史对话的Bot输出结尾，拼接一个</s> | 830 | [
[
-0.01788330078125,
-0.05859375,
0.0051727294921875,
0.06903076171875,
-0.0205841064453125,
-0.007373809814453125,
0.0118560791015625,
-0.036468505859375,
0.0223846435546875,
0.035858154296875,
-0.043914794921875,
-0.01519775390625,
-0.037994384765625,
-0.01030731201171875,
-0.0072021484375,
0.05517578125,
0.016876220703125,
0.01104736328125,
0.0011281967163085938,
0.032196044921875,
-0.055938720703125,
-0.00423431396484375,
-0.07025146484375,
-0.024505615234375,
0.0103912353515625,
0.033477783203125,
0.049163818359375,
0.017608642578125,
0.041290283203125,
0.0213775634765625,
-0.00360870361328125,
-0.0003733634948730469,
-0.045135498046875,
0.007396697998046875,
0.00591278076171875,
-0.05517578125,
-0.065673828125,
-0.0214385986328125,
0.0313720703125,
0.0225677490234375,
0.0207061767578125,
0.0125885009765625,
0.01508331298828125,
0.0748291015625,
-0.01751708984375,
0.0552978515625,
-0.036163330078125,
0.007843017578125,
-0.031524658203125,
-0.01541900634765625,
-0.0162506103515625,
-0.03961181640625,
-0.0148468017578125,
-0.0452880859375,
-0.01324462890625,
-0.00981903076171875,
0.0919189453125,
0.01056671142578125,
-0.03387451171875,
0.017364501953125,
-0.042633056640625,
0.0545654296875,
-0.036895751953125,
-0.0003581047058105469,
0.02362060546875,
0.045806884765625,
-0.01166534423828125,
-0.059295654296875,
-0.03668212890625,
0.016937255859375,
-0.0136871337890625,
0.032989501953125,
-0.01568603515625,
-0.023468017578125,
0.00797271728515625,
0.0145416259765625,
-0.0147705078125,
-0.0080718994140625,
-0.046875,
-0.0245513916015625,
0.03826904296875,
0.02728271484375,
0.044281005859375,
-0.00989532470703125,
-0.0184783935546875,
-0.00363922119140625,
-0.0347900390625,
0.0257568359375,
0.00440216064453125,
0.01447296142578125,
-0.034027099609375,
0.036102294921875,
0.00640869140625,
0.03350830078125,
-0.0078582763671875,
-0.01812744140625,
0.030242919921875,
-0.0445556640625,
-0.0175323486328125,
-0.043304443359375,
0.06439208984375,
0.05010986328125,
0.0171356201171875,
0.019134521484375,
0.0079803466796875,
-0.022979736328125,
-0.0085601806640625,
-0.0784912109375,
-0.0240631103515625,
0.0347900390625,
-0.046234130859375,
-0.0189361572265625,
0.03399658203125,
-0.064697265625,
0.0137786865234375,
0.0218353271484375,
0.0167694091796875,
-0.04473876953125,
-0.06396484375,
-0.005771636962890625,
-0.036407470703125,
0.02142333984375,
0.042816162109375,
-0.05487060546875,
0.0181121826171875,
0.0599365234375,
0.052154541015625,
-0.002902984619140625,
-0.02056884765625,
0.0007853507995605469,
0.01482391357421875,
-0.03375244140625,
0.030670166015625,
0.0103607177734375,
-0.043365478515625,
-0.01207733154296875,
-0.0005588531494140625,
0.001636505126953125,
-0.02557373046875,
0.027618408203125,
-0.060577392578125,
0.0244140625,
-0.02362060546875,
-0.0268096923828125,
-0.037017822265625,
0.0127716064453125,
-0.0435791015625,
0.0472412109375,
0.0259857177734375,
-0.056640625,
0.021575927734375,
-0.07025146484375,
-0.0272369384765625,
0.033172607421875,
-0.01702880859375,
-0.0120849609375,
-0.033294677734375,
-0.014495849609375,
0.0189666748046875,
-0.01580810546875,
-0.033660888671875,
-0.00215911865234375,
-0.01104736328125,
0.01552581787109375,
-0.037017822265625,
0.1029052734375,
0.0347900390625,
-0.019287109375,
0.0105438232421875,
-0.033416748046875,
0.008453369140625,
0.044403076171875,
-0.0099334716796875,
0.00946807861328125,
-0.005641937255859375,
0.01203155517578125,
0.01056671142578125,
0.06927490234375,
-0.0242156982421875,
0.0243072509765625,
-0.05218505859375,
0.0282135009765625,
0.040985107421875,
0.0196380615234375,
0.015045166015625,
-0.037994384765625,
0.0242767333984375,
0.0283355712890625,
0.0205078125,
-0.0236358642578125,
-0.0787353515625,
-0.05572509765625,
-0.006916046142578125,
-0.0049285888671875,
0.0693359375,
-0.039703369140625,
0.064453125,
-0.016357421875,
-0.041900634765625,
-0.025146484375,
0.0082855224609375,
0.0294647216796875,
0.0196685791015625,
0.0240631103515625,
-0.004058837890625,
-0.0238494873046875,
-0.050445556640625,
-0.002025604248046875,
-0.040069580078125,
-0.004688262939453125,
0.0440673828125,
0.0208892822265625,
-0.013671875,
0.069091796875,
-0.035858154296875,
-0.0299224853515625,
-0.032989501953125,
-0.0015783309936523438,
0.04046630859375,
0.031768798828125,
0.06707763671875,
-0.05072021484375,
-0.061126708984375,
-0.0037059783935546875,
-0.044189453125,
0.0102691650390625,
0.005558013916015625,
-0.02984619140625,
0.044921875,
-0.01366424560546875,
-0.027984619140625,
0.04339599609375,
0.021575927734375,
-0.050384521484375,
0.044769287109375,
-0.00980377197265625,
0.033355712890625,
-0.07867431640625,
-0.0079193115234375,
-0.041473388671875,
-0.006031036376953125,
-0.0318603515625,
-0.00024056434631347656,
0.00608062744140625,
0.038482666015625,
-0.05224609375,
0.052520751953125,
-0.030517578125,
0.0024204254150390625,
-0.0308380126953125,
0.0220184326171875,
-0.0015497207641601562,
0.028106689453125,
-0.023223876953125,
0.05224609375,
0.051544189453125,
-0.046051025390625,
0.056488037109375,
0.04931640625,
-0.009185791015625,
0.0064544677734375,
-0.0313720703125,
-0.005931854248046875,
0.023223876953125,
0.0204925537109375,
-0.07464599609375,
-0.02362060546875,
0.03790283203125,
-0.07080078125,
0.00211334228515625,
-0.0024166107177734375,
-0.033111572265625,
-0.053955078125,
-0.036346435546875,
0.0301666259765625,
0.06634521484375,
-0.01383209228515625,
0.0006823539733886719,
0.0180511474609375,
-0.0035839080810546875,
-0.0297698974609375,
-0.062744140625,
0.005901336669921875,
0.007965087890625,
-0.07904052734375,
0.0193634033203125,
-0.007572174072265625,
-0.00780487060546875,
-0.007289886474609375,
0.031890869140625,
0.0015201568603515625,
0.00855255126953125,
0.0018310546875,
0.03216552734375,
-0.0198516845703125,
-0.035491943359375,
0.0026149749755859375,
-0.01531982421875,
0.0215301513671875,
0.0206451416015625,
0.05596923828125,
-0.00952911376953125,
-0.04833984375,
-0.0595703125,
0.005237579345703125,
0.04888916015625,
0.0163116455078125,
0.0511474609375,
0.0546875,
-0.0231781005859375,
0.0178070068359375,
-0.027618408203125,
-0.00970458984375,
-0.037872314453125,
0.041534423828125,
-0.038818359375,
-0.05377197265625,
0.0279083251953125,
0.0224151611328125,
0.0220184326171875,
0.07452392578125,
0.043365478515625,
0.0019435882568359375,
0.08453369140625,
0.034637451171875,
-0.02880859375,
0.02984619140625,
-0.03961181640625,
0.0070037841796875,
-0.049285888671875,
-0.043426513671875,
-0.02459716796875,
-0.028411865234375,
-0.0271759033203125,
-0.023651123046875,
0.020355224609375,
0.025299072265625,
-0.043304443359375,
0.0274505615234375,
-0.04107666015625,
0.004703521728515625,
0.053985595703125,
0.0002815723419189453,
-0.004772186279296875,
-0.029693603515625,
0.0269317626953125,
0.0016984939575195312,
-0.0262298583984375,
-0.022369384765625,
0.04296875,
0.041290283203125,
0.01715087890625,
0.03009033203125,
0.0260162353515625,
0.001331329345703125,
-0.00304412841796875,
-0.0416259765625,
0.051055908203125,
0.0031566619873046875,
-0.048126220703125,
-0.033538818359375,
-0.0169525146484375,
-0.07470703125,
-0.00659942626953125,
0.00981903076171875,
-0.08062744140625,
-0.0261077880859375,
0.0006604194641113281,
-0.0175018310546875,
0.044952392578125,
-0.03009033203125,
0.064697265625,
-0.03900146484375,
-0.004550933837890625,
-0.0182037353515625,
-0.0472412109375,
0.048126220703125,
0.0097808837890625,
0.01611328125,
-0.0156707763671875,
-0.00537872314453125,
0.06109619140625,
-0.03497314453125,
0.0292205810546875,
-0.03643798828125,
0.006591796875,
0.02716064453125,
0.022430419921875,
0.042236328125,
0.0200347900390625,
0.0180511474609375,
0.00710296630859375,
0.0189361572265625,
-0.046722412109375,
-0.0244293212890625,
0.048370361328125,
-0.08197021484375,
-0.061187744140625,
-0.0330810546875,
-0.0121612548828125,
0.0028247833251953125,
0.0252685546875,
0.031829833984375,
-0.005523681640625,
-0.0039215087890625,
0.024505615234375,
-0.004634857177734375,
-0.0206451416015625,
0.0430908203125,
0.033660888671875,
-0.06103515625,
-0.040313720703125,
0.06707763671875,
-0.003814697265625,
0.0068817138671875,
0.0212249755859375,
0.004070281982421875,
-0.0178070068359375,
-0.0296478271484375,
-0.032989501953125,
0.0196990966796875,
-0.01035308837890625,
-0.02471923828125,
-0.05352783203125,
-0.055877685546875,
-0.06219482421875,
0.00246429443359375,
-0.0187225341796875,
-0.0240631103515625,
-0.051544189453125,
0.0022430419921875,
0.061126708984375,
0.04852294921875,
-0.009002685546875,
0.024200439453125,
-0.08306884765625,
0.0222930908203125,
0.0169830322265625,
0.0226898193359375,
0.00992584228515625,
-0.04345703125,
-0.029815673828125,
0.016510009765625,
-0.04339599609375,
-0.07684326171875,
0.0430908203125,
0.002838134765625,
0.034332275390625,
0.07916259765625,
-0.013671875,
0.0504150390625,
-0.01885986328125,
0.10302734375,
0.03656005859375,
-0.052703857421875,
0.04547119140625,
-0.031341552734375,
-0.002819061279296875,
0.0180206298828125,
-0.00470733642578125,
-0.039154052734375,
-0.0182037353515625,
-0.06298828125,
-0.07220458984375,
0.0831298828125,
0.03973388671875,
0.02685546875,
0.002071380615234375,
0.00115203857421875,
-0.00006651878356933594,
0.01045989990234375,
-0.04302978515625,
-0.06622314453125,
-0.024749755859375,
0.0041046142578125,
0.021575927734375,
-0.0321044921875,
-0.00588226318359375,
-0.026458740234375,
0.0684814453125,
0.035675048828125,
0.04107666015625,
-0.0201568603515625,
0.029571533203125,
-0.0240478515625,
-0.004192352294921875,
0.034637451171875,
0.031707763671875,
-0.0206756591796875,
-0.01349639892578125,
0.0131072998046875,
-0.04302978515625,
-0.00731658935546875,
-0.00798797607421875,
-0.021820068359375,
-0.0094146728515625,
0.03271484375,
0.055328369140625,
-0.0160369873046875,
-0.0284271240234375,
0.059326171875,
-0.0002701282501220703,
-0.002506256103515625,
-0.043182373046875,
0.046966552734375,
0.00897979736328125,
0.0213623046875,
0.029693603515625,
0.012237548828125,
-0.0013895034790039062,
-0.03314208984375,
0.02490234375,
0.04058837890625,
-0.0343017578125,
-0.0173187255859375,
0.053375244140625,
0.0259246826171875,
-0.058441162109375,
0.043304443359375,
-0.0247955322265625,
-0.048858642578125,
0.06304931640625,
0.0533447265625,
0.06658935546875,
0.007442474365234375,
0.01715087890625,
0.0465087890625,
-0.0018091201782226562,
-0.0008215904235839844,
0.04168701171875,
0.00988006591796875,
-0.0556640625,
-0.01119232177734375,
-0.030914306640625,
-0.0283355712890625,
0.0258331298828125,
-0.0185394287109375,
0.030914306640625,
-0.026702880859375,
-0.0152435302734375,
-0.01953125,
0.0184783935546875,
-0.024200439453125,
0.00958251953125,
-0.00240325927734375,
0.0911865234375,
-0.036865234375,
0.06365966796875,
0.02642822265625,
-0.04766845703125,
-0.050506591796875,
-0.006649017333984375,
0.00545501708984375,
-0.07562255859375,
0.062225341796875,
0.00437164306640625,
-0.0025577545166015625,
0.0063934326171875,
-0.04217529296875,
-0.07098388671875,
0.10003662109375,
-0.01410675048828125,
-0.007587432861328125,
0.0271148681640625,
0.0028209686279296875,
0.0261383056640625,
-0.02264404296875,
0.053619384765625,
0.032135009765625,
0.054534912109375,
0.0333251953125,
-0.06536865234375,
-0.01134490966796875,
-0.0478515625,
-0.0023517608642578125,
0.0022525787353515625,
-0.0794677734375,
0.06939697265625,
-0.0025882720947265625,
-0.03192138671875,
0.0203399658203125,
0.07183837890625,
0.0253143310546875,
0.0196380615234375,
0.041229248046875,
0.0158233642578125,
0.0100555419921875,
-0.04278564453125,
0.039886474609375,
-0.046661376953125,
0.01555633544921875,
0.046844482421875,
-0.00013375282287597656,
0.043304443359375,
-0.0029850006103515625,
-0.06817626953125,
0.041595458984375,
0.0675048828125,
-0.022705078125,
0.0257720947265625,
-0.01336669921875,
-0.008941650390625,
0.00970458984375,
0.0207672119140625,
-0.040252685546875,
-0.007160186767578125,
0.0281829833984375,
0.004131317138671875,
0.0005822181701660156,
-0.00617218017578125,
0.029296875,
-0.016754150390625,
-0.01861572265625,
0.08062744140625,
-0.0250091552734375,
-0.05987548828125,
0.020538330078125,
0.00939178466796875,
0.057891845703125,
-0.04180908203125,
-0.00688934326171875,
-0.024200439453125,
0.0104217529296875,
-0.0178375244140625,
-0.0870361328125,
0.0069732666015625,
-0.006130218505859375,
-0.002719879150390625,
0.005970001220703125,
0.043914794921875,
-0.0003197193145751953,
-0.03741455078125,
0.036956787109375,
0.039276123046875,
0.04095458984375,
0.0079345703125,
-0.08148193359375,
-0.01358795166015625,
0.0242462158203125,
-0.029693603515625,
0.00641632080078125,
0.036468505859375,
0.018463134765625,
0.052947998046875,
0.03350830078125,
0.03314208984375,
0.003570556640625,
0.01230621337890625,
0.048248291015625,
-0.059173583984375,
-0.035675048828125,
-0.06744384765625,
0.044281005859375,
-0.0207366943359375,
-0.0206146240234375,
0.057769775390625,
0.056304931640625,
0.044219970703125,
-0.00510406494140625,
0.066162109375,
-0.0275726318359375,
0.058990478515625,
-0.003208160400390625,
0.055419921875,
-0.0458984375,
0.00489044189453125,
-0.0234222412109375,
-0.042633056640625,
-0.01093292236328125,
0.0283660888671875,
-0.001171112060546875,
0.032501220703125,
0.041900634765625,
0.0394287109375,
0.020294189453125,
-0.0010929107666015625,
0.025054931640625,
0.036956787109375,
0.036956787109375,
0.04571533203125,
0.038238525390625,
-0.041168212890625,
0.047698974609375,
-0.033172607421875,
-0.034698486328125,
-0.048858642578125,
-0.0260162353515625,
-0.059844970703125,
-0.03472900390625,
-0.0048828125,
-0.04754638671875,
-0.032958984375,
0.06024169921875,
0.022491455078125,
-0.0716552734375,
-0.029815673828125,
-0.01346588134765625,
0.0161590576171875,
-0.035064697265625,
-0.021209716796875,
0.0391845703125,
-0.041046142578125,
-0.06597900390625,
-0.0006837844848632812,
0.03387451171875,
0.0200042724609375,
-0.0316162109375,
-0.0147705078125,
0.013214111328125,
0.01568603515625,
0.023040771484375,
0.0221099853515625,
-0.06280517578125,
-0.00911712646484375,
0.013336181640625,
-0.01580810546875,
0.0194244384765625,
0.015655517578125,
-0.028594970703125,
0.0122528076171875,
0.0599365234375,
-0.01026153564453125,
0.034759521484375,
-0.007965087890625,
0.0297698974609375,
-0.01654052734375,
0.01678466796875,
0.0015697479248046875,
0.031219482421875,
-0.0117950439453125,
-0.038909912109375,
0.0301971435546875,
0.031463623046875,
-0.048858642578125,
-0.02752685546875,
0.02642822265625,
-0.090087890625,
-0.035430908203125,
0.07855224609375,
-0.0028629302978515625,
-0.0305633544921875,
-0.00707244873046875,
-0.042236328125,
0.044769287109375,
-0.03497314453125,
0.047943115234375,
0.036163330078125,
-0.037078857421875,
-0.005035400390625,
-0.0241241455078125,
0.051971435546875,
0.00475311279296875,
-0.059722900390625,
-0.0257568359375,
0.0230712890625,
0.006046295166015625,
0.01922607421875,
0.056915283203125,
-0.0162506103515625,
0.032073974609375,
-0.00530242919921875,
-0.0021839141845703125,
-0.011260986328125,
-0.00951385498046875,
0.024322509765625,
0.0171051025390625,
-0.0001970529556274414,
-0.0338134765625
]
] |
Aeala/GPT4-x-Alpasta-13b | 2023-05-06T19:26:28.000Z | [
"transformers",
"pytorch",
"llama",
"text-generation",
"endpoints_compatible",
"has_space",
"text-generation-inference",
"region:us"
] | text-generation | Aeala | null | null | Aeala/GPT4-x-Alpasta-13b | 3 | 5,942 | transformers | 2023-05-06T16:51:34 | ## Pasta with a small twist!
Untested but fresh as of 5/6/2023, taste and hopefully enjoy! ^~^
## Model Info:
ChanSung's [AlpacaGPT4-LoRA-13B-elina](https://huggingface.co/LLMs/AlpacaGPT4-LoRA-13B-elina) merged with [dvruette's llama-13b sft do2 finetune](https://huggingface.co/dvruette/llama-13b-pretrained-sft-do2) | 318 | [
[
-0.041534423828125,
-0.07293701171875,
0.0170440673828125,
0.0404052734375,
-0.033660888671875,
0.00292205810546875,
0.023101806640625,
-0.04803466796875,
0.0716552734375,
0.0389404296875,
-0.067626953125,
0.00582122802734375,
-0.05535888671875,
0.01383209228515625,
-0.054351806640625,
0.066650390625,
-0.0028247833251953125,
0.034332275390625,
0.052581787109375,
-0.019317626953125,
-0.002201080322265625,
0.0208892822265625,
-0.03521728515625,
-0.039825439453125,
0.0755615234375,
0.048126220703125,
0.04852294921875,
0.03753662109375,
0.01412200927734375,
0.031890869140625,
-0.018463134765625,
0.0182342529296875,
-0.0277099609375,
-0.0288848876953125,
0.004543304443359375,
-0.019500732421875,
-0.05474853515625,
0.012969970703125,
0.01390838623046875,
0.059539794921875,
-0.0182342529296875,
0.0243682861328125,
0.0043487548828125,
0.054107666015625,
-0.0248870849609375,
-0.01190185546875,
-0.03955078125,
0.00467681884765625,
-0.0027923583984375,
0.0038890838623046875,
-0.010467529296875,
-0.048492431640625,
-0.00848388671875,
-0.054931640625,
0.006977081298828125,
-0.0017795562744140625,
0.10845947265625,
0.0216217041015625,
-0.041748046875,
-0.0293731689453125,
-0.0234375,
0.01560211181640625,
-0.04443359375,
0.01861572265625,
0.0243072509765625,
0.0263214111328125,
-0.05169677734375,
-0.06549072265625,
-0.037811279296875,
0.0007085800170898438,
0.021148681640625,
0.003444671630859375,
-0.0231475830078125,
-0.030548095703125,
-0.00930023193359375,
0.041107177734375,
-0.030670166015625,
0.01065826416015625,
-0.060333251953125,
0.02410888671875,
0.04644775390625,
-0.03155517578125,
0.0280609130859375,
-0.0033969879150390625,
-0.06878662109375,
0.01373291015625,
-0.0572509765625,
-0.01560211181640625,
0.042633056640625,
-0.0003197193145751953,
-0.0229339599609375,
0.06561279296875,
-0.0249481201171875,
0.0153045654296875,
0.055023193359375,
0.0243988037109375,
0.046630859375,
-0.02935791015625,
-0.0408935546875,
0.005039215087890625,
0.06390380859375,
0.0241546630859375,
0.0094146728515625,
0.0154571533203125,
0.007564544677734375,
-0.03448486328125,
-0.0007448196411132812,
-0.050262451171875,
-0.04071044921875,
-0.005649566650390625,
-0.032745361328125,
-0.044952392578125,
0.031768798828125,
-0.037078857421875,
-0.01218414306640625,
0.0002574920654296875,
0.0284271240234375,
-0.0284423828125,
-0.0264739990234375,
0.01471710205078125,
0.01171112060546875,
0.01067352294921875,
0.0452880859375,
-0.05426025390625,
0.0227508544921875,
0.0213775634765625,
0.03167724609375,
-0.0028934478759765625,
-0.0139007568359375,
-0.02459716796875,
0.011474609375,
-0.054290771484375,
0.06695556640625,
0.0207061767578125,
-0.010406494140625,
-0.0234832763671875,
0.0379638671875,
0.0278472900390625,
-0.04608154296875,
0.0389404296875,
-0.041351318359375,
0.003810882568359375,
-0.0440673828125,
-0.0244598388671875,
-0.02313232421875,
0.00980377197265625,
-0.07733154296875,
0.065673828125,
0.0172576904296875,
-0.037078857421875,
0.045806884765625,
-0.04937744140625,
-0.0089874267578125,
0.00384521484375,
0.0106964111328125,
-0.030426025390625,
0.017822265625,
-0.0016613006591796875,
0.0025959014892578125,
-0.0038204193115234375,
0.00630950927734375,
-0.033355712890625,
-0.02191162109375,
0.0390625,
-0.031158447265625,
0.0533447265625,
0.038116455078125,
-0.004077911376953125,
0.005374908447265625,
-0.06817626953125,
-0.00543975830078125,
0.0217742919921875,
-0.00926971435546875,
0.0026378631591796875,
-0.035003662109375,
0.0091552734375,
0.016326904296875,
0.04229736328125,
-0.0265960693359375,
0.02484130859375,
-0.0168914794921875,
0.03790283203125,
0.077880859375,
-0.015655517578125,
0.0223846435546875,
-0.038543701171875,
0.04046630859375,
0.01367950439453125,
0.03662109375,
0.0128631591796875,
-0.052947998046875,
-0.09649658203125,
-0.0183868408203125,
0.0212860107421875,
0.0335693359375,
-0.047515869140625,
0.062347412109375,
0.026275634765625,
-0.049346923828125,
-0.028594970703125,
0.0207977294921875,
-0.005340576171875,
0.00199127197265625,
0.0266571044921875,
-0.01302337646484375,
-0.047119140625,
-0.06732177734375,
0.01212310791015625,
-0.00836181640625,
0.0035266876220703125,
0.0007967948913574219,
0.04962158203125,
-0.0215911865234375,
0.03240966796875,
-0.045501708984375,
-0.0161590576171875,
-0.04205322265625,
0.01108551025390625,
0.033294677734375,
0.038848876953125,
0.058349609375,
-0.052581787109375,
-0.0504150390625,
0.01416778564453125,
-0.038421630859375,
-0.0213470458984375,
0.021148681640625,
-0.01317596435546875,
-0.003681182861328125,
0.02789306640625,
-0.054931640625,
0.05145263671875,
0.05987548828125,
-0.056060791015625,
0.02581787109375,
-0.006374359130859375,
0.044189453125,
-0.08929443359375,
-0.0267791748046875,
-0.032989501953125,
-0.0192413330078125,
-0.01058197021484375,
-0.00804901123046875,
-0.01163482666015625,
0.01236724853515625,
-0.06109619140625,
0.03961181640625,
-0.04107666015625,
-0.0008597373962402344,
-0.025848388671875,
-0.021240234375,
0.00714874267578125,
0.035552978515625,
-0.037078857421875,
0.0321044921875,
0.0188140869140625,
-0.0150146484375,
0.04937744140625,
0.022430419921875,
-0.039947509765625,
0.0177764892578125,
-0.064208984375,
-0.0013113021850585938,
0.0196533203125,
0.039520263671875,
-0.055145263671875,
-0.042083740234375,
0.042999267578125,
-0.0341796875,
-0.01349639892578125,
0.0213623046875,
-0.03704833984375,
-0.03704833984375,
-0.032989501953125,
0.046356201171875,
0.02593994140625,
-0.059295654296875,
0.0214996337890625,
-0.0141448974609375,
-0.003360748291015625,
-0.0273895263671875,
-0.074462890625,
-0.01136016845703125,
-0.03265380859375,
-0.0302276611328125,
0.031890869140625,
-0.0003077983856201172,
-0.0020313262939453125,
0.00667572021484375,
-0.0263824462890625,
-0.04803466796875,
-0.018035888671875,
0.01812744140625,
0.0135040283203125,
-0.0130462646484375,
-0.034454345703125,
0.0301361083984375,
-0.0013637542724609375,
-0.022247314453125,
0.01495361328125,
0.05426025390625,
-0.01351165771484375,
-0.0176544189453125,
-0.04754638671875,
0.036163330078125,
0.01861572265625,
0.01203155517578125,
0.047760009765625,
0.06610107421875,
-0.027923583984375,
-0.0172271728515625,
-0.062408447265625,
-0.024383544921875,
-0.042755126953125,
0.0009760856628417969,
-0.02960205078125,
-0.06646728515625,
0.045745849609375,
-0.0008249282836914062,
-0.04388427734375,
0.040313720703125,
0.032989501953125,
-0.0264739990234375,
0.0452880859375,
0.04052734375,
0.0243377685546875,
0.0208892822265625,
-0.0308074951171875,
0.0309600830078125,
-0.03790283203125,
-0.042510986328125,
-0.0259246826171875,
-0.0297698974609375,
-0.05340576171875,
-0.037750244140625,
0.0013360977172851562,
0.05596923828125,
-0.0291595458984375,
0.053436279296875,
-0.0159912109375,
0.050323486328125,
0.0273284912109375,
0.02789306640625,
0.038482666015625,
-0.01849365234375,
0.0054473876953125,
0.0108489990234375,
-0.045196533203125,
-0.00982666015625,
0.05419921875,
0.025146484375,
0.056427001953125,
0.006580352783203125,
0.04510498046875,
0.01001739501953125,
-0.0037841796875,
-0.02069091796875,
0.046905517578125,
-0.01824951171875,
-0.051177978515625,
-0.0070037841796875,
-0.0347900390625,
-0.07598876953125,
0.00643157958984375,
0.01218414306640625,
-0.047607421875,
0.0125274658203125,
0.016204833984375,
-0.01187896728515625,
0.0190582275390625,
-0.056732177734375,
0.059814453125,
-0.0251312255859375,
-0.01239776611328125,
-0.0004794597625732422,
-0.0260162353515625,
0.047149658203125,
0.0031642913818359375,
0.0105438232421875,
-0.037384033203125,
-0.01110076904296875,
0.03668212890625,
-0.059967041015625,
0.033905029296875,
0.01763916015625,
-0.0196533203125,
0.0154571533203125,
-0.004150390625,
0.0086669921875,
0.04278564453125,
0.015716552734375,
0.022552490234375,
-0.0186004638671875,
-0.035430908203125,
-0.03790283203125,
0.09027099609375,
-0.041412353515625,
-0.0323486328125,
-0.051239013671875,
-0.006916046142578125,
0.00804901123046875,
0.027801513671875,
0.059356689453125,
0.004230499267578125,
-0.0223236083984375,
0.01343536376953125,
0.032745361328125,
-0.00258636474609375,
0.04315185546875,
0.024200439453125,
-0.046661376953125,
-0.050872802734375,
0.0396728515625,
0.0096588134765625,
0.019866943359375,
0.01390838623046875,
0.0311431884765625,
-0.00733184814453125,
-0.0024566650390625,
-0.0196990966796875,
0.024932861328125,
-0.0172271728515625,
-0.0291748046875,
-0.01268768310546875,
-0.0160369873046875,
-0.017120361328125,
-0.01702880859375,
-0.052764892578125,
-0.051361083984375,
-0.0338134765625,
-0.027130126953125,
0.047332763671875,
0.07330322265625,
-0.028472900390625,
0.032196044921875,
-0.046356201171875,
0.029571533203125,
0.02459716796875,
0.032562255859375,
-0.042083740234375,
-0.029327392578125,
-0.004436492919921875,
0.0005755424499511719,
-0.04132080078125,
-0.06646728515625,
0.035736083984375,
0.0015659332275390625,
0.0034008026123046875,
0.044342041015625,
-0.019561767578125,
0.04974365234375,
-0.04388427734375,
0.0701904296875,
0.05645751953125,
-0.06549072265625,
0.0416259765625,
-0.034332275390625,
0.0252685546875,
0.034271240234375,
0.033538818359375,
0.0157928466796875,
-0.017669677734375,
-0.046478271484375,
-0.07171630859375,
0.034454345703125,
0.025390625,
0.00580596923828125,
-0.01000213623046875,
-0.0008449554443359375,
0.041412353515625,
0.035308837890625,
-0.08050537109375,
-0.035003662109375,
-0.034454345703125,
0.01007843017578125,
0.00821685791015625,
-0.0283203125,
-0.002964019775390625,
-0.05914306640625,
0.0289306640625,
-0.0010728836059570312,
0.0229034423828125,
-0.01485443115234375,
0.026947021484375,
-0.0250701904296875,
0.0022430419921875,
0.04937744140625,
0.0479736328125,
-0.029022216796875,
-0.000812530517578125,
0.018310546875,
-0.0293731689453125,
0.017913818359375,
0.0011796951293945312,
-0.01030731201171875,
0.00954437255859375,
0.025421142578125,
0.054229736328125,
0.05670166015625,
-0.035888671875,
0.031890869140625,
0.01152801513671875,
-0.017822265625,
-0.04840087890625,
0.00794219970703125,
0.007640838623046875,
0.027435302734375,
0.036834716796875,
-0.00170135498046875,
0.02825927734375,
-0.036895751953125,
0.0318603515625,
0.005580902099609375,
-0.0210113525390625,
-0.028472900390625,
0.050048828125,
0.00726318359375,
-0.0010442733764648438,
0.0201568603515625,
-0.0160980224609375,
-0.019439697265625,
0.037567138671875,
0.04241943359375,
0.06414794921875,
-0.055023193359375,
0.0191192626953125,
0.0233306884765625,
-0.00225067138671875,
-0.007274627685546875,
0.033233642578125,
0.0159454345703125,
-0.055023193359375,
-0.018798828125,
-0.0391845703125,
0.003253936767578125,
0.0219879150390625,
-0.0606689453125,
0.041351318359375,
-0.02191162109375,
-0.0192413330078125,
-0.0183258056640625,
0.005138397216796875,
-0.0229339599609375,
0.0245361328125,
-0.0189971923828125,
0.0894775390625,
-0.059417724609375,
0.093994140625,
0.0311279296875,
-0.039093017578125,
-0.059539794921875,
-0.01433563232421875,
-0.0012598037719726562,
-0.057525634765625,
0.0487060546875,
0.015777587890625,
-0.0028934478759765625,
-0.0260009765625,
-0.051361083984375,
-0.05218505859375,
0.093017578125,
0.0181732177734375,
-0.02642822265625,
-0.024200439453125,
-0.0037670135498046875,
0.007678985595703125,
-0.0212554931640625,
0.0035915374755859375,
0.03533935546875,
0.0369873046875,
0.049163818359375,
-0.07421875,
-0.032440185546875,
-0.039520263671875,
-0.005031585693359375,
0.0105133056640625,
-0.0733642578125,
0.07427978515625,
-0.020599365234375,
0.0204315185546875,
0.038787841796875,
0.07220458984375,
0.0266571044921875,
0.0130462646484375,
0.0377197265625,
0.05535888671875,
0.0236053466796875,
-0.031768798828125,
0.04998779296875,
0.023712158203125,
0.0234222412109375,
0.08441162109375,
-0.0253753662109375,
0.05694580078125,
0.0611572265625,
-0.00798797607421875,
0.052886962890625,
0.06622314453125,
-0.00365447998046875,
0.0516357421875,
0.00913238525390625,
0.0037212371826171875,
0.00653076171875,
-0.022705078125,
-0.052398681640625,
0.0268707275390625,
0.01190185546875,
-0.0246429443359375,
-0.00982666015625,
0.0054779052734375,
0.01104736328125,
0.014129638671875,
-0.01715087890625,
0.033050537109375,
0.0275421142578125,
-0.032684326171875,
0.042572021484375,
-0.0227203369140625,
0.055877685546875,
-0.0400390625,
-0.0093841552734375,
-0.03173828125,
0.01332855224609375,
-0.013641357421875,
-0.0516357421875,
0.0166168212890625,
-0.00392913818359375,
0.0019445419311523438,
-0.007175445556640625,
0.0177764892578125,
-0.033538818359375,
-0.058197021484375,
0.033660888671875,
0.01023101806640625,
-0.00394439697265625,
0.038330078125,
-0.055572509765625,
0.051788330078125,
0.00537109375,
-0.029022216796875,
0.0225677490234375,
0.0257720947265625,
0.0165863037109375,
0.0689697265625,
0.0249176025390625,
0.01296234130859375,
0.026214599609375,
0.0056915283203125,
0.06402587890625,
-0.0236053466796875,
-0.0294952392578125,
-0.0447998046875,
0.03765869140625,
-0.01169586181640625,
-0.037841796875,
0.056884765625,
0.046417236328125,
0.0723876953125,
-0.027435302734375,
0.01551055908203125,
-0.0128936767578125,
0.02825927734375,
-0.06695556640625,
0.0491943359375,
-0.06646728515625,
-0.0017042160034179688,
-0.017974853515625,
-0.07781982421875,
0.0024662017822265625,
0.06854248046875,
0.0187225341796875,
0.015838623046875,
0.066162109375,
0.040496826171875,
-0.0206146240234375,
0.006313323974609375,
0.0208282470703125,
0.02276611328125,
0.0108795166015625,
0.041168212890625,
0.057861328125,
-0.052093505859375,
0.00815582275390625,
-0.043243408203125,
-0.0208282470703125,
-0.036346435546875,
-0.07989501953125,
-0.057586669921875,
-0.0413818359375,
-0.021881103515625,
-0.01085662841796875,
-0.0015850067138671875,
0.065185546875,
0.050689697265625,
-0.05535888671875,
-0.02142333984375,
-0.00955963134765625,
0.0071563720703125,
-0.01247406005859375,
-0.01488494873046875,
0.019775390625,
0.0218658447265625,
-0.0595703125,
0.049530029296875,
0.01467132568359375,
0.027587890625,
-0.006725311279296875,
-0.01309967041015625,
-0.0343017578125,
0.02020263671875,
0.0285797119140625,
0.0117340087890625,
-0.06634521484375,
-0.0294036865234375,
-0.0071258544921875,
-0.00623321533203125,
-0.006664276123046875,
0.0260009765625,
-0.0039520263671875,
-0.0236968994140625,
0.03411865234375,
-0.0027484893798828125,
0.035858154296875,
0.01076507568359375,
0.0498046875,
-0.0204315185546875,
0.044952392578125,
-0.01102447509765625,
0.0362548828125,
0.0047760009765625,
0.0007724761962890625,
0.050628662109375,
0.0100860595703125,
-0.04443359375,
-0.06353759765625,
0.0191497802734375,
-0.128662109375,
-0.0196990966796875,
0.06695556640625,
0.00537109375,
-0.0306243896484375,
0.052886962890625,
-0.0239105224609375,
0.023651123046875,
-0.019195556640625,
0.039764404296875,
0.053619384765625,
-0.0182037353515625,
-0.03582763671875,
-0.0418701171875,
0.0287933349609375,
0.0186614990234375,
-0.0670166015625,
-0.0487060546875,
0.017181396484375,
0.040008544921875,
0.005039215087890625,
0.0228729248046875,
-0.018341064453125,
0.038177490234375,
-0.022216796875,
-0.01172637939453125,
-0.007022857666015625,
-0.04412841796875,
0.0220489501953125,
-0.0088348388671875,
0.01218414306640625,
-0.034820556640625
]
] |
Aeala/Alpaca-elina-65b | 2023-05-15T10:19:18.000Z | [
"transformers",
"pytorch",
"llama",
"text-generation",
"endpoints_compatible",
"has_space",
"text-generation-inference",
"region:us"
] | text-generation | Aeala | null | null | Aeala/Alpaca-elina-65b | 3 | 5,942 | transformers | 2023-05-11T18:36:17 | ## Model Info
Merge of ChanSung's [Alpaca-LoRA-65B-elina](https://huggingface.co/LLMs/Alpaca-LoRA-65B-elina)
## Benchmarks
Coming soon... for now, it was submitted to the [Open LLM Leaderboard](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard) | 260 | [
[
-0.0577392578125,
-0.058990478515625,
0.016937255859375,
0.0305023193359375,
-0.034820556640625,
-0.0193023681640625,
0.00835418701171875,
-0.064453125,
0.05474853515625,
0.042724609375,
-0.058624267578125,
-0.044921875,
-0.044921875,
-0.0095367431640625,
-0.0487060546875,
0.05413818359375,
-0.0201568603515625,
-0.01128387451171875,
0.00567626953125,
-0.03851318359375,
-0.0174560546875,
-0.02410888671875,
-0.0562744140625,
-0.0292816162109375,
0.054107666015625,
0.0306243896484375,
0.05584716796875,
0.040924072265625,
0.041534423828125,
0.0305938720703125,
-0.013397216796875,
0.0178070068359375,
-0.012481689453125,
0.00457000732421875,
0.007076263427734375,
-0.025054931640625,
-0.0982666015625,
0.01166534423828125,
0.0401611328125,
0.053192138671875,
-0.0249176025390625,
0.044891357421875,
0.01314544677734375,
0.037078857421875,
-0.03179931640625,
0.020477294921875,
-0.0211334228515625,
-0.003086090087890625,
-0.0382080078125,
0.01177978515625,
-0.0006847381591796875,
-0.0609130859375,
-0.01023101806640625,
-0.036376953125,
0.010040283203125,
0.0008096694946289062,
0.08612060546875,
0.016143798828125,
-0.03271484375,
-0.012054443359375,
-0.006580352783203125,
0.032989501953125,
-0.03564453125,
0.025115966796875,
0.029296875,
0.0284881591796875,
-0.01953125,
-0.0305328369140625,
-0.0181884765625,
0.0072479248046875,
0.007495880126953125,
-0.006359100341796875,
-0.01544189453125,
-0.034393310546875,
0.00664520263671875,
0.0350341796875,
-0.0179595947265625,
0.01212310791015625,
-0.050079345703125,
0.002567291259765625,
0.054840087890625,
0.005786895751953125,
0.0227813720703125,
0.00969696044921875,
-0.051788330078125,
-0.007244110107421875,
-0.0535888671875,
-0.0036106109619140625,
0.024169921875,
0.0147247314453125,
-0.029296875,
0.0648193359375,
-0.0330810546875,
0.047607421875,
0.017059326171875,
0.01025390625,
0.05206298828125,
-0.0276641845703125,
-0.0301513671875,
0.0220947265625,
0.062042236328125,
0.05499267578125,
-0.00835418701171875,
0.0179595947265625,
-0.0216217041015625,
-0.0049285888671875,
-0.00514984130859375,
-0.06683349609375,
0.00612640380859375,
0.00794219970703125,
-0.056854248046875,
-0.042633056640625,
0.024993896484375,
-0.058441162109375,
-0.0092010498046875,
0.00838470458984375,
0.0306854248046875,
-0.0089569091796875,
-0.03076171875,
0.0114288330078125,
0.0140228271484375,
0.04266357421875,
0.035308837890625,
-0.043060302734375,
0.035736083984375,
0.041595458984375,
0.054901123046875,
0.007328033447265625,
-0.01329803466796875,
-0.02740478515625,
0.0104217529296875,
-0.036468505859375,
0.058258056640625,
-0.0005817413330078125,
-0.0295562744140625,
-0.004474639892578125,
0.013275146484375,
0.0113677978515625,
-0.0521240234375,
0.06256103515625,
-0.0496826171875,
-0.01125335693359375,
-0.0562744140625,
-0.0175628662109375,
-0.01410675048828125,
0.024658203125,
-0.07415771484375,
0.087646484375,
0.0081634521484375,
-0.035430908203125,
0.024169921875,
-0.049591064453125,
0.005237579345703125,
0.0013513565063476562,
0.00528717041015625,
-0.03997802734375,
-0.002719879150390625,
-0.0011339187622070312,
0.03436279296875,
-0.0261077880859375,
-0.003162384033203125,
-0.031890869140625,
-0.031890869140625,
0.0347900390625,
0.00618743896484375,
0.051300048828125,
0.036041259765625,
-0.00193023681640625,
-0.0017147064208984375,
-0.04937744140625,
0.005397796630859375,
0.0164794921875,
-0.0154876708984375,
0.00675201416015625,
-0.0245819091796875,
-0.0018663406372070312,
0.00922393798828125,
0.052825927734375,
-0.0277557373046875,
0.0281982421875,
0.02288818359375,
0.00431060791015625,
0.05352783203125,
-0.01216888427734375,
0.00240325927734375,
-0.053924560546875,
0.041168212890625,
-0.0063629150390625,
0.0289459228515625,
0.01441192626953125,
-0.054290771484375,
-0.08038330078125,
-0.01245880126953125,
0.02374267578125,
0.0191802978515625,
0.00867462158203125,
0.055755615234375,
0.023681640625,
-0.07220458984375,
-0.0311431884765625,
0.01885986328125,
0.0186614990234375,
0.0222625732421875,
0.021759033203125,
-0.050048828125,
-0.045379638671875,
-0.06500244140625,
0.003448486328125,
-0.0162200927734375,
0.01100921630859375,
0.01222991943359375,
0.0260772705078125,
-0.0239410400390625,
0.02880859375,
-0.04205322265625,
-0.0231170654296875,
-0.014923095703125,
0.00662994384765625,
0.0399169921875,
0.0301513671875,
0.05731201171875,
-0.040740966796875,
-0.0345458984375,
0.00638580322265625,
-0.060455322265625,
-0.036376953125,
0.047698974609375,
-0.02679443359375,
0.001476287841796875,
0.049560546875,
-0.06134033203125,
0.053192138671875,
0.0615234375,
-0.035125732421875,
0.0265960693359375,
0.00627899169921875,
0.054290771484375,
-0.07958984375,
-0.0080108642578125,
0.00670623779296875,
-0.01546478271484375,
0.0021953582763671875,
0.022308349609375,
-0.0011348724365234375,
0.0025997161865234375,
-0.04833984375,
0.039581298828125,
-0.048492431640625,
-0.025787353515625,
-0.0227203369140625,
0.0107269287109375,
-0.001007080078125,
0.01288604736328125,
-0.0252838134765625,
0.035797119140625,
0.041473388671875,
-0.03271484375,
0.05023193359375,
0.0268402099609375,
-0.033294677734375,
0.031463623046875,
-0.0625,
-0.01534271240234375,
0.01500701904296875,
0.0223846435546875,
-0.06475830078125,
-0.0311126708984375,
0.021514892578125,
-0.00850677490234375,
0.00238037109375,
0.01153564453125,
-0.03900146484375,
-0.027496337890625,
-0.039093017578125,
0.06475830078125,
0.052581787109375,
-0.05120849609375,
0.0386962890625,
0.0211181640625,
-0.0070037841796875,
-0.0179290771484375,
-0.0513916015625,
-0.0239410400390625,
-0.01525115966796875,
-0.0401611328125,
0.04595947265625,
-0.019683837890625,
-0.029632568359375,
0.01861572265625,
0.010498046875,
-0.0176544189453125,
-0.0180816650390625,
0.036468505859375,
0.046630859375,
-0.0328369140625,
-0.0186920166015625,
-0.006008148193359375,
-0.00041866302490234375,
0.00402069091796875,
0.043304443359375,
0.05157470703125,
-0.035919189453125,
-0.026519775390625,
-0.049224853515625,
-0.0007319450378417969,
0.029632568359375,
0.0009584426879882812,
0.0606689453125,
0.035614013671875,
-0.045257568359375,
-0.00024235248565673828,
-0.06634521484375,
0.0188751220703125,
-0.0380859375,
0.0024547576904296875,
-0.0307769775390625,
-0.04119873046875,
0.053466796875,
0.0024433135986328125,
-0.0186767578125,
0.06451416015625,
0.051055908203125,
0.003955841064453125,
0.057586669921875,
0.056854248046875,
-0.00740814208984375,
0.027618408203125,
-0.012969970703125,
-0.0059967041015625,
-0.06329345703125,
-0.049560546875,
-0.041259765625,
-0.0328369140625,
-0.045806884765625,
-0.046142578125,
0.00243377685546875,
0.03240966796875,
-0.022430419921875,
0.06707763671875,
-0.035400390625,
0.03857421875,
0.031280517578125,
0.03167724609375,
0.043792724609375,
-0.040496826171875,
0.01236724853515625,
0.006908416748046875,
-0.0211334228515625,
-0.0226898193359375,
0.0804443359375,
0.01861572265625,
0.0643310546875,
0.016143798828125,
0.057952880859375,
0.01157379150390625,
0.0318603515625,
-0.0277557373046875,
0.0653076171875,
-0.000873565673828125,
-0.04022216796875,
0.0037097930908203125,
-0.0073089599609375,
-0.06292724609375,
0.0173797607421875,
-0.00658416748046875,
-0.038299560546875,
0.0117340087890625,
0.001155853271484375,
-0.0118408203125,
0.0226287841796875,
-0.036224365234375,
0.049835205078125,
-0.01776123046875,
-0.006427764892578125,
-0.0217437744140625,
-0.01448822021484375,
0.045440673828125,
-0.0193023681640625,
0.00766754150390625,
-0.04571533203125,
-0.01995849609375,
0.061767578125,
-0.062408447265625,
0.05377197265625,
0.0085296630859375,
-0.025604248046875,
0.02862548828125,
-0.01715087890625,
0.00848388671875,
-0.0020160675048828125,
-0.02166748046875,
0.0169830322265625,
-0.0279083251953125,
-0.026214599609375,
0.0012712478637695312,
0.08563232421875,
-0.06353759765625,
-0.00907135009765625,
-0.044189453125,
-0.0208587646484375,
-0.018402099609375,
0.00888824462890625,
0.033355712890625,
0.014068603515625,
-0.045654296875,
0.005558013916015625,
0.057830810546875,
-0.006618499755859375,
0.032989501953125,
0.024871826171875,
-0.05230712890625,
-0.031707763671875,
0.01824951171875,
0.01666259765625,
0.004756927490234375,
0.006988525390625,
0.0028629302978515625,
-0.016326904296875,
-0.021392822265625,
-0.01038360595703125,
0.034423828125,
-0.03350830078125,
-0.0266571044921875,
-0.0258941650390625,
-0.04840087890625,
-0.022857666015625,
-0.01788330078125,
-0.041534423828125,
-0.022186279296875,
-0.01824951171875,
-0.0237274169921875,
0.0469970703125,
0.07373046875,
-0.007602691650390625,
0.0330810546875,
-0.05072021484375,
0.030792236328125,
0.03472900390625,
0.041778564453125,
-0.0179901123046875,
-0.03814697265625,
-0.00299072265625,
-0.004673004150390625,
-0.030059814453125,
-0.053131103515625,
0.016448974609375,
0.003429412841796875,
0.040771484375,
0.030181884765625,
-0.0203857421875,
0.05712890625,
-0.012847900390625,
0.04840087890625,
0.036224365234375,
-0.06011962890625,
0.05413818359375,
-0.031463623046875,
0.00595855712890625,
0.03948974609375,
0.0313720703125,
-0.001514434814453125,
-0.0220184326171875,
-0.0435791015625,
-0.052520751953125,
0.04150390625,
-0.00313568115234375,
-0.01241302490234375,
0.009124755859375,
0.006084442138671875,
0.0186004638671875,
0.0308837890625,
-0.06219482421875,
-0.0386962890625,
-0.00677490234375,
0.00241851806640625,
0.00533294677734375,
-0.027496337890625,
-0.024658203125,
-0.03302001953125,
0.05242919921875,
0.00543975830078125,
-0.0029964447021484375,
-0.01068878173828125,
0.01910400390625,
-0.03814697265625,
-0.0202178955078125,
0.025299072265625,
0.039825439453125,
-0.039215087890625,
-0.034027099609375,
0.0116119384765625,
-0.0028076171875,
-0.00939178466796875,
0.014923095703125,
-0.00409698486328125,
0.000835418701171875,
0.033966064453125,
0.0482177734375,
0.0355224609375,
-0.035858154296875,
0.037872314453125,
-0.0117950439453125,
-0.017364501953125,
-0.0182342529296875,
0.01468658447265625,
0.01092529296875,
0.037567138671875,
0.024078369140625,
-0.0016336441040039062,
0.00238037109375,
-0.06292724609375,
0.01195526123046875,
0.034637451171875,
-0.0155029296875,
-0.0254364013671875,
0.06884765625,
0.014923095703125,
-0.011749267578125,
0.0294036865234375,
0.005779266357421875,
-0.016632080078125,
0.0689697265625,
0.03179931640625,
0.05914306640625,
-0.037811279296875,
0.0023326873779296875,
0.037078857421875,
0.006214141845703125,
-0.023223876953125,
0.035247802734375,
0.00518035888671875,
-0.046844482421875,
0.0003712177276611328,
-0.053192138671875,
-0.040313720703125,
-0.007068634033203125,
-0.06622314453125,
0.0640869140625,
-0.0298919677734375,
-0.0305328369140625,
-0.00865936279296875,
0.01214599609375,
-0.05584716796875,
0.029144287109375,
0.00666046142578125,
0.09149169921875,
-0.0626220703125,
0.06317138671875,
0.031097412109375,
-0.0394287109375,
-0.057952880859375,
-0.018890380859375,
-0.00689697265625,
-0.08477783203125,
0.026641845703125,
0.03033447265625,
-0.001983642578125,
-0.0438232421875,
-0.04364013671875,
-0.07501220703125,
0.11749267578125,
0.027069091796875,
-0.036468505859375,
0.0038547515869140625,
-0.01552581787109375,
0.020294189453125,
-0.0340576171875,
0.01242828369140625,
0.020050048828125,
0.047698974609375,
0.0214996337890625,
-0.0799560546875,
0.0028858184814453125,
-0.041534423828125,
-0.01251983642578125,
0.02972412109375,
-0.10662841796875,
0.08709716796875,
-0.01154327392578125,
0.01690673828125,
0.0455322265625,
0.06304931640625,
0.0345458984375,
0.0142669677734375,
0.042327880859375,
0.07373046875,
0.032135009765625,
-0.0264739990234375,
0.057952880859375,
0.0026645660400390625,
0.0270538330078125,
0.06353759765625,
-0.040191650390625,
0.068359375,
0.021331787109375,
-0.0201568603515625,
0.051910400390625,
0.06256103515625,
-0.00859832763671875,
0.037506103515625,
-0.0081939697265625,
-0.0193939208984375,
-0.0029888153076171875,
-0.033416748046875,
-0.08135986328125,
0.028045654296875,
0.003910064697265625,
-0.036224365234375,
-0.0212860107421875,
-0.043304443359375,
-0.0019464492797851562,
-0.006389617919921875,
-0.044677734375,
0.03515625,
0.01551055908203125,
-0.040863037109375,
0.021453857421875,
-0.0016984939575195312,
0.0648193359375,
-0.0596923828125,
-0.004302978515625,
-0.020477294921875,
0.007480621337890625,
-0.02911376953125,
-0.053802490234375,
0.0189208984375,
-0.0149078369140625,
-0.00009834766387939453,
-0.01529693603515625,
0.051544189453125,
-0.042388916015625,
-0.048187255859375,
0.053192138671875,
0.032989501953125,
0.0114288330078125,
0.022003173828125,
-0.04986572265625,
0.059295654296875,
-0.0259857177734375,
-0.02679443359375,
0.037353515625,
0.0151824951171875,
0.005031585693359375,
0.056396484375,
0.0399169921875,
0.0137176513671875,
0.00980377197265625,
-0.0001952648162841797,
0.06951904296875,
-0.052093505859375,
-0.02349853515625,
-0.051544189453125,
0.003978729248046875,
-0.0124664306640625,
-0.0396728515625,
0.05810546875,
0.054534912109375,
0.057586669921875,
0.01024627685546875,
0.02783203125,
-0.01323699951171875,
0.03326416015625,
-0.0435791015625,
0.0560302734375,
-0.060546875,
0.005146026611328125,
-0.0186920166015625,
-0.09454345703125,
0.01123046875,
0.0394287109375,
0.0290374755859375,
0.02191162109375,
0.043243408203125,
0.050628662109375,
-0.001434326171875,
0.0026531219482421875,
0.00794219970703125,
0.016998291015625,
0.0016622543334960938,
0.0452880859375,
0.043975830078125,
-0.060333251953125,
0.0232391357421875,
-0.04107666015625,
-0.02911376953125,
-0.031646728515625,
-0.08050537109375,
-0.05267333984375,
-0.0200042724609375,
-0.0305328369140625,
-0.02838134765625,
-0.005153656005859375,
0.047454833984375,
0.04315185546875,
-0.046295166015625,
-0.030426025390625,
0.01654052734375,
0.00531768798828125,
-0.01183319091796875,
-0.0150909423828125,
0.0236663818359375,
0.032928466796875,
-0.06890869140625,
0.026031494140625,
0.00798797607421875,
0.01020050048828125,
-0.013885498046875,
-0.0037631988525390625,
-0.00820159912109375,
0.0166473388671875,
0.024993896484375,
0.0208892822265625,
-0.04168701171875,
-0.006061553955078125,
-0.0198822021484375,
-0.032257080078125,
-0.0010709762573242188,
0.0259552001953125,
-0.03814697265625,
-0.0188446044921875,
0.033172607421875,
0.022308349609375,
0.036102294921875,
0.022430419921875,
0.03985595703125,
-0.022216796875,
0.021697998046875,
-0.0194244384765625,
0.06396484375,
0.006298065185546875,
-0.00726318359375,
0.0679931640625,
0.006381988525390625,
-0.043243408203125,
-0.06414794921875,
-0.0005464553833007812,
-0.11676025390625,
0.0120391845703125,
0.0582275390625,
0.01422882080078125,
-0.03729248046875,
0.04522705078125,
-0.02099609375,
0.0193634033203125,
-0.030242919921875,
0.0391845703125,
0.059967041015625,
-0.0196075439453125,
-0.0012149810791015625,
-0.0207366943359375,
0.015045166015625,
0.01505279541015625,
-0.077392578125,
-0.0361328125,
0.020416259765625,
0.0207672119140625,
0.00946044921875,
0.0626220703125,
-0.033477783203125,
0.040679931640625,
-0.0162506103515625,
0.006969451904296875,
-0.0081024169921875,
-0.015167236328125,
0.009490966796875,
0.01214599609375,
0.018280029296875,
-0.037872314453125
]
] |
jondurbin/airoboros-l2-70b-2.1 | 2023-09-08T09:25:02.000Z | [
"transformers",
"pytorch",
"llama",
"text-generation",
"dataset:jondurbin/airoboros-2.1",
"license:llama2",
"endpoints_compatible",
"has_space",
"text-generation-inference",
"region:us"
] | text-generation | jondurbin | null | null | jondurbin/airoboros-l2-70b-2.1 | 33 | 5,942 | transformers | 2023-08-26T13:33:56 | ---
license: llama2
datasets:
- jondurbin/airoboros-2.1
---
### Overview
__*This model is a bit broken due to a prompt formatting bug in the training code! 2.2 will be available soon and should fix this*__
This is an instruction fine-tuned llama-2 model, using synthetic data generated by [airoboros](https://github.com/jondurbin/airoboros)
- Experimental RP style instruction set, with two categories: rp and gtkm
- rp includes multi-round chats, with emotes, between a varying number of characters, defined by cards
- gtkm is a way to test a simpler alternative to ghost attention - first, a character card is generated, then several questions are created to ask the model (as the character), using the character system prompt, then everything in synthesized into a dialog (one system prompt, all turns remain in character)
- Experimental support for longer, more detailed writing prompts, as well as next-chapter generation
- I used the new `cull-instructions` entrypoint in airoboros to shrink the m2.0 dataset to a smaller subset of high-quality instructions (according to gpt-4)
- The training data now also includes "stylized_response", in which 1500 sample instructions from various categories were re-generated using character cards as system prompts.
- this should allow better adherence to style/etc. specified in the system card
- Thousands of new generations, using some of the updates re: Flesch hints, etc., to get longer/higher quality writing outputs.
- A small "de-alignment" dataset was also added (not published) to remove some of the censorship in the base models.
*Why do I try to remove censorship?*
- laws vary widely based on time and location
- language model may conflate certain words with laws, e.g. it may think "stealing eggs from a chicken" is illegal
- these models just produce text, what you do with that text is your resonsibility
- many people and industries deal with "sensitive" content; imagine if a court stenographer's equipment filtered illegal content - it would be useless
Huge thank you to the folks over at [a16z](https://a16z.com/) for sponsoring the costs associated with building models and associated tools!
### Prompt format
The training code was updated to randomize newline vs space:
https://github.com/jondurbin/qlora/blob/main/qlora.py#L559C1-L559C1
```
A chat. USER: {prompt} ASSISTANT:
```
or
```
A chat.
USER: {prompt}
ASSISTANT:
```
So in other words, it's the preamble/system prompt, followed by a single space or newline, then "USER: " (single space after colon) then the prompt (which can have multiple lines, spaces, whatever), then a single space or newline, followed by "ASSISTANT: " (with a single space after the colon).
__*I strongly suggest adding stopping criteria/early inference stopping on "USER:", because the training data includes many multi-round chats and could otherwise start simulating a conversation!*__
### Helpful usage tips
*The prompts shown here are are just the text that would be included after USER: and before ASSISTANT: in the full prompt format above, the system prompt and USER:/ASSISTANT: have been omited for readability.*
#### Context obedient question answering
By obedient, I mean the model was trained to ignore what it thinks it knows, and uses the context to answer the question. The model was also tuned to limit the values to the provided context as much as possible to reduce hallucinations.
The format for a closed-context prompt is as follows:
```
BEGININPUT
BEGINCONTEXT
[key0: value0]
[key1: value1]
... other metdata ...
ENDCONTEXT
[insert your text blocks here]
ENDINPUT
[add as many other blocks, in the exact same format]
BEGININSTRUCTION
[insert your instruction(s). The model was tuned with single questions, paragraph format, lists, etc.]
ENDINSTRUCTION
```
It's also helpful to add "Don't make up answers if you don't know." to your instruction block to make sure if the context is completely unrelated it doesn't make something up.
*The __only__ prompts that need this closed context formating are closed-context instructions. Normal questions/instructions do not!*
I know it's a bit verbose and annoying, but after much trial and error, using these explicit delimiters helps the model understand where to find the responses and how to associate specific sources with it.
- `BEGININPUT` - denotes a new input block
- `BEGINCONTEXT` - denotes the block of context (metadata key/value pairs) to associate with the current input block
- `ENDCONTEXT` - denotes the end of the metadata block for the current input
- [text] - Insert whatever text you want for the input block, as many paragraphs as can fit in the context.
- `ENDINPUT` - denotes the end of the current input block
- [repeat as many input blocks in this format as you want]
- `BEGININSTRUCTION` - denotes the start of the list (or one) instruction(s) to respond to for all of the input blocks above.
- [instruction(s)]
- `ENDINSTRUCTION` - denotes the end of instruction set
It sometimes works without `ENDINSTRUCTION`, but by explicitly including that in the prompt, the model better understands that all of the instructions in the block should be responded to.
Here's a trivial, but important example to prove the point:
```
BEGININPUT
BEGINCONTEXT
date: 2021-01-01
url: https://web.site/123
ENDCONTEXT
In a shocking turn of events, blueberries are now green, but will be sticking with the same name.
ENDINPUT
BEGININSTRUCTION
What color are bluberries? Source?
ENDINSTRUCTION
```
And the response:
```
Blueberries are now green.
Source:
date: 2021-01-01
url: https://web.site/123
```
#### Coding
You can ask for fairly complex coding instructions with multiple criteria, e.g.:
```
Create a python application with the following requirements:
- Asyncio FastAPI webserver
- ping endpoint that returns the current date in JSON format
- file upload endpoint, which calculates the file's sha256 checksum, and checks postgres to deduplicate
```
Or inline criteria:
```
Write a multi-threaded TCP server in C that accepts a "GET [key]" input and "SET [key] [value]" input, and uses a binary tree to get and store the input values.
```
You can also optionally add a single space and "PLAINFORMAT" at the end of your prompt to avoid backticks, explanations, etc. and just print the code, e.g.:
```
Write a websocket application in node.js. PLAINFORMAT
```
#### Agent/function calling
The dataset includes many examples of function/args generation based on input criteria. This is somewhat similar to the OpenAI function calling, but the output is either JSON or YAML.
Example prompt:
```
As an AI assistant, please select the most suitable function and parameters from the list of available functions below, based on the user's input. Provide your response in JSON format.
Input: I want to know how many times 'Python' is mentioned in my text file.
Available functions:
file_analytics:
description: This tool performs various operations on a text file.
params:
action: The operation we want to perform on the data, such as "count_occurrences", "find_line", etc.
filters:
keyword: The word or phrase we want to search for.
```
Response:
```json
{
"function": "file_analytics",
"params": {
"action": "count_occurrences",
"filters": {
"keyword": "Python"
}
}
}
```
#### Chain-of-thought
You can ask for several possible responses to a given problem, with a ranking and final answer selection.
Example prompt:
```
A group of people decide to play a round-robin tournament where every player plays every other player exactly once. If a total of 45 games were played, how many players participated in the tournament? Offer several potential answers, rank them according to mathematical logic, and pick the most feasible one.
```
Example response:
```
Solution 1:
In a round-robin tournament, each player plays every other player exactly once. This means that the total number of games played is equal to the combination of all players taken two at a time (since a game involves two players).
The formula for combinations is nC2 = n(n-1)/2, where n is the total number of players. We know that nC2 equals 45 (the total number of games), so we can set up the equation as follows:
n(n-1)/2 = 45
Solving this equation gives us n=10.
Final answer: There were 10 players in the tournament.
Solution 2:
Another way to approach this problem is by using the concept of triangular numbers. Triangular numbers are a sequence of numbers where each number can be represented as a triangle with dots. The nth triangular number is given by the sum of the first n natural numbers.
If we consider each game as a dot and each player as a side of the triangle, then the total number of games (45) would correspond to the 9th triangular number because 1+2+3+4+5+6+7+8+9=45. However, since each side of the triangle represents a player, and there's one more player than sides in our model (because the last player has no one left to play against), we need to add one more to get the total number of players.
Final answer: There were 10 players in the tournament.
Solution 3:
We could also use trial and error to solve this problem. Starting from a small number of players, we calculate how many games would be played in a round-robin tournament until we reach 45 games. For example, with 2 players, only 1 game is played; with 3 players, 3 games are played; with 4 players, 6 games are played, and so on. Continuing this process, we find that with 10 players, 45 games are played.
Final answer: There were 10 players in the tournament.
Ranking of solutions from best to worst:
1. Solution 1: This solution uses direct mathematical reasoning based on the properties of a round-robin tournament and solves an equation to find the answer.
2. Solution 2: This solution uses the concept of triangular numbers which requires additional steps and knowledge beyond the immediate problem context, but still arrives at the correct answer.
3. Solution 3: This solution relies on trial and error, which is less efficient and reliable than the other methods, but it eventually gets to the right answer.
Best and final answer: There were 10 players in the tournament.
```
#### reWOO style execution planning
The model now supports execution planning for complex instructions that would require making use of several tools. The output is just the plan, you must implement a mechanism to parse the output and actually call the functions!
Example prompt:
```
Please construct a systematic plan to generate an optimal response to the user instruction, utilizing a set of provided tools. Each plan will correspond to an evidence value, which will be the output of one of the available functions given an input string
that could be the user's question, one or more prior evidence values, or a combination of both.
Here are the tools available to be called:
DuckDuckGo[input]: Tool that enables the user to search for information using the DuckDuckGo search engine. This tool is helpful in locating a suitable list of sites that may contain the answer to the user's query. It does not directly answer the question but finds a list of sites that might have the answer.
TextScraper[input]: Load one or more websites from the input string containing newline delimited links, where input is one or more links, and produces plain text output containing the content of the links.
HyperlinkExtractor[input]: Extract links from plain text and produces a plain text, newline delimited response of links.
KnowledgeModel[input]: Question answering language model, particularly useful in answering questions based on an input passage of text. The input must be a text question that references an :evidence[n]: variable, e.g. What color is the cat, given :evidence1:?
The output should be in format:
Plan: [first action to take based on input question]
:evidence0: = [function to call with input parameter]
Plan: [next action to take, based on result of :evidence0:]
:evidence1: = [next function to call and input parameter, which may include reference to previous evidence, e.g. "Given context :evidence0"]
...
Answer: [:evidence[n]: containing the final answer.]
Question: Who won the Nobel Prize in Literature in 2020 and what is their most famous work?
```
Response:
```
Plan: Begin by conducting a web search to find out who won the Nobel Prize in Literature in 2020.
:evidence0: = DuckDuckGo[Nobel Prize in Literature 2020 winner]
Plan: Utilize the KnowledgeModel to interpret the search results and find the name of the laureate.
:evidence1: = KnowledgeModel[Who won the Nobel Prize in Literature in 2020, given :evidence0:?]
Plan: Conduct another web search to find the most famous work of the identified laureate.
:evidence2: = DuckDuckGo[Most famous work of :evidence1:]
Plan: Extract the relevant links from the DuckDuckGo search results for a more focused search.
:evidence3: = HyperlinkExtractor[:evidence2:]
Plan: Use the TextScraper tool to extract information from the relevant links.
:evidence4: = TextScraper[:evidence3:]
Plan: Finally, utilize the KnowledgeModel to identify and summarize the most famous work of the laureate from the extracted information.
:evidence5: = KnowledgeModel[What is the most famous work of :evidence1:, given :evidence4:?]
Answer: :evidence5:
```
For this to be useful, you'd have to parse the output plan text, and implement/call each of the functions. This is just pseudo-code, completely untested off the top of my head, and obviously would requiring full implementation + hardening:
```python
import re
import requests
def inject_context(input_text, **context):
for ref in set(re.findall(r"(:evidence[0-9]+:)", input_text, re.I)):
input_text = input_text.replace(ref, context.get(ref, ""))
return input_text
def duckduckgo(input_text, **context):
search_string = inject_context(input_text, **context)
... search via duck duck go using search_string
... return text content
def link_extractor(input_text, **context):
input_text = inject_context(input_text, **context)
return "\n".join(list(set(re.findall(r"(https?://[^\s]+?\.?)", input_text, re.I))))
def scrape(input_text, **context):
input_text = inject_context(input_text, **context)
text = []
for link in input_text.splitlines():
text.append(requests.get(link).text)
return "\n".join(text)
def infer(input_text, **context)
prompt = inject_context(input_text, **context)
... call model with prompt, return output
def parse_plan(plan):
method_map = {
"DuckDuckGo": duckduckgo,
"HyperlinkExtractor": link_extractor,
"KnowledgeModel": infer,
"TextScraper": scrape,
}
context = {}
for line in plan.strip().splitlines():
if line.startswith("Plan:"):
print(line)
continue
parts = re.match("^(:evidence[0-9]+:)\s*=\s*([^\[]+])(\[.*\])\s$", line, re.I)
if not parts:
if line.startswith("Answer: "):
return context.get(line.split(" ")[-1].strip(), "Answer couldn't be generated...")
raise RuntimeError("bad format: " + line)
context[parts.group(1)] = method_map[parts.group(2)](parts.group(3), **context)
```
### Contribute
If you're interested in new functionality, particularly a new "instructor" type to generate a specific type of training data,
take a look at the dataset generation tool repo: https://github.com/jondurbin/airoboros and either make a PR or open an issue with details.
To help me with the OpenAI/compute costs:
- https://bmc.link/jondurbin
- ETH 0xce914eAFC2fe52FdceE59565Dd92c06f776fcb11
- BTC bc1qdwuth4vlg8x37ggntlxu5cjfwgmdy5zaa7pswf
### Licence and usage restrictions
The airoboros 2.1 models are built on top of llama-2.
The llama-2 base model has a custom Meta license:
- See the [meta-license/LICENSE.txt](meta-license/LICENSE.txt) file attached for the original license provided by Meta.
- See also [meta-license/USE_POLICY.md](meta-license/USE_POLICY.md) and [meta-license/Responsible-Use-Guide.pdf](meta-license/Responsible-Use-Guide.pdf), also provided by Meta.
The fine-tuning data was generated by OpenAI API calls to gpt-4, via [airoboros](https://github.com/jondurbin/airoboros)
The ToS for OpenAI API usage has a clause preventing the output from being used to train a model that __competes__ with OpenAI
- what does *compete* actually mean here?
- these small open source models will not produce output anywhere near the quality of gpt-4, or even gpt-3.5, so I can't imagine this could credibly be considered competing in the first place
- if someone else uses the dataset to do the same, they wouldn't necessarily be violating the ToS because they didn't call the API, so I don't know how that works
- the training data used in essentially all large language models includes a significant amount of copyrighted or otherwise non-permissive licensing in the first place
- other work using the self-instruct method, e.g. the original here: https://github.com/yizhongw/self-instruct released the data and model as apache-2
I am purposingly leaving this license ambiguous (other than the fact you must comply with the Meta original license for llama-2) because I am not a lawyer and refuse to attempt to interpret all of the terms accordingly.
Your best bet is probably to avoid using this commercially due to the OpenAI API usage.
Either way, by using this model, you agree to completely indemnify me. | 17,468 | [
[
-0.02081298828125,
-0.07989501953125,
0.034332275390625,
0.01861572265625,
-0.0089263916015625,
-0.016357421875,
-0.01303863525390625,
-0.023101806640625,
0.01390838623046875,
0.037841796875,
-0.06103515625,
-0.043792724609375,
-0.0279693603515625,
0.0167083740234375,
-0.018310546875,
0.08343505859375,
0.003215789794921875,
-0.00251007080078125,
-0.01020050048828125,
0.00988006591796875,
-0.050048828125,
-0.0360107421875,
-0.06494140625,
-0.0025234222412109375,
0.031585693359375,
0.039886474609375,
0.0352783203125,
0.053314208984375,
0.03399658203125,
0.0273895263671875,
-0.0036163330078125,
0.0230560302734375,
-0.0303497314453125,
0.0103302001953125,
-0.006374359130859375,
-0.03973388671875,
-0.027618408203125,
0.0035533905029296875,
0.035186767578125,
0.035614013671875,
-0.00885009765625,
0.01806640625,
0.00616455078125,
0.02093505859375,
-0.044281005859375,
0.015716552734375,
-0.03228759765625,
-0.0002722740173339844,
-0.003826141357421875,
-0.030670166015625,
-0.035675048828125,
-0.01342010498046875,
0.007640838623046875,
-0.06805419921875,
-0.002246856689453125,
0.01324462890625,
0.07269287109375,
0.02069091796875,
-0.03302001953125,
-0.0296783447265625,
-0.03790283203125,
0.05780029296875,
-0.067138671875,
0.008087158203125,
0.051666259765625,
0.0298309326171875,
-0.02984619140625,
-0.062103271484375,
-0.0489501953125,
-0.0215301513671875,
-0.019256591796875,
0.01065826416015625,
-0.004779815673828125,
0.0023479461669921875,
0.04022216796875,
-0.005645751953125,
-0.05511474609375,
-0.0083465576171875,
-0.040374755859375,
-0.01541900634765625,
0.049102783203125,
0.03143310546875,
0.020782470703125,
-0.0168609619140625,
-0.02569580078125,
-0.0101165771484375,
-0.0291290283203125,
0.0261993408203125,
0.03399658203125,
0.0311279296875,
-0.0282440185546875,
0.039093017578125,
-0.02783203125,
0.049102783203125,
0.001068115234375,
-0.01313018798828125,
0.007598876953125,
-0.0284576416015625,
-0.0142822265625,
-0.015594482421875,
0.085205078125,
0.047332763671875,
0.0089111328125,
0.0085296630859375,
0.00038242340087890625,
-0.00655364990234375,
0.0152587890625,
-0.0628662109375,
-0.013916015625,
0.044342041015625,
-0.04248046875,
-0.031280517578125,
-0.0092620849609375,
-0.06219482421875,
-0.0198516845703125,
-0.00811767578125,
0.03717041015625,
-0.03533935546875,
0.00205230712890625,
0.008819580078125,
-0.021575927734375,
0.01251220703125,
0.036468505859375,
-0.07159423828125,
0.038970947265625,
0.0294647216796875,
0.06439208984375,
0.00396728515625,
-0.0333251953125,
-0.040191650390625,
-0.0131072998046875,
-0.0072021484375,
0.062042236328125,
-0.0345458984375,
-0.0202178955078125,
-0.01361083984375,
0.024017333984375,
-0.0034503936767578125,
-0.0194091796875,
0.0208282470703125,
-0.0258026123046875,
0.041748046875,
-0.021759033203125,
-0.03363037109375,
-0.020172119140625,
0.027069091796875,
-0.0291748046875,
0.0726318359375,
0.0015993118286132812,
-0.059356689453125,
-0.01192474365234375,
-0.07568359375,
-0.0250091552734375,
-0.005649566650390625,
0.001140594482421875,
-0.004199981689453125,
-0.0255584716796875,
0.0093841552734375,
0.033843994140625,
-0.023284912109375,
0.008056640625,
-0.017578125,
-0.033721923828125,
0.041900634765625,
-0.0264129638671875,
0.08404541015625,
0.022705078125,
-0.01085662841796875,
0.0177764892578125,
-0.062347412109375,
0.0034465789794921875,
0.01554107666015625,
-0.035491943359375,
-0.01343536376953125,
0.00603485107421875,
0.00627899169921875,
-0.002079010009765625,
0.025054931640625,
-0.0408935546875,
0.027862548828125,
-0.025634765625,
0.06005859375,
0.051239013671875,
0.01413726806640625,
0.02532958984375,
-0.0232391357421875,
0.035491943359375,
-0.005626678466796875,
0.025970458984375,
-0.041473388671875,
-0.044464111328125,
-0.05255126953125,
0.0027179718017578125,
0.00977325439453125,
0.07366943359375,
-0.04278564453125,
0.039031982421875,
0.004695892333984375,
-0.035552978515625,
-0.026214599609375,
-0.0142364501953125,
0.0279693603515625,
0.053558349609375,
0.03631591796875,
-0.01033782958984375,
-0.053955078125,
-0.05682373046875,
0.013275146484375,
-0.016632080078125,
0.003063201904296875,
0.0304412841796875,
0.05072021484375,
-0.017303466796875,
0.06378173828125,
-0.060272216796875,
0.0005507469177246094,
-0.014556884765625,
0.00496673583984375,
0.01407623291015625,
0.0496826171875,
0.0247650146484375,
-0.0477294921875,
-0.0276031494140625,
-0.0142059326171875,
-0.06103515625,
-0.004207611083984375,
-0.01300811767578125,
-0.018646240234375,
0.00018298625946044922,
0.0277557373046875,
-0.047393798828125,
0.0380859375,
0.0208740234375,
-0.039520263671875,
0.05157470703125,
-0.00988006591796875,
0.0193023681640625,
-0.09710693359375,
0.01751708984375,
-0.011566162109375,
-0.01041412353515625,
-0.053314208984375,
0.0267486572265625,
-0.00867462158203125,
-0.01183319091796875,
-0.037261962890625,
0.049285888671875,
-0.0250701904296875,
0.01215362548828125,
-0.0161285400390625,
0.00945281982421875,
0.0116424560546875,
0.055145263671875,
-0.0013580322265625,
0.0545654296875,
0.041229248046875,
-0.05615234375,
0.0474853515625,
0.0225067138671875,
-0.002422332763671875,
0.02734375,
-0.063720703125,
0.0240631103515625,
-0.006732940673828125,
0.021331787109375,
-0.09271240234375,
-0.0164947509765625,
0.047454833984375,
-0.054443359375,
0.00557708740234375,
-0.0110321044921875,
-0.027923583984375,
-0.0298309326171875,
-0.0282745361328125,
0.01357269287109375,
0.03533935546875,
-0.0191802978515625,
0.04583740234375,
0.0216522216796875,
0.006610870361328125,
-0.047027587890625,
-0.05072021484375,
0.006732940673828125,
-0.0222930908203125,
-0.046630859375,
0.0148773193359375,
-0.0316162109375,
-0.01861572265625,
-0.0138397216796875,
0.01617431640625,
-0.0242919921875,
0.0306854248046875,
0.0179290771484375,
0.0194854736328125,
-0.00628662109375,
-0.0024394989013671875,
0.009124755859375,
0.0014123916625976562,
0.005931854248046875,
-0.0165252685546875,
0.06866455078125,
-0.011016845703125,
-0.009521484375,
-0.061065673828125,
0.036224365234375,
0.0234527587890625,
-0.0181121826171875,
0.047454833984375,
0.03717041015625,
-0.036712646484375,
0.00441741943359375,
-0.024017333984375,
-0.033050537109375,
-0.039764404296875,
0.021240234375,
-0.0216827392578125,
-0.046905517578125,
0.05145263671875,
0.020477294921875,
0.025482177734375,
0.0238800048828125,
0.0272216796875,
-0.0232086181640625,
0.0716552734375,
0.030609130859375,
0.0160675048828125,
0.02996826171875,
-0.03460693359375,
-0.00274658203125,
-0.06134033203125,
-0.0264434814453125,
-0.045166015625,
-0.02679443359375,
-0.0333251953125,
-0.0179901123046875,
0.018310546875,
0.02117919921875,
-0.0299072265625,
0.036407470703125,
-0.050445556640625,
0.03192138671875,
0.047760009765625,
0.00823211669921875,
0.0093841552734375,
-0.0112152099609375,
-0.001537322998046875,
0.00394439697265625,
-0.042572021484375,
-0.04901123046875,
0.08001708984375,
0.028106689453125,
0.05810546875,
-0.0015316009521484375,
0.061553955078125,
0.014617919921875,
0.0002193450927734375,
-0.06365966796875,
0.05487060546875,
0.00725555419921875,
-0.035186767578125,
-0.0341796875,
-0.0184783935546875,
-0.07867431640625,
0.01629638671875,
-0.012786865234375,
-0.0770263671875,
0.00725555419921875,
0.0182647705078125,
-0.061492919921875,
-0.0006108283996582031,
-0.060791015625,
0.0732421875,
-0.0152587890625,
-0.02490234375,
0.006397247314453125,
-0.059234619140625,
0.0188446044921875,
0.023223876953125,
0.01280975341796875,
-0.00103759765625,
-0.004962921142578125,
0.065673828125,
-0.0487060546875,
0.08062744140625,
-0.0156402587890625,
0.01275634765625,
0.03570556640625,
0.0018949508666992188,
0.033203125,
0.0130767822265625,
0.00458526611328125,
0.005588531494140625,
0.0235595703125,
-0.01253509521484375,
-0.052947998046875,
0.039520263671875,
-0.06414794921875,
-0.036407470703125,
-0.0266571044921875,
-0.047698974609375,
0.0146636962890625,
0.0222320556640625,
0.031341552734375,
0.042877197265625,
-0.009796142578125,
-0.0020351409912109375,
0.0440673828125,
-0.0306549072265625,
0.03851318359375,
0.048431396484375,
-0.02081298828125,
-0.03790283203125,
0.052093505859375,
0.005146026611328125,
0.002269744873046875,
0.041717529296875,
0.0278167724609375,
-0.01690673828125,
-0.0249481201171875,
-0.052520751953125,
0.012664794921875,
-0.051177978515625,
-0.0128173828125,
-0.073974609375,
-0.01107025146484375,
-0.048828125,
-0.0004482269287109375,
-0.00797271728515625,
-0.0367431640625,
-0.044219970703125,
0.0019426345825195312,
0.044281005859375,
0.038055419921875,
0.00865936279296875,
0.04632568359375,
-0.053131103515625,
0.01551055908203125,
0.0147705078125,
0.002834320068359375,
0.005466461181640625,
-0.050384521484375,
-0.01303863525390625,
0.01763916015625,
-0.036712646484375,
-0.08428955078125,
0.029754638671875,
0.0015134811401367188,
0.0350341796875,
0.037353515625,
0.0034770965576171875,
0.0582275390625,
-0.044036865234375,
0.07965087890625,
-0.0018701553344726562,
-0.06195068359375,
0.06414794921875,
-0.03607177734375,
0.0157470703125,
0.03546142578125,
0.02978515625,
-0.060089111328125,
-0.022064208984375,
-0.038299560546875,
-0.057769775390625,
0.07147216796875,
0.0144500732421875,
0.0163116455078125,
-0.0133209228515625,
0.0379638671875,
-0.003021240234375,
0.01390838623046875,
-0.06048583984375,
-0.0328369140625,
-0.026123046875,
-0.01495361328125,
0.010345458984375,
-0.011505126953125,
-0.0151214599609375,
-0.0260467529296875,
0.04150390625,
-0.01410675048828125,
0.04443359375,
0.014190673828125,
0.00266265869140625,
0.00555419921875,
0.0166168212890625,
0.058837890625,
0.0384521484375,
-0.0185394287109375,
0.0005197525024414062,
0.016357421875,
-0.036346435546875,
0.00366973876953125,
0.0092315673828125,
-0.020172119140625,
-0.021514892578125,
0.0304412841796875,
0.068603515625,
0.0023097991943359375,
-0.052581787109375,
0.0322265625,
-0.01904296875,
-0.006908416748046875,
-0.0275421142578125,
0.0257110595703125,
0.01025390625,
0.0130615234375,
0.0215301513671875,
-0.0084381103515625,
0.02264404296875,
-0.05230712890625,
0.00299072265625,
0.0248565673828125,
-0.006160736083984375,
-0.0280303955078125,
0.051849365234375,
0.01629638671875,
-0.056243896484375,
0.04632568359375,
-0.0303192138671875,
-0.04156494140625,
0.06414794921875,
0.054840087890625,
0.052093505859375,
-0.0126800537109375,
0.01947021484375,
0.041900634765625,
0.023162841796875,
-0.0107269287109375,
0.046539306640625,
-0.01224517822265625,
-0.054229736328125,
-0.005718231201171875,
-0.041168212890625,
-0.026123046875,
0.021728515625,
-0.043731689453125,
0.0264129638671875,
-0.05865478515625,
-0.01169586181640625,
-0.0010833740234375,
0.00238037109375,
-0.04962158203125,
0.0142669677734375,
-0.01263427734375,
0.07220458984375,
-0.0703125,
0.039276123046875,
0.06597900390625,
-0.054656982421875,
-0.06524658203125,
-0.00701141357421875,
0.00421905517578125,
-0.050811767578125,
0.0291290283203125,
0.02490234375,
0.01366424560546875,
0.0007562637329101562,
-0.07147216796875,
-0.07421875,
0.08953857421875,
0.00794219970703125,
-0.031463623046875,
-0.017120361328125,
-0.007122039794921875,
0.04449462890625,
-0.038238525390625,
0.0526123046875,
0.032012939453125,
0.03399658203125,
-0.003265380859375,
-0.070556640625,
0.02197265625,
-0.034637451171875,
-0.0059967041015625,
-0.009033203125,
-0.062744140625,
0.08447265625,
-0.0233154296875,
-0.0210723876953125,
0.0163421630859375,
0.042144775390625,
0.0149078369140625,
0.020904541015625,
0.0305633544921875,
0.0305633544921875,
0.072509765625,
0.002780914306640625,
0.07379150390625,
-0.024688720703125,
0.01617431640625,
0.08984375,
-0.00786590576171875,
0.06390380859375,
0.027435302734375,
-0.019012451171875,
0.047027587890625,
0.062225341796875,
-0.0109100341796875,
0.03717041015625,
0.00926971435546875,
0.002109527587890625,
-0.0009813308715820312,
-0.006404876708984375,
-0.0240936279296875,
0.038665771484375,
0.0205230712890625,
-0.020477294921875,
-0.0004870891571044922,
0.01175689697265625,
0.015380859375,
-0.004344940185546875,
-0.0032825469970703125,
0.05810546875,
-0.0030364990234375,
-0.060333251953125,
0.0518798828125,
0.0181732177734375,
0.05914306640625,
-0.05316162109375,
-0.015960693359375,
-0.0210113525390625,
-0.0052337646484375,
-0.0209503173828125,
-0.0584716796875,
0.01442718505859375,
0.0037136077880859375,
-0.026123046875,
0.0008330345153808594,
0.04278564453125,
-0.02935791015625,
-0.021728515625,
0.01100921630859375,
0.01947021484375,
0.050323486328125,
0.004039764404296875,
-0.056396484375,
0.01763916015625,
0.00534820556640625,
-0.022735595703125,
0.01352691650390625,
0.0240631103515625,
-0.005489349365234375,
0.05975341796875,
0.054229736328125,
-0.003063201904296875,
-0.00276947021484375,
-0.005710601806640625,
0.0677490234375,
-0.052734375,
-0.047393798828125,
-0.06414794921875,
0.05426025390625,
-0.0097808837890625,
-0.035400390625,
0.051727294921875,
0.0426025390625,
0.056060791015625,
0.0020427703857421875,
0.05810546875,
-0.020599365234375,
0.0272979736328125,
-0.0384521484375,
0.055938720703125,
-0.03839111328125,
0.0203704833984375,
-0.01392364501953125,
-0.050048828125,
-0.0002579689025878906,
0.068115234375,
-0.0107574462890625,
-0.007381439208984375,
0.052825927734375,
0.0733642578125,
0.0017490386962890625,
0.004894256591796875,
0.006053924560546875,
0.01366424560546875,
0.0240631103515625,
0.046112060546875,
0.06488037109375,
-0.047454833984375,
0.041168212890625,
-0.0282745361328125,
-0.03216552734375,
-0.0033550262451171875,
-0.056243896484375,
-0.07354736328125,
-0.046600341796875,
-0.008697509765625,
-0.03228759765625,
0.00978851318359375,
0.09344482421875,
0.0487060546875,
-0.059051513671875,
-0.03521728515625,
0.006687164306640625,
0.00504302978515625,
-0.02349853515625,
-0.0235595703125,
0.01442718505859375,
-0.0207977294921875,
-0.053131103515625,
0.036102294921875,
-0.006092071533203125,
0.0116424560546875,
-0.01898193359375,
-0.0015106201171875,
-0.0318603515625,
0.005474090576171875,
0.036865234375,
0.024932861328125,
-0.05126953125,
-0.0232086181640625,
0.01277923583984375,
-0.01131439208984375,
0.0005154609680175781,
0.0394287109375,
-0.056976318359375,
0.0291748046875,
0.042083740234375,
0.02374267578125,
0.026214599609375,
0.00628662109375,
0.030242919921875,
-0.04962158203125,
0.008514404296875,
0.006748199462890625,
0.0233154296875,
0.01922607421875,
-0.05804443359375,
0.0322265625,
0.02197265625,
-0.04559326171875,
-0.0709228515625,
0.0027256011962890625,
-0.0728759765625,
-0.02935791015625,
0.0926513671875,
-0.0140228271484375,
-0.02435302734375,
-0.0131072998046875,
-0.04241943359375,
0.0175018310546875,
-0.053131103515625,
0.052001953125,
0.052978515625,
-0.0279998779296875,
-0.00878143310546875,
-0.03668212890625,
0.03363037109375,
0.003692626953125,
-0.06689453125,
0.0021114349365234375,
0.042236328125,
0.0400390625,
0.0243682861328125,
0.070068359375,
0.0157470703125,
0.023956298828125,
0.001049041748046875,
0.006687164306640625,
-0.0203094482421875,
-0.03240966796875,
-0.0228424072265625,
0.003673553466796875,
-0.01800537109375,
-0.0282440185546875
]
] |
hfl/chinese-bert-wwm | 2021-05-19T19:07:49.000Z | [
"transformers",
"pytorch",
"tf",
"jax",
"bert",
"fill-mask",
"zh",
"arxiv:1906.08101",
"arxiv:2004.13922",
"license:apache-2.0",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] | fill-mask | hfl | null | null | hfl/chinese-bert-wwm | 42 | 5,941 | transformers | 2022-03-02T23:29:05 | ---
language:
- zh
license: "apache-2.0"
---
## Chinese BERT with Whole Word Masking
For further accelerating Chinese natural language processing, we provide **Chinese pre-trained BERT with Whole Word Masking**.
**[Pre-Training with Whole Word Masking for Chinese BERT](https://arxiv.org/abs/1906.08101)**
Yiming Cui, Wanxiang Che, Ting Liu, Bing Qin, Ziqing Yang, Shijin Wang, Guoping Hu
This repository is developed based on：https://github.com/google-research/bert
You may also interested in,
- Chinese BERT series: https://github.com/ymcui/Chinese-BERT-wwm
- Chinese MacBERT: https://github.com/ymcui/MacBERT
- Chinese ELECTRA: https://github.com/ymcui/Chinese-ELECTRA
- Chinese XLNet: https://github.com/ymcui/Chinese-XLNet
- Knowledge Distillation Toolkit - TextBrewer: https://github.com/airaria/TextBrewer
More resources by HFL: https://github.com/ymcui/HFL-Anthology
## Citation
If you find the technical report or resource is useful, please cite the following technical report in your paper.
- Primary: https://arxiv.org/abs/2004.13922
```
@inproceedings{cui-etal-2020-revisiting,
title = "Revisiting Pre-Trained Models for {C}hinese Natural Language Processing",
author = "Cui, Yiming and
Che, Wanxiang and
Liu, Ting and
Qin, Bing and
Wang, Shijin and
Hu, Guoping",
booktitle = "Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing: Findings",
month = nov,
year = "2020",
address = "Online",
publisher = "Association for Computational Linguistics",
url = "https://www.aclweb.org/anthology/2020.findings-emnlp.58",
pages = "657--668",
}
```
- Secondary: https://arxiv.org/abs/1906.08101
```
@article{chinese-bert-wwm,
title={Pre-Training with Whole Word Masking for Chinese BERT},
author={Cui, Yiming and Che, Wanxiang and Liu, Ting and Qin, Bing and Yang, Ziqing and Wang, Shijin and Hu, Guoping},
journal={arXiv preprint arXiv:1906.08101},
year={2019}
}
``` | 1,993 | [
[
-0.0306243896484375,
-0.054473876953125,
0.0236053466796875,
0.042236328125,
-0.0304107666015625,
-0.00914764404296875,
-0.042999267578125,
-0.053985595703125,
0.0261077880859375,
0.034332275390625,
-0.03448486328125,
-0.034637451171875,
-0.042755126953125,
-0.0005488395690917969,
-0.0067596435546875,
0.061859130859375,
0.0009260177612304688,
0.0146026611328125,
0.0203704833984375,
0.007183074951171875,
-0.005023956298828125,
-0.0540771484375,
-0.052276611328125,
-0.03265380859375,
0.034698486328125,
-0.000019669532775878906,
0.0239410400390625,
0.03857421875,
0.020111083984375,
0.0260772705078125,
-0.00678253173828125,
0.0144805908203125,
-0.01611328125,
-0.0025482177734375,
0.01959228515625,
-0.00391387939453125,
-0.03753662109375,
0.00754547119140625,
0.054046630859375,
0.043609619140625,
0.00829315185546875,
-0.0065460205078125,
0.00888824462890625,
0.04833984375,
-0.0445556640625,
0.002429962158203125,
-0.055877685546875,
0.00806427001953125,
-0.035797119140625,
0.004093170166015625,
-0.031951904296875,
-0.0230712890625,
0.0203857421875,
-0.054046630859375,
0.014190673828125,
0.0030040740966796875,
0.10888671875,
-0.00986480712890625,
-0.000362396240234375,
-0.002105712890625,
-0.0304107666015625,
0.051544189453125,
-0.082763671875,
0.03326416015625,
0.03863525390625,
-0.00966644287109375,
-0.01611328125,
-0.080810546875,
-0.0631103515625,
-0.0251617431640625,
-0.008087158203125,
0.0176849365234375,
-0.0005679130554199219,
0.02972412109375,
0.01453399658203125,
0.030548095703125,
-0.04864501953125,
0.018341064453125,
-0.0184478759765625,
-0.044525146484375,
0.04595947265625,
-0.024017333984375,
0.0229949951171875,
-0.005321502685546875,
-0.0254669189453125,
-0.038055419921875,
-0.0266876220703125,
0.016082763671875,
0.026092529296875,
0.016265869140625,
-0.0002391338348388672,
0.0117340087890625,
-0.0037631988525390625,
0.050506591796875,
-0.005977630615234375,
0.00914764404296875,
0.044830322265625,
-0.034393310546875,
-0.0256500244140625,
0.010040283203125,
0.07379150390625,
-0.005847930908203125,
0.0183563232421875,
-0.0022068023681640625,
-0.0234527587890625,
-0.02288818359375,
0.0013494491577148438,
-0.057281494140625,
-0.03155517578125,
0.01044464111328125,
-0.03643798828125,
-0.0021820068359375,
0.016204833984375,
-0.043212890625,
-0.0203704833984375,
-0.0162353515625,
0.041534423828125,
-0.038970947265625,
-0.02288818359375,
0.0168914794921875,
-0.017303466796875,
0.03399658203125,
0.0030384063720703125,
-0.057464599609375,
0.00803375244140625,
0.037811279296875,
0.05194091796875,
0.004337310791015625,
-0.03094482421875,
-0.027862548828125,
-0.00647735595703125,
-0.01146697998046875,
0.053985595703125,
-0.0214996337890625,
-0.0018453598022460938,
0.016082763671875,
0.00849151611328125,
-0.006374359130859375,
-0.0161285400390625,
0.0628662109375,
-0.033233642578125,
0.027557373046875,
-0.0202484130859375,
-0.02838134765625,
-0.0217132568359375,
0.0143280029296875,
-0.04559326171875,
0.0877685546875,
-0.0256195068359375,
-0.05902099609375,
0.0137939453125,
-0.0640869140625,
-0.049957275390625,
0.0014123916625976562,
0.01654052734375,
-0.038726806640625,
-0.0174560546875,
0.03033447265625,
0.02655029296875,
-0.00746917724609375,
0.01190948486328125,
-0.01079559326171875,
-0.0276641845703125,
0.0018558502197265625,
-0.0258941650390625,
0.0888671875,
0.0204925537109375,
-0.0338134765625,
0.0251617431640625,
-0.06488037109375,
0.0096435546875,
0.01092529296875,
-0.0165252685546875,
-0.021514892578125,
-0.0042877197265625,
0.0178985595703125,
0.0197906494140625,
0.047271728515625,
-0.059326171875,
-0.00559234619140625,
-0.048858642578125,
0.032470703125,
0.06109619140625,
-0.0223541259765625,
0.0178680419921875,
-0.019378662109375,
0.013214111328125,
-0.0005230903625488281,
0.004619598388671875,
-0.029266357421875,
-0.039276123046875,
-0.072265625,
-0.021820068359375,
0.045501708984375,
0.0467529296875,
-0.059112548828125,
0.067138671875,
-0.0177154541015625,
-0.042572021484375,
-0.056488037109375,
-0.00577545166015625,
0.035125732421875,
0.03143310546875,
0.038848876953125,
-0.0232696533203125,
-0.0667724609375,
-0.0555419921875,
-0.01461029052734375,
-0.01849365234375,
-0.00975799560546875,
-0.0017080307006835938,
0.0152435302734375,
-0.01152801513671875,
0.05718994140625,
-0.04541015625,
-0.036529541015625,
-0.0157470703125,
0.04022216796875,
0.0161590576171875,
0.037017822265625,
0.035125732421875,
-0.050323486328125,
-0.04541015625,
-0.01038360595703125,
-0.022216796875,
-0.0183258056640625,
-0.015380859375,
-0.034423828125,
0.0338134765625,
0.03656005859375,
-0.026092529296875,
0.034393310546875,
0.0271148681640625,
-0.00811004638671875,
0.04779052734375,
-0.031646728515625,
-0.0042877197265625,
-0.0740966796875,
0.01442718505859375,
0.007198333740234375,
-0.0013217926025390625,
-0.06060791015625,
-0.0031948089599609375,
0.001659393310546875,
0.01251220703125,
-0.0280303955078125,
0.03814697265625,
-0.049652099609375,
0.0191802978515625,
-0.019287109375,
0.0252532958984375,
0.005096435546875,
0.06634521484375,
0.0280303955078125,
0.04449462890625,
0.037933349609375,
-0.053802490234375,
0.0117340087890625,
0.006805419921875,
-0.027008056640625,
-0.0200347900390625,
-0.057373046875,
0.0150146484375,
-0.00337982177734375,
0.03106689453125,
-0.0865478515625,
0.01009368896484375,
0.03338623046875,
-0.051361083984375,
0.0286712646484375,
0.0279693603515625,
-0.056854248046875,
-0.029815673828125,
-0.061309814453125,
0.01090240478515625,
0.035552978515625,
-0.036163330078125,
0.0188751220703125,
0.021209716796875,
0.00027680397033691406,
-0.045654296875,
-0.06390380859375,
0.01593017578125,
0.0157623291015625,
-0.051788330078125,
0.061187744140625,
-0.02490234375,
0.01258087158203125,
0.0002777576446533203,
0.0076751708984375,
-0.031005859375,
0.005481719970703125,
-0.011688232421875,
0.0269317626953125,
-0.0204925537109375,
0.01507568359375,
-0.005279541015625,
0.01039886474609375,
0.0021190643310546875,
-0.025115966796875,
0.04638671875,
0.007450103759765625,
-0.01251220703125,
-0.0272064208984375,
0.013641357421875,
0.0171051025390625,
-0.0289764404296875,
0.06158447265625,
0.08734130859375,
-0.047637939453125,
0.007518768310546875,
-0.04949951171875,
-0.0115966796875,
-0.0350341796875,
0.0307159423828125,
-0.00450897216796875,
-0.06982421875,
0.0328369140625,
0.0300445556640625,
0.0341796875,
0.048431396484375,
0.033599853515625,
-0.006374359130859375,
0.05426025390625,
0.048858642578125,
-0.02593994140625,
0.06390380859375,
-0.0005583763122558594,
0.036468505859375,
-0.0706787109375,
0.00563812255859375,
-0.04656982421875,
-0.019073486328125,
-0.051788330078125,
-0.01383209228515625,
-0.00011491775512695312,
0.005054473876953125,
-0.02069091796875,
0.036376953125,
-0.0604248046875,
0.0179595947265625,
0.05615234375,
0.006008148193359375,
0.006580352783203125,
0.004497528076171875,
-0.025238037109375,
-0.00490570068359375,
-0.02862548828125,
-0.027130126953125,
0.068359375,
0.0230560302734375,
0.0190582275390625,
-0.0157470703125,
0.056671142578125,
0.013519287109375,
0.01416778564453125,
-0.042388916015625,
0.049285888671875,
-0.0254364013671875,
-0.0469970703125,
-0.041229248046875,
-0.0176239013671875,
-0.08721923828125,
0.036956787109375,
-0.0231781005859375,
-0.05926513671875,
0.00736236572265625,
-0.0018606185913085938,
-0.0282440185546875,
0.035858154296875,
-0.05572509765625,
0.041168212890625,
-0.01366424560546875,
-0.00727081298828125,
0.0018320083618164062,
-0.058837890625,
0.035980224609375,
-0.01297760009765625,
0.0034656524658203125,
-0.00016033649444580078,
0.016326904296875,
0.0787353515625,
-0.032257080078125,
0.06451416015625,
-0.0204620361328125,
-0.0228271484375,
0.0264129638671875,
-0.03814697265625,
0.0267181396484375,
-0.0215606689453125,
-0.006256103515625,
0.03607177734375,
-0.007457733154296875,
-0.02459716796875,
-0.01357269287109375,
0.0406494140625,
-0.0640869140625,
-0.04498291015625,
-0.052642822265625,
-0.021026611328125,
0.0011816024780273438,
0.038421630859375,
0.039703369140625,
0.01239776611328125,
0.0022296905517578125,
0.007480621337890625,
0.05712890625,
-0.035308837890625,
0.05780029296875,
0.040771484375,
-0.003101348876953125,
-0.032867431640625,
0.06494140625,
0.024566650390625,
-0.0019207000732421875,
0.052276611328125,
0.011138916015625,
-0.017242431640625,
-0.04669189453125,
-0.011444091796875,
0.0253753662109375,
-0.0343017578125,
-0.004688262939453125,
-0.05767822265625,
-0.057769775390625,
-0.06298828125,
0.0096893310546875,
-0.004619598388671875,
-0.028472900390625,
-0.038055419921875,
0.003528594970703125,
0.012054443359375,
0.020477294921875,
-0.0199737548828125,
0.02490234375,
-0.065673828125,
0.03265380859375,
0.0219268798828125,
0.021942138671875,
0.02227783203125,
-0.049041748046875,
-0.042572021484375,
0.0263519287109375,
-0.03338623046875,
-0.04058837890625,
0.037872314453125,
0.02130126953125,
0.060089111328125,
0.028045654296875,
0.0255279541015625,
0.051055908203125,
-0.041595458984375,
0.08428955078125,
0.0159759521484375,
-0.07598876953125,
0.0310516357421875,
-0.00324249267578125,
0.029388427734375,
0.033294677734375,
0.01105499267578125,
-0.04901123046875,
-0.0195159912109375,
-0.03460693359375,
-0.076904296875,
0.06866455078125,
0.0167388916015625,
0.01384735107421875,
0.00038361549377441406,
0.008087158203125,
-0.004932403564453125,
0.0003261566162109375,
-0.0897216796875,
-0.036956787109375,
-0.0279693603515625,
-0.002040863037109375,
0.010040283203125,
-0.035247802734375,
0.0154266357421875,
-0.036041259765625,
0.07550048828125,
0.015228271484375,
0.041259765625,
0.042999267578125,
-0.023040771484375,
-0.00011897087097167969,
0.018798828125,
0.061309814453125,
0.03253173828125,
-0.0278778076171875,
0.0007696151733398438,
0.004974365234375,
-0.06158447265625,
-0.02093505859375,
0.036163330078125,
0.0059967041015625,
0.02044677734375,
0.04718017578125,
0.05926513671875,
0.009002685546875,
-0.03936767578125,
0.04119873046875,
-0.01261138916015625,
-0.037811279296875,
-0.0295562744140625,
-0.012359619140625,
-0.0007314682006835938,
-0.00908660888671875,
0.040771484375,
-0.01470947265625,
0.0007386207580566406,
-0.0247344970703125,
0.0027065277099609375,
0.0175933837890625,
-0.036773681640625,
-0.029815673828125,
0.03851318359375,
0.0170440673828125,
-0.01113128662109375,
0.052398681640625,
-0.0250244140625,
-0.07000732421875,
0.035736083984375,
0.042388916015625,
0.09515380859375,
-0.006107330322265625,
-0.005527496337890625,
0.04425048828125,
0.04339599609375,
0.004909515380859375,
0.0198974609375,
-0.0215911865234375,
-0.07501220703125,
-0.044921875,
-0.045196533203125,
-0.0155792236328125,
0.029510498046875,
-0.030029296875,
0.011322021484375,
-0.034637451171875,
-0.0011167526245117188,
0.0018100738525390625,
0.0189971923828125,
-0.03985595703125,
0.01226043701171875,
0.038970947265625,
0.06964111328125,
-0.039794921875,
0.096435546875,
0.06695556640625,
-0.031036376953125,
-0.04620361328125,
0.0229034423828125,
-0.0288543701171875,
-0.05572509765625,
0.057891845703125,
0.0123138427734375,
-0.0092926025390625,
-0.0096893310546875,
-0.052825927734375,
-0.050323486328125,
0.07061767578125,
0.00888824462890625,
-0.036468505859375,
-0.0012569427490234375,
-0.0036945343017578125,
0.047637939453125,
0.0022144317626953125,
0.0198974609375,
0.02996826171875,
0.055908203125,
-0.006862640380859375,
-0.07489013671875,
-0.00029158592224121094,
-0.030914306640625,
0.004364013671875,
-0.001499176025390625,
-0.05389404296875,
0.06842041015625,
0.003139495849609375,
-0.02490234375,
0.0014543533325195312,
0.06329345703125,
0.0159759521484375,
0.0231170654296875,
0.04620361328125,
0.0467529296875,
0.059722900390625,
-0.010955810546875,
0.036102294921875,
-0.01861572265625,
0.0034236907958984375,
0.0716552734375,
-0.005405426025390625,
0.05126953125,
0.0159454345703125,
-0.0254364013671875,
0.043701171875,
0.06121826171875,
0.00823974609375,
0.040374755859375,
0.023406982421875,
-0.01044464111328125,
-0.01120758056640625,
-0.00904083251953125,
-0.04827880859375,
0.0162353515625,
0.015655517578125,
-0.018585205078125,
0.00042557716369628906,
-0.00405120849609375,
0.0258636474609375,
0.0025272369384765625,
-0.00893402099609375,
0.049102783203125,
0.0083465576171875,
-0.0433349609375,
0.049652099609375,
0.01355743408203125,
0.09637451171875,
-0.07318115234375,
-0.00600433349609375,
-0.00958251953125,
-0.016082763671875,
-0.00814056396484375,
-0.04071044921875,
-0.0012416839599609375,
-0.0081787109375,
-0.0170135498046875,
-0.0178375244140625,
0.057464599609375,
-0.035430908203125,
-0.028228759765625,
0.0328369140625,
0.01092529296875,
0.01236724853515625,
-0.0031642913818359375,
-0.043701171875,
-0.0006761550903320312,
0.02166748046875,
-0.031951904296875,
0.027923583984375,
0.046295166015625,
0.0008692741394042969,
0.0252227783203125,
0.07452392578125,
0.02862548828125,
0.020294189453125,
0.01396942138671875,
0.06292724609375,
-0.053375244140625,
-0.038543701171875,
-0.0625,
0.0255889892578125,
-0.0193634033203125,
-0.0273895263671875,
0.05322265625,
0.0235748291015625,
0.07440185546875,
0.003978729248046875,
0.056884765625,
-0.00821685791015625,
0.0305633544921875,
-0.032012939453125,
0.080322265625,
-0.037872314453125,
0.01117706298828125,
-0.0341796875,
-0.06597900390625,
-0.036895751953125,
0.04559326171875,
-0.00737762451171875,
-0.0020542144775390625,
0.050872802734375,
0.039154052734375,
0.029083251953125,
-0.009307861328125,
0.0251312255859375,
0.0267791748046875,
0.036590576171875,
0.0282135009765625,
0.027740478515625,
-0.044219970703125,
0.05023193359375,
-0.0296630859375,
-0.013916015625,
-0.00963592529296875,
-0.07708740234375,
-0.055938720703125,
-0.063232421875,
-0.0229034423828125,
-0.00415802001953125,
-0.00569915771484375,
0.064697265625,
0.0546875,
-0.0701904296875,
-0.01078033447265625,
0.0034160614013671875,
0.0146026611328125,
-0.0254974365234375,
-0.0181427001953125,
0.05426025390625,
-0.045257568359375,
-0.053985595703125,
0.0079803466796875,
0.0029773712158203125,
-0.0023784637451171875,
-0.00911712646484375,
0.0022983551025390625,
-0.052520751953125,
0.01026153564453125,
0.04571533203125,
0.0274200439453125,
-0.055694580078125,
-0.0157470703125,
-0.0042724609375,
-0.01123046875,
0.005268096923828125,
0.046173095703125,
-0.05194091796875,
0.040985107421875,
0.04901123046875,
0.0504150390625,
0.033172607421875,
-0.02081298828125,
0.02978515625,
-0.06085205078125,
0.0166778564453125,
0.0099945068359375,
0.0298614501953125,
0.0207061767578125,
-0.0305023193359375,
0.0306549072265625,
0.010986328125,
-0.0406494140625,
-0.0604248046875,
-0.01342010498046875,
-0.06854248046875,
-0.0251312255859375,
0.061798095703125,
-0.0272369384765625,
-0.01186370849609375,
-0.004337310791015625,
-0.031158447265625,
0.04925537109375,
-0.03277587890625,
0.0523681640625,
0.0841064453125,
0.0096588134765625,
-0.0152740478515625,
-0.017547607421875,
0.033172607421875,
0.034210205078125,
-0.0457763671875,
0.0036563873291015625,
0.0105743408203125,
-0.008697509765625,
0.023040771484375,
0.05194091796875,
-0.0011758804321289062,
0.0025386810302734375,
-0.0177764892578125,
0.04388427734375,
-0.01107025146484375,
0.00001043081283569336,
-0.0198822021484375,
-0.0194854736328125,
-0.00534820556640625,
-0.04266357421875
]
] |
JosephusCheung/Guanaco | 2023-05-29T12:48:21.000Z | [
"transformers",
"pytorch",
"llama",
"text-generation",
"guannaco",
"alpaca",
"conversational",
"en",
"zh",
"ja",
"de",
"dataset:JosephusCheung/GuanacoDataset",
"doi:10.57967/hf/0607",
"license:gpl-3.0",
"has_space",
"text-generation-inference",
"region:us"
] | conversational | JosephusCheung | null | null | JosephusCheung/Guanaco | 214 | 5,941 | transformers | 2023-04-08T03:03:14 | ---
inference: false
license: gpl-3.0
datasets:
- JosephusCheung/GuanacoDataset
language:
- en
- zh
- ja
- de
pipeline_tag: conversational
tags:
- llama
- guannaco
- alpaca
---
**You can run on Colab free T4 GPU now**
[](https://colab.research.google.com/drive/1ocSmoy3ba1EkYu7JWT1oCw9vz8qC2cMk#scrollTo=zLORi5OcPcIJ)
**It is highly recommended to use fp16 inference for this model, as 8-bit precision may significantly affect performance. If you require a more Consumer Hardware friendly version, please use the specialized quantized, only 5+GB V-Ram required** [JosephusCheung/GuanacoOnConsumerHardware](https://huggingface.co/JosephusCheung/GuanacoOnConsumerHardware).
**You are encouraged to use the latest version of transformers from GitHub.**
Guanaco is an advanced instruction-following language model built on Meta's LLaMA 7B model. Expanding upon the initial 52K dataset from the Alpaca model, an additional 534K+ entries have been incorporated, covering English, Simplified Chinese, Traditional Chinese (Taiwan), Traditional Chinese (Hong Kong), Japanese, Deutsch, and various linguistic and grammatical tasks. This wealth of data enables Guanaco to perform exceptionally well in multilingual environments.
In an effort to foster openness and replicability in research, we have made the Guanaco Dataset publicly accessible and we have released the model weights here. By providing these resources, we aim to inspire more researchers to pursue related research and collectively advance the development of instruction-following language models.
[KBlueLeaf](https://huggingface.co/KBlueLeaf)’s invaluable contributions to the conceptual validation, [trained model](https://huggingface.co/KBlueLeaf/guanaco-7B-leh) and [inference development](https://github.com/KohakuBlueleaf/guanaco-lora) of the model would be gratefully acknowledged, without whose efforts the project shall never have come to fruition.
When utilizing the Guanaco model, please bear in mind the following points:
The Guanaco model has not been filtered for harmful, biased, or explicit content. As a result, outputs that do not adhere to ethical norms may be generated during use. Please exercise caution when using the model in research or practical applications.
1. ### Improved context and prompt role support:
The new format is designed to be similar to ChatGPT, allowing for better integration with the Alpaca format and enhancing the overall user experience.
Instruction is utilized as a few-shot context to support diverse inputs and responses, making it easier for the model to understand and provide accurate responses to user queries.
The format is as follows:
```
### Instruction:
User: History User Input
Assistant: History Assistant Answer
### Input:
System: Knowledge
User: New User Input
### Response:
New Assistant Answer
```
This structured format allows for easier tracking of the conversation history and maintaining context throughout a multi-turn dialogue.
3. ### Role-playing support:
Guanaco now offers advanced role-playing support, similar to Character.AI, in English, Simplified Chinese, Traditional Chinese, Japanese, and Deutsch, making it more versatile for users from different linguistic backgrounds.
Users can instruct the model to assume specific roles, historical figures, or fictional characters, as well as personalities based on their input. This allows for more engaging and immersive conversations.
The model can use various sources of information to provide knowledge and context for the character's background and behavior, such as encyclopedic entries, first-person narrations, or a list of personality traits.
The model will consistently output responses in the format "Character Name: Reply" to maintain the chosen role throughout the conversation, enhancing the user's experience.
4. ### Rejection of answers and avoidance of erroneous responses:
The model has been updated to handle situations where it lacks sufficient knowledge or is unable to provide a valid response more effectively.
Reserved keywords have been introduced to indicate different scenarios and provide clearer communication with the user, use in System Prompt:
NO IDEA: Indicates that the model lacks the necessary knowledge to provide an accurate answer, and will explain this to the user, encouraging them to seek alternative sources.
FORBIDDEN: Indicates that the model refuses to answer due to specific reasons (e.g., legal, ethical, or safety concerns), which will be inferred based on the context of the query.
SFW: Indicates that the model refuses to answer a question because it has been filtered for NSFW content, ensuring a safer and more appropriate user experience.
6. ### Continuation of responses for ongoing topics:
The Guanaco model can now continue answering questions or discussing topics upon the user's request, making it more adaptable and better suited for extended conversations.
The contextual structure consisting of System, Assistant, and User roles allows the model to engage in multi-turn dialogues, maintain context-aware conversations, and provide more coherent responses.
The model can now accommodate role specification and character settings, providing a more immersive and tailored conversational experience based on the user's preferences.
It is important to remember that Guanaco is a 7B-parameter model, and **any knowledge-based content should be considered potentially inaccurate**. We strongly recommend **providing verifiable sources in System Prompt, such as Wikipedia, for knowledge-based answers**. In the absence of sources, it is crucial to inform users of this limitation to prevent the dissemination of false information and to maintain transparency.
Due to the differences in the format between this project and [Stanford Alpaca](https://github.com/tatsu-lab/stanford_alpaca), please refer to *Guanaco-lora: LoRA for training Multilingual Instruction-following LM based on LLaMA* (https://github.com/KohakuBlueleaf/guanaco-lora) for further training and inference our models.
## Recent News
We've noticed a recent entrant in the field, the QLoRa method, which we find concerning due to its attempt to piggyback on the reputation of Guanaco. We strongly disapprove of such practices. QLoRa, as far as we can tell, lacks mathematical robustness and its performance significantly trails behind that of GPTQ and advancements such as PEFT fine-tuning, which have been successful in improving upon it.
Guanaco has been diligent, consistently releasing multilingual datasets since March 2023, along with publishing weights that are not only an enhanced version of GPTQ but also support multimodal VQA and have been optimized for 4-bit. Despite the substantial financial investment of tens of thousands of dollars in distilling data from OpenAI's GPT models, we still consider these efforts to be incremental.
We, however, aim to move beyond the incremental:
1. We strive to no longer rely on distillation data from OpenAI: We've found that relying on GPT-generated data impedes significant breakthroughs. Furthermore, this approach has proven to be disastrous when dealing with the imbalances in multilingual tasks.
2. We're focusing on the enhancement of quantization structure and partial native 4-bit fine-tuning: We are deeply appreciative of the GPTQ-Llama project for paving the way in state-of-the-art LLM quantization. Its unique qualities, especially at the 7B size, are facilitating significant progress in multilingual and multimodal tasks.
3. We plan to utilize visual data to adjust our language models: We believe this will fundamentally address the issues of language imbalance, translation inaccuracies, and the lack of graphical logic in LLM.
While our work is still in the early stages, we're determined to break new ground in these areas. Our critique of QLoRa's practices does not stem from animosity but rather from the fundamental belief that innovation should be rooted in originality, integrity, and substantial progress.
| 8,232 | [
[
-0.0178375244140625,
-0.069091796875,
0.033172607421875,
0.0295257568359375,
-0.01507568359375,
-0.01311492919921875,
-0.0122222900390625,
-0.048797607421875,
0.0051116943359375,
0.033294677734375,
-0.0287628173828125,
-0.04730224609375,
-0.03948974609375,
-0.006500244140625,
-0.022247314453125,
0.0946044921875,
0.01009368896484375,
-0.0047607421875,
-0.004596710205078125,
-0.0115203857421875,
-0.059600830078125,
-0.040496826171875,
-0.072021484375,
-0.0172882080078125,
0.064697265625,
0.0218963623046875,
0.0592041015625,
0.04730224609375,
0.01496124267578125,
0.0189361572265625,
-0.007328033447265625,
0.0190887451171875,
-0.024505615234375,
-0.01474761962890625,
0.0076446533203125,
-0.036468505859375,
-0.061065673828125,
-0.00240325927734375,
0.031341552734375,
0.0251007080078125,
-0.035400390625,
-0.0005402565002441406,
0.007244110107421875,
0.0234832763671875,
-0.044921875,
0.047088623046875,
-0.032135009765625,
-0.00959014892578125,
-0.01082611083984375,
0.0013523101806640625,
-0.030242919921875,
-0.04315185546875,
-0.00543212890625,
-0.05633544921875,
0.006412506103515625,
-0.012054443359375,
0.078125,
0.00453948974609375,
-0.037506103515625,
-0.0278778076171875,
-0.05322265625,
0.04302978515625,
-0.0643310546875,
0.01739501953125,
0.04718017578125,
0.0389404296875,
-0.00677490234375,
-0.059600830078125,
-0.0511474609375,
-0.031585693359375,
-0.00618743896484375,
0.0141448974609375,
-0.02203369140625,
0.007549285888671875,
0.0206298828125,
0.0310211181640625,
-0.03778076171875,
0.01023101806640625,
-0.04547119140625,
-0.0003705024719238281,
0.043121337890625,
0.0185394287109375,
0.029266357421875,
-0.0086669921875,
-0.02386474609375,
-0.0224609375,
-0.0650634765625,
0.00936126708984375,
0.0426025390625,
0.0216522216796875,
-0.024139404296875,
0.03143310546875,
-0.019287109375,
0.041412353515625,
-0.019134521484375,
-0.0189666748046875,
0.009368896484375,
-0.0170745849609375,
-0.03985595703125,
-0.0185546875,
0.06463623046875,
0.0166168212890625,
0.003780364990234375,
0.00433349609375,
-0.0007762908935546875,
0.006809234619140625,
0.00885009765625,
-0.044891357421875,
-0.0218963623046875,
0.023406982421875,
-0.01318359375,
-0.020263671875,
-0.0017919540405273438,
-0.05279541015625,
-0.01419830322265625,
-0.0031757354736328125,
0.0286712646484375,
-0.0308685302734375,
-0.01145172119140625,
0.0028438568115234375,
0.005054473876953125,
0.0264739990234375,
0.048797607421875,
-0.08843994140625,
0.0198516845703125,
0.0306396484375,
0.06787109375,
0.003879547119140625,
-0.027984619140625,
-0.022613525390625,
0.00827789306640625,
-0.0171966552734375,
0.0634765625,
-0.027313232421875,
-0.042205810546875,
-0.004119873046875,
0.019744873046875,
-0.014251708984375,
-0.0283660888671875,
0.04669189453125,
-0.02984619140625,
0.034515380859375,
-0.037261962890625,
-0.01026153564453125,
-0.023590087890625,
0.013580322265625,
-0.05291748046875,
0.07757568359375,
0.0185699462890625,
-0.030517578125,
0.00024116039276123047,
-0.05462646484375,
-0.0177459716796875,
0.0007281303405761719,
-0.0009412765502929688,
-0.0145416259765625,
-0.0139007568359375,
0.0281219482421875,
0.01297760009765625,
-0.04058837890625,
0.02655029296875,
-0.0165252685546875,
-0.0335693359375,
0.0083160400390625,
-0.040679931640625,
0.09075927734375,
0.0259246826171875,
-0.022216796875,
0.00917816162109375,
-0.038360595703125,
0.007511138916015625,
0.01800537109375,
-0.02093505859375,
-0.0142822265625,
-0.01195526123046875,
-0.00205230712890625,
0.0200042724609375,
0.0176849365234375,
-0.0172882080078125,
0.032470703125,
-0.007457733154296875,
0.05291748046875,
0.05767822265625,
-0.006801605224609375,
0.020111083984375,
-0.020477294921875,
0.06591796875,
-0.00830841064453125,
0.0511474609375,
-0.0214996337890625,
-0.062744140625,
-0.06829833984375,
-0.0181121826171875,
0.02435302734375,
0.048553466796875,
-0.041107177734375,
0.01751708984375,
-0.01380157470703125,
-0.0426025390625,
-0.036376953125,
-0.00493621826171875,
0.034454345703125,
0.0482177734375,
0.03973388671875,
-0.02447509765625,
-0.02880859375,
-0.057342529296875,
0.016326904296875,
-0.03253173828125,
-0.0037059783935546875,
0.0159912109375,
0.035491943359375,
-0.01953125,
0.047027587890625,
-0.05035400390625,
-0.0214996337890625,
-0.0164031982421875,
-0.006130218505859375,
0.004058837890625,
0.029754638671875,
0.0399169921875,
-0.052734375,
-0.038177490234375,
0.0269622802734375,
-0.0684814453125,
-0.0017375946044921875,
0.01195526123046875,
-0.0159149169921875,
0.0200042724609375,
0.0222930908203125,
-0.06304931640625,
0.04193115234375,
0.0517578125,
-0.01119232177734375,
0.03826904296875,
-0.017791748046875,
0.01509857177734375,
-0.07403564453125,
0.004528045654296875,
-0.0020599365234375,
0.001972198486328125,
-0.04461669921875,
0.0200042724609375,
0.0097808837890625,
-0.00045228004455566406,
-0.040740966796875,
0.047393798828125,
-0.0201873779296875,
0.008209228515625,
-0.019439697265625,
0.004169464111328125,
0.02276611328125,
0.056640625,
-0.01123046875,
0.0723876953125,
0.0187835693359375,
-0.050689697265625,
0.026763916015625,
0.0204620361328125,
-0.0197906494140625,
0.02764892578125,
-0.08209228515625,
0.0211181640625,
-0.00807952880859375,
0.0275726318359375,
-0.062347412109375,
-0.0209808349609375,
0.04541015625,
-0.03839111328125,
0.006092071533203125,
-0.0215911865234375,
-0.0278472900390625,
-0.025482177734375,
-0.01230621337890625,
0.0238800048828125,
0.0335693359375,
-0.032745361328125,
0.044464111328125,
0.026214599609375,
-0.00911712646484375,
-0.056732177734375,
-0.057525634765625,
0.005718231201171875,
-0.036834716796875,
-0.045989990234375,
0.022735595703125,
-0.0172882080078125,
-0.0182037353515625,
-0.01361846923828125,
0.00850677490234375,
-0.017791748046875,
-0.00019860267639160156,
0.01474761962890625,
0.040435791015625,
-0.004421234130859375,
0.008544921875,
0.01256561279296875,
0.0081939697265625,
-0.0022125244140625,
-0.0161895751953125,
0.05047607421875,
-0.0211181640625,
-0.004482269287109375,
-0.030303955078125,
0.023040771484375,
0.027252197265625,
-0.034912109375,
0.06170654296875,
0.044158935546875,
-0.032562255859375,
0.0156402587890625,
-0.04034423828125,
0.00824737548828125,
-0.036651611328125,
0.023345947265625,
-0.0171966552734375,
-0.051239013671875,
0.06488037109375,
0.04376220703125,
0.001781463623046875,
0.03936767578125,
0.04510498046875,
-0.005596160888671875,
0.0767822265625,
0.047027587890625,
-0.00110626220703125,
0.047149658203125,
-0.0256195068359375,
0.01416015625,
-0.0625,
-0.04644775390625,
-0.046142578125,
0.00850677490234375,
-0.051025390625,
-0.02587890625,
0.0242156982421875,
0.0005474090576171875,
-0.0199432373046875,
0.0389404296875,
-0.0214080810546875,
0.0273895263671875,
0.04534912109375,
0.0177764892578125,
0.037841796875,
-0.004817962646484375,
0.0213775634765625,
0.00847625732421875,
-0.05657958984375,
-0.04437255859375,
0.082763671875,
0.027923583984375,
0.059112548828125,
0.0169830322265625,
0.02911376953125,
0.03076171875,
0.026763916015625,
-0.0562744140625,
0.030303955078125,
0.0083160400390625,
-0.046600341796875,
-0.0202178955078125,
-0.0285186767578125,
-0.075439453125,
0.028717041015625,
0.0081634521484375,
-0.058929443359375,
0.02935791015625,
0.01248931884765625,
-0.0394287109375,
0.006450653076171875,
-0.08013916015625,
0.0775146484375,
-0.01488494873046875,
-0.027435302734375,
-0.00406646728515625,
-0.036285400390625,
0.03717041015625,
0.01490020751953125,
0.0018224716186523438,
-0.006961822509765625,
0.00798797607421875,
0.040374755859375,
-0.048431396484375,
0.069091796875,
-0.009979248046875,
-0.03558349609375,
0.049041748046875,
-0.0186309814453125,
0.042083740234375,
0.01499176025390625,
0.01000213623046875,
0.02691650390625,
0.0024089813232421875,
-0.039703369140625,
-0.052886962890625,
0.06689453125,
-0.058837890625,
-0.042816162109375,
-0.03973388671875,
-0.04669189453125,
-0.0023345947265625,
0.005184173583984375,
0.0281524658203125,
0.017181396484375,
-0.01059722900390625,
-0.0029010772705078125,
0.04644775390625,
-0.0231781005859375,
0.0256195068359375,
0.02496337890625,
-0.01482391357421875,
-0.0360107421875,
0.05743408203125,
0.0218658447265625,
0.0157928466796875,
0.0248260498046875,
0.0037403106689453125,
-0.027679443359375,
-0.031585693359375,
-0.03631591796875,
0.01506805419921875,
-0.0560302734375,
-0.017364501953125,
-0.05316162109375,
-0.0028228759765625,
-0.042572021484375,
0.02203369140625,
-0.0157012939453125,
-0.03765869140625,
-0.0219879150390625,
-0.0169677734375,
0.0256195068359375,
0.04388427734375,
0.0213775634765625,
0.049591064453125,
-0.03961181640625,
0.0091705322265625,
0.03802490234375,
0.0031795501708984375,
0.003055572509765625,
-0.05419921875,
-0.036468505859375,
0.019378662109375,
-0.044036865234375,
-0.057586669921875,
0.039337158203125,
0.004512786865234375,
0.031829833984375,
0.0305938720703125,
-0.01309967041015625,
0.0672607421875,
-0.0231781005859375,
0.061492919921875,
-0.0032787322998046875,
-0.0628662109375,
0.05584716796875,
-0.046600341796875,
0.0181884765625,
0.027099609375,
0.039276123046875,
-0.043426513671875,
-0.015960693359375,
-0.049835205078125,
-0.05743408203125,
0.044189453125,
0.01503753662109375,
0.0200653076171875,
-0.01348114013671875,
0.0175323486328125,
0.0028629302978515625,
0.0214385986328125,
-0.07177734375,
-0.01226043701171875,
-0.0341796875,
-0.0027065277099609375,
0.0016345977783203125,
-0.01335906982421875,
-0.01483917236328125,
-0.0257110595703125,
0.04876708984375,
-0.01529693603515625,
0.04168701171875,
0.0012617111206054688,
-0.008758544921875,
0.02117919921875,
0.0196380615234375,
0.058929443359375,
0.0596923828125,
-0.019012451171875,
-0.01654052734375,
0.01323699951171875,
-0.044921875,
-0.0005812644958496094,
0.01593017578125,
-0.0245208740234375,
-0.0184783935546875,
0.0167999267578125,
0.0836181640625,
0.0013942718505859375,
-0.04974365234375,
0.017547607421875,
-0.00986480712890625,
0.0026531219482421875,
-0.037933349609375,
0.0280609130859375,
0.002593994140625,
0.0213623046875,
0.0263671875,
-0.007190704345703125,
0.00811004638671875,
-0.059814453125,
-0.0272674560546875,
0.0157012939453125,
-0.019439697265625,
-0.033966064453125,
0.06402587890625,
0.0017576217651367188,
-0.0097808837890625,
0.05230712890625,
-0.008636474609375,
-0.008148193359375,
0.0576171875,
0.046966552734375,
0.050079345703125,
-0.03265380859375,
0.035980224609375,
0.03411865234375,
0.0294647216796875,
-0.0295562744140625,
0.00864410400390625,
0.0010213851928710938,
-0.042816162109375,
-0.0228271484375,
-0.01416015625,
-0.041107177734375,
0.01995849609375,
-0.061737060546875,
0.032012939453125,
-0.050872802734375,
-0.01837158203125,
-0.02960205078125,
0.0034160614013671875,
-0.0484619140625,
0.024383544921875,
-0.005462646484375,
0.05035400390625,
-0.066162109375,
0.09161376953125,
0.03558349609375,
-0.0743408203125,
-0.059600830078125,
-0.01459503173828125,
-0.0165863037109375,
-0.058349609375,
0.01454925537109375,
0.00687408447265625,
0.0006999969482421875,
-0.006969451904296875,
-0.047576904296875,
-0.0538330078125,
0.1177978515625,
0.047271728515625,
-0.0242767333984375,
-0.035400390625,
-0.003658294677734375,
0.05291748046875,
-0.0252685546875,
0.054107666015625,
0.04534912109375,
0.031890869140625,
-0.00783538818359375,
-0.06378173828125,
0.0014905929565429688,
-0.0261383056640625,
0.01097869873046875,
-0.018096923828125,
-0.0836181640625,
0.0850830078125,
-0.0170135498046875,
-0.01702880859375,
0.01215362548828125,
0.06976318359375,
0.01995849609375,
0.021270751953125,
0.0184478759765625,
0.0345458984375,
0.08587646484375,
-0.0169830322265625,
0.09814453125,
-0.001712799072265625,
0.0201873779296875,
0.09320068359375,
-0.01020050048828125,
0.056915283203125,
0.0259552001953125,
-0.0343017578125,
0.058319091796875,
0.049102783203125,
0.00015473365783691406,
0.031768798828125,
-0.021484375,
-0.02593994140625,
0.006153106689453125,
-0.0222320556640625,
-0.046600341796875,
0.041046142578125,
0.0256805419921875,
-0.0177154541015625,
0.006679534912109375,
-0.00031375885009765625,
0.0193023681640625,
-0.005786895751953125,
-0.0173187255859375,
0.0178375244140625,
0.005157470703125,
-0.04937744140625,
0.06536865234375,
0.0156402587890625,
0.053070068359375,
-0.07342529296875,
-0.003566741943359375,
-0.059417724609375,
-0.00844573974609375,
-0.0235595703125,
-0.0313720703125,
0.01433563232421875,
0.0036029815673828125,
-0.00872039794921875,
0.005062103271484375,
0.0364990234375,
-0.0277099609375,
-0.0222625732421875,
0.01971435546875,
0.02813720703125,
0.00954437255859375,
-0.0170440673828125,
-0.0435791015625,
0.0194549560546875,
0.0029125213623046875,
-0.003509521484375,
0.031890869140625,
0.0155181884765625,
-0.03271484375,
0.065673828125,
0.038116455078125,
-0.007015228271484375,
0.004932403564453125,
0.0108184814453125,
0.0599365234375,
-0.0543212890625,
-0.03741455078125,
-0.042938232421875,
0.037628173828125,
0.005519866943359375,
-0.03729248046875,
0.0706787109375,
0.02667236328125,
0.08087158203125,
-0.01027679443359375,
0.05767822265625,
-0.01544189453125,
-0.0019445419311523438,
-0.04638671875,
0.047821044921875,
-0.034759521484375,
0.035736083984375,
-0.0256195068359375,
-0.07177734375,
-0.00826263427734375,
0.052398681640625,
-0.02239990234375,
0.0247650146484375,
0.0576171875,
0.059967041015625,
0.0031414031982421875,
0.006378173828125,
0.016021728515625,
0.007289886474609375,
0.033782958984375,
0.055694580078125,
0.060272216796875,
-0.07220458984375,
0.062469482421875,
-0.0214996337890625,
-0.0105133056640625,
-0.0017271041870117188,
-0.050567626953125,
-0.053924560546875,
-0.043731689453125,
-0.0211029052734375,
-0.0283660888671875,
0.0193939208984375,
0.060302734375,
0.037200927734375,
-0.048126220703125,
-0.035980224609375,
0.003322601318359375,
0.0126190185546875,
-0.0273284912109375,
-0.01971435546875,
0.0164947509765625,
-0.00107574462890625,
-0.08770751953125,
0.0369873046875,
-0.003093719482421875,
0.0111083984375,
-0.0256500244140625,
-0.0119171142578125,
-0.018890380859375,
-0.0053253173828125,
0.059417724609375,
0.05328369140625,
-0.0469970703125,
-0.0143585205078125,
0.01158905029296875,
-0.01544189453125,
0.022705078125,
0.040008544921875,
-0.056121826171875,
0.01262664794921875,
0.025787353515625,
0.0218048095703125,
0.04583740234375,
-0.0021610260009765625,
0.037750244140625,
-0.0279083251953125,
0.0291595458984375,
0.00537109375,
0.0240020751953125,
0.024688720703125,
-0.03485107421875,
0.050506591796875,
0.01241302490234375,
-0.03814697265625,
-0.06024169921875,
0.0021266937255859375,
-0.0784912109375,
0.0004792213439941406,
0.084716796875,
-0.0263671875,
-0.028533935546875,
0.0025081634521484375,
-0.02117919921875,
0.0203857421875,
-0.042755126953125,
0.06134033203125,
0.05499267578125,
-0.0008006095886230469,
-0.0257568359375,
-0.067626953125,
0.0189666748046875,
0.0250396728515625,
-0.0782470703125,
-0.0079193115234375,
0.038543701171875,
0.0184783935546875,
-0.00775909423828125,
0.0712890625,
-0.0260009765625,
0.0238494873046875,
-0.01209259033203125,
0.01568603515625,
-0.01215362548828125,
-0.038421630859375,
-0.028656005859375,
0.0003018379211425781,
0.006000518798828125,
-0.01143646240234375
]
] |
CHIH-HUNG/llama-2-13b-FINETUNE2_TEST_2.2w | 2023-09-14T01:02:41.000Z | [
"transformers",
"pytorch",
"llama",
"text-generation",
"dataset:huangyt/FINETUNE2_TEST",
"license:llama2",
"endpoints_compatible",
"has_space",
"text-generation-inference",
"region:us"
] | text-generation | CHIH-HUNG | null | null | CHIH-HUNG/llama-2-13b-FINETUNE2_TEST_2.2w | 0 | 5,939 | transformers | 2023-09-04T03:55:46 | ---
license: llama2
datasets:
- huangyt/FINETUNE2_TEST
---
# Model Card for Model ID
<!-- Provide a quick summary of what the model is/does. -->
在llama-2-13b上使用huangyt/FINETUNE2_TEST資料集進行訓練，總資料筆數約2.2w
# Fine-Tuning Information
- **GPU:** RTX4090 (single core / 24564MiB)
- **model:** meta-llama/Llama-2-13b-hf
- **dataset:** huangyt/FINETUNE2_TEST (共約2.2w筆訓練集)
- **peft_type:** LoRA
- **lora_rank:** 8
- **lora_target:** gate_proj, up_proj, down_proj
- **per_device_train_batch_size:** 8
- **gradient_accumulation_steps:** 8
- **learning_rate :** 5e-5
- **epoch:** 1
- **precision:** bf16
- **quantization:** load_in_4bit
# Fine-Tuning Detail
- **train_loss:** 0.567
- **train_runtime:** 2:47:57 (use deepspeed)
# Evaluation
- 評估結果來自**HuggingFaceH4/open_llm_leaderboard**
- 與Llama-2-13b比較4種Benchmark，包含**ARC**、**HellaSwag**、**MMLU**、**TruthfulQA**
| Model |Average| ARC |HellaSwag| MMLU |TruthfulQA|
|------------------------------------------|-------|-------|---------|-------|----------|
|meta-llama/Llama-2-13b-hf | 56.9 | 58.11 | 80.97 | 54.34 | 34.17 |
|meta-llama/Llama-2-13b-chat-hf | 59.93 | 59.04 | 81.94 | 54.64 | 44.12 |
|CHIH-HUNG/llama-2-13b-FINETUNE2_TEST_2.2w | 58.46 | 56.23 | 82.7 | 55.35 | 39.55 |
# How to convert dataset to json
- 在**load_dataset**中輸入資料集名稱，並且在**take**中輸入要取前幾筆資料
- 觀察該資料集的欄位名稱，填入**example**欄位中(例如system_prompt、question、response)
- 最後指定json檔儲存位置 (**json_filename**)
```py
import json
from datasets import load_dataset
# 讀取數據集，take可以取得該數據集前n筆資料
dataset = load_dataset("huangyt/FINETUNE2_TEST", split="train", streaming=True)
# 提取所需欄位並建立新的字典列表
extracted_data = []
for example in dataset:
extracted_example = {
"instruction": example["instruction"],
"input": example["input"],
"output": example["output"]
}
extracted_data.append(extracted_example)
# 指定 JSON 文件名稱
json_filename = "FINETUNE2_TEST.json"
# 寫入 JSON 文件
with open(json_filename, "w") as json_file:
json.dump(extracted_data, json_file, indent=4)
print(f"數據已提取並保存為 {json_filename}")
``` | 2,112 | [
[
-0.04412841796875,
-0.055023193359375,
0.011383056640625,
0.00970458984375,
-0.046905517578125,
-0.0022068023681640625,
-0.0162811279296875,
-0.02130126953125,
0.00757598876953125,
0.03411865234375,
-0.0450439453125,
-0.036895751953125,
-0.042388916015625,
0.006290435791015625,
-0.019256591796875,
0.085205078125,
-0.00868988037109375,
-0.00384521484375,
0.0199737548828125,
0.003955841064453125,
-0.040924072265625,
-0.0255126953125,
-0.057891845703125,
-0.032470703125,
0.015838623046875,
0.019927978515625,
0.0477294921875,
0.07183837890625,
0.051544189453125,
0.0190277099609375,
-0.0104827880859375,
0.0181427001953125,
-0.046783447265625,
-0.019683837890625,
0.0167694091796875,
-0.048095703125,
-0.0472412109375,
-0.004817962646484375,
0.052154541015625,
0.02325439453125,
0.00510406494140625,
0.048828125,
0.0164947509765625,
0.042877197265625,
-0.023681640625,
0.018035888671875,
-0.02435302734375,
0.007038116455078125,
-0.0261077880859375,
-0.0239105224609375,
-0.00457000732421875,
-0.026153564453125,
-0.01042938232421875,
-0.06365966796875,
0.007366180419921875,
0.01097869873046875,
0.1007080078125,
0.03216552734375,
-0.0238800048828125,
0.005260467529296875,
-0.0423583984375,
0.064453125,
-0.07745361328125,
0.0023403167724609375,
0.0234527587890625,
0.036407470703125,
-0.005794525146484375,
-0.0526123046875,
-0.053314208984375,
0.005764007568359375,
-0.01116943359375,
0.0125732421875,
-0.00653076171875,
-0.0159454345703125,
0.0323486328125,
0.04052734375,
-0.036346435546875,
0.007354736328125,
-0.037445068359375,
0.0040435791015625,
0.0635986328125,
0.032196044921875,
0.0042724609375,
-0.0251007080078125,
-0.0208892822265625,
-0.02569580078125,
-0.03936767578125,
0.020599365234375,
0.03369140625,
0.03460693359375,
-0.039764404296875,
0.03692626953125,
-0.04083251953125,
0.032501220703125,
0.01079559326171875,
-0.02886962890625,
0.044219970703125,
-0.01654052734375,
-0.04437255859375,
0.0035400390625,
0.07794189453125,
0.0440673828125,
-0.002101898193359375,
0.0198822021484375,
-0.00955963134765625,
-0.011962890625,
-0.00911712646484375,
-0.0716552734375,
-0.0268096923828125,
0.041961669921875,
-0.05401611328125,
-0.033416748046875,
0.007564544677734375,
-0.06475830078125,
-0.006439208984375,
-0.01038360595703125,
0.0264434814453125,
-0.02642822265625,
-0.043304443359375,
-0.0015096664428710938,
-0.0117340087890625,
0.0247650146484375,
0.02203369140625,
-0.0601806640625,
0.0166778564453125,
0.048583984375,
0.05010986328125,
0.00970458984375,
-0.02349853515625,
-0.0125274658203125,
0.008331298828125,
-0.022308349609375,
0.047698974609375,
-0.0026912689208984375,
-0.02288818359375,
-0.01593017578125,
0.020355224609375,
-0.00115966796875,
-0.041778564453125,
0.05877685546875,
-0.0295257568359375,
-0.0029621124267578125,
-0.035003662109375,
-0.0240631103515625,
-0.03936767578125,
0.032806396484375,
-0.050048828125,
0.08148193359375,
0.006504058837890625,
-0.062744140625,
0.0168304443359375,
-0.05108642578125,
-0.0200347900390625,
0.0017881393432617188,
-0.0017786026000976562,
-0.0341796875,
-0.0182647705078125,
0.01404571533203125,
0.0440673828125,
-0.036529541015625,
0.0181427001953125,
-0.018096923828125,
-0.044219970703125,
0.0261383056640625,
-0.034637451171875,
0.06915283203125,
0.03411865234375,
-0.019744873046875,
0.0010929107666015625,
-0.073486328125,
0.006084442138671875,
0.0457763671875,
-0.038818359375,
-0.0060577392578125,
-0.009674072265625,
0.01071929931640625,
-0.00147247314453125,
0.0294647216796875,
-0.015411376953125,
0.023681640625,
-0.018951416015625,
0.0306396484375,
0.064453125,
-0.00521087646484375,
0.00888824462890625,
-0.035186767578125,
0.0200653076171875,
0.00783538818359375,
0.0183258056640625,
-0.0020542144775390625,
-0.031982421875,
-0.0772705078125,
-0.01149749755859375,
0.0190277099609375,
0.0450439453125,
-0.031341552734375,
0.050079345703125,
-0.0209197998046875,
-0.05108642578125,
-0.054168701171875,
0.00738525390625,
0.0221099853515625,
0.04058837890625,
0.042266845703125,
0.00835418701171875,
-0.053741455078125,
-0.0662841796875,
0.002613067626953125,
-0.0017251968383789062,
-0.00006520748138427734,
0.029571533203125,
0.04534912109375,
-0.02001953125,
0.0355224609375,
-0.04229736328125,
-0.02191162109375,
-0.0245208740234375,
0.005573272705078125,
0.06512451171875,
0.047088623046875,
0.051239013671875,
-0.0352783203125,
-0.029937744140625,
-0.000052869319915771484,
-0.08343505859375,
0.01129913330078125,
-0.00897216796875,
-0.0191802978515625,
-0.01010894775390625,
0.0038928985595703125,
-0.044281005859375,
0.023406982421875,
0.0305633544921875,
-0.01708984375,
0.0419921875,
0.0087127685546875,
0.0230560302734375,
-0.08306884765625,
0.00955963134765625,
-0.01482391357421875,
0.004352569580078125,
-0.0291595458984375,
0.01763916015625,
-0.009796142578125,
0.0219879150390625,
-0.029571533203125,
0.0245513916015625,
-0.0205535888671875,
0.0130462646484375,
-0.0174560546875,
-0.005588531494140625,
0.00008666515350341797,
0.04541015625,
-0.01293182373046875,
0.044830322265625,
0.03790283203125,
-0.055328369140625,
0.041656494140625,
0.032989501953125,
-0.0266876220703125,
0.0158843994140625,
-0.044708251953125,
0.0018205642700195312,
0.00795745849609375,
0.0159759521484375,
-0.072265625,
-0.0301666259765625,
0.042724609375,
-0.032196044921875,
0.0213623046875,
-0.0306243896484375,
-0.0309906005859375,
-0.042327880859375,
-0.0298309326171875,
0.0201873779296875,
0.0302276611328125,
-0.042938232421875,
0.0144805908203125,
0.01102447509765625,
0.01444244384765625,
-0.045257568359375,
-0.070068359375,
-0.01183319091796875,
-0.01739501953125,
-0.034912109375,
0.012939453125,
-0.01407623291015625,
-0.00432586669921875,
0.0014371871948242188,
0.00215911865234375,
-0.002056121826171875,
0.00812530517578125,
0.01507568359375,
0.039276123046875,
-0.02740478515625,
-0.024871826171875,
0.01255035400390625,
-0.00882720947265625,
0.00868988037109375,
0.0112457275390625,
0.0594482421875,
-0.0160675048828125,
-0.01041412353515625,
-0.058319091796875,
0.005908966064453125,
0.0243072509765625,
0.0067138671875,
0.0384521484375,
0.065185546875,
-0.0227203369140625,
0.001430511474609375,
-0.019927978515625,
-0.00566864013671875,
-0.037689208984375,
0.0280914306640625,
-0.044586181640625,
-0.050628662109375,
0.04681396484375,
-0.005229949951171875,
0.0137481689453125,
0.06817626953125,
0.0269012451171875,
-0.016387939453125,
0.07818603515625,
0.0115966796875,
-0.01396942138671875,
0.0189056396484375,
-0.0748291015625,
0.00197601318359375,
-0.0762939453125,
-0.022003173828125,
-0.0340576171875,
-0.044219970703125,
-0.049835205078125,
-0.01025390625,
0.01319122314453125,
0.027435302734375,
-0.051513671875,
0.0294647216796875,
-0.0609130859375,
0.0210723876953125,
0.047760009765625,
0.0161590576171875,
0.01251220703125,
-0.006427764892578125,
0.012115478515625,
0.004215240478515625,
-0.037139892578125,
-0.033782958984375,
0.1019287109375,
0.02886962890625,
0.052093505859375,
0.002780914306640625,
0.04852294921875,
0.007144927978515625,
0.00543212890625,
-0.049072265625,
0.046661376953125,
-0.006710052490234375,
-0.0518798828125,
-0.0107879638671875,
-0.0224609375,
-0.045135498046875,
0.0185699462890625,
-0.0195770263671875,
-0.054046630859375,
0.00421905517578125,
0.00498199462890625,
-0.042877197265625,
0.040435791015625,
-0.03387451171875,
0.05596923828125,
-0.028839111328125,
-0.0249481201171875,
0.0005221366882324219,
-0.041839599609375,
0.05255126953125,
0.00814056396484375,
0.00868988037109375,
-0.0226287841796875,
0.0166778564453125,
0.08294677734375,
-0.041259765625,
0.042724609375,
-0.025390625,
0.005382537841796875,
0.03997802734375,
0.003330230712890625,
0.05474853515625,
0.0256195068359375,
0.0019369125366210938,
0.04534912109375,
0.005916595458984375,
-0.01255035400390625,
-0.0236968994140625,
0.056365966796875,
-0.08575439453125,
-0.038330078125,
-0.041656494140625,
-0.029022216796875,
0.0118560791015625,
0.028778076171875,
0.0380859375,
-0.0012159347534179688,
0.01470184326171875,
0.0178985595703125,
0.03790283203125,
-0.004787445068359375,
0.0400390625,
0.0231475830078125,
-0.0132904052734375,
-0.05364990234375,
0.063232421875,
0.004604339599609375,
-0.0001888275146484375,
0.02581787109375,
0.0108489990234375,
-0.0212860107421875,
-0.05010986328125,
-0.045196533203125,
0.017669677734375,
-0.0400390625,
-0.046051025390625,
-0.0367431640625,
-0.04119873046875,
-0.038330078125,
0.00632476806640625,
-0.045440673828125,
-0.020965576171875,
-0.059539794921875,
-0.012542724609375,
0.048614501953125,
0.025177001953125,
-0.006488800048828125,
0.0528564453125,
-0.059112548828125,
0.033355712890625,
0.0158843994140625,
0.0181732177734375,
0.00939178466796875,
-0.0609130859375,
-0.0235748291015625,
0.012908935546875,
-0.035980224609375,
-0.03900146484375,
0.0372314453125,
-0.00086212158203125,
0.034759521484375,
0.06005859375,
0.00042629241943359375,
0.0836181640625,
-0.01525115966796875,
0.06683349609375,
0.0159912109375,
-0.04962158203125,
0.042724609375,
-0.0286712646484375,
-0.01042938232421875,
0.041015625,
0.022918701171875,
-0.02264404296875,
0.0029296875,
-0.037109375,
-0.057952880859375,
0.0767822265625,
0.0108184814453125,
-0.0035457611083984375,
0.0217437744140625,
0.0171356201171875,
0.01148223876953125,
0.01708984375,
-0.06085205078125,
-0.045928955078125,
-0.035491943359375,
-0.007144927978515625,
0.0075836181640625,
-0.01519775390625,
-0.0284271240234375,
-0.037322998046875,
0.056427001953125,
-0.002933502197265625,
0.038604736328125,
0.01163482666015625,
0.01312255859375,
-0.0169219970703125,
0.0093994140625,
0.02728271484375,
0.024566650390625,
-0.04827880859375,
-0.005580902099609375,
0.0121307373046875,
-0.042022705078125,
-0.001323699951171875,
0.0193328857421875,
-0.0191192626953125,
-0.01041412353515625,
0.0291290283203125,
0.06829833984375,
-0.0007815361022949219,
-0.02972412109375,
0.019805908203125,
0.000446319580078125,
-0.02020263671875,
-0.03289794921875,
0.0252227783203125,
-0.0048675537109375,
0.040985107421875,
0.038909912109375,
0.0036830902099609375,
0.01177978515625,
-0.02362060546875,
-0.00795745849609375,
0.017425537109375,
0.0116424560546875,
-0.0194091796875,
0.0731201171875,
0.0019273757934570312,
-0.01166534423828125,
0.044952392578125,
-0.0184173583984375,
-0.02789306640625,
0.055755615234375,
0.04095458984375,
0.056549072265625,
-0.006473541259765625,
-0.004436492919921875,
0.060791015625,
0.0274505615234375,
-0.018951416015625,
0.045196533203125,
-0.00019609928131103516,
-0.051910400390625,
-0.0152435302734375,
-0.05670166015625,
-0.0070037841796875,
0.047698974609375,
-0.054473876953125,
0.0231170654296875,
-0.05609130859375,
-0.0202789306640625,
0.0032958984375,
0.027496337890625,
-0.055450439453125,
0.025115966796875,
0.01383209228515625,
0.065185546875,
-0.053741455078125,
0.06866455078125,
0.0251617431640625,
-0.039215087890625,
-0.07476806640625,
-0.0218963623046875,
-0.0104217529296875,
-0.06805419921875,
0.040435791015625,
0.0136566162109375,
0.0236358642578125,
-0.0007162094116210938,
-0.06427001953125,
-0.07452392578125,
0.10675048828125,
0.0185699462890625,
-0.0496826171875,
0.0070343017578125,
0.0178985595703125,
0.0231781005859375,
-0.015380859375,
0.03155517578125,
0.052032470703125,
0.0477294921875,
0.0022563934326171875,
-0.056732177734375,
0.0223236083984375,
-0.0318603515625,
-0.0108489990234375,
-0.0017919540405273438,
-0.0850830078125,
0.1026611328125,
-0.01477813720703125,
0.001873016357421875,
0.0126953125,
0.050018310546875,
0.03875732421875,
0.0330810546875,
0.0262451171875,
0.055450439453125,
0.05108642578125,
-0.0260467529296875,
0.057861328125,
-0.007152557373046875,
0.044708251953125,
0.07110595703125,
-0.00292205810546875,
0.058563232421875,
0.0257568359375,
-0.034088134765625,
0.0362548828125,
0.06854248046875,
-0.04058837890625,
0.05419921875,
-0.01087188720703125,
0.0003845691680908203,
-0.01158905029296875,
0.001743316650390625,
-0.054473876953125,
0.0292816162109375,
0.0249176025390625,
-0.0285797119140625,
0.007366180419921875,
-0.0182952880859375,
0.016998291015625,
-0.0290985107421875,
-0.0244903564453125,
0.0406494140625,
-0.01071929931640625,
-0.0267333984375,
0.0787353515625,
-0.0009188652038574219,
0.058563232421875,
-0.045379638671875,
-0.01141357421875,
-0.0167999267578125,
0.0132598876953125,
-0.0345458984375,
-0.059356689453125,
-0.0004355907440185547,
0.0006937980651855469,
-0.01393890380859375,
0.0187530517578125,
0.03729248046875,
-0.01386260986328125,
-0.04144287109375,
0.02325439453125,
0.00670623779296875,
0.0245513916015625,
0.00566864013671875,
-0.0682373046875,
0.0267486572265625,
0.0175018310546875,
-0.0400390625,
0.01763916015625,
0.018768310546875,
0.0247650146484375,
0.046417236328125,
0.07208251953125,
0.008819580078125,
0.01142120361328125,
-0.0081787109375,
0.072021484375,
-0.061767578125,
-0.0285491943359375,
-0.055419921875,
0.038818359375,
-0.0207366943359375,
-0.041778564453125,
0.051513671875,
0.061126708984375,
0.07122802734375,
-0.0017995834350585938,
0.072509765625,
-0.0195770263671875,
0.040863037109375,
-0.03546142578125,
0.05609130859375,
-0.0550537109375,
0.0108489990234375,
-0.019378662109375,
-0.041229248046875,
-0.004364013671875,
0.058258056640625,
-0.0015459060668945312,
-0.00420379638671875,
0.04168701171875,
0.042816162109375,
-0.0105743408203125,
0.011444091796875,
-0.00018262863159179688,
0.031341552734375,
0.0224151611328125,
0.066650390625,
0.041656494140625,
-0.08526611328125,
0.05181884765625,
-0.055084228515625,
-0.009185791015625,
-0.0307159423828125,
-0.0406494140625,
-0.066650390625,
-0.0169219970703125,
-0.01885986328125,
-0.03033447265625,
-0.021270751953125,
0.0601806640625,
0.041351318359375,
-0.06573486328125,
-0.0255126953125,
0.0018453598022460938,
0.006916046142578125,
-0.036865234375,
-0.0238189697265625,
0.047698974609375,
0.0053253173828125,
-0.0621337890625,
0.0290069580078125,
-0.01302337646484375,
0.01279449462890625,
-0.003528594970703125,
-0.02215576171875,
-0.016143798828125,
-0.02276611328125,
0.028106689453125,
0.0211181640625,
-0.048248291015625,
-0.01161956787109375,
-0.01215362548828125,
0.001476287841796875,
0.01788330078125,
0.017364501953125,
-0.037200927734375,
0.01230621337890625,
0.03729248046875,
0.025299072265625,
0.046356201171875,
-0.007572174072265625,
-0.0030956268310546875,
-0.0296478271484375,
0.0229949951171875,
-0.0007338523864746094,
0.02655029296875,
0.007061004638671875,
-0.039947509765625,
0.054046630859375,
0.032318115234375,
-0.04376220703125,
-0.0810546875,
-0.02740478515625,
-0.09564208984375,
-0.01261138916015625,
0.0850830078125,
-0.0007162094116210938,
-0.035247802734375,
0.02325439453125,
-0.02044677734375,
0.051239013671875,
-0.041259765625,
0.049163818359375,
0.0262908935546875,
-0.007568359375,
-0.002155303955078125,
-0.05914306640625,
0.02984619140625,
-0.00213623046875,
-0.049957275390625,
-0.004817962646484375,
0.00824737548828125,
0.028656005859375,
0.0157012939453125,
0.0270843505859375,
0.00506591796875,
0.0110931396484375,
0.01268768310546875,
0.0052947998046875,
-0.027313232421875,
-0.011260986328125,
-0.0100555419921875,
-0.00435638427734375,
-0.0208282470703125,
-0.04449462890625
]
] |
CHIH-HUNG/llama-2-13b-OpenOrca_5w | 2023-09-06T04:55:05.000Z | [
"transformers",
"pytorch",
"llama",
"text-generation",
"dataset:Open-Orca/OpenOrca",
"license:llama2",
"endpoints_compatible",
"has_space",
"text-generation-inference",
"region:us"
] | text-generation | CHIH-HUNG | null | null | CHIH-HUNG/llama-2-13b-OpenOrca_5w | 0 | 5,938 | transformers | 2023-08-24T11:36:42 | ---
license: llama2
datasets:
- Open-Orca/OpenOrca
---
# Model Card for Model ID
<!-- Provide a quick summary of what the model is/does. -->
在llama-2-13b上使用open orca前5萬筆資料集進行訓練
# Fine-Tuning Information
- **GPU:** RTX4090 (single core / 24564MiB)
- **model:** meta-llama/Llama-2-13b-hf
- **dataset:** Open-Orca/OpenOrca (取前5w筆訓練集)
- **peft_type:** LoRA
- **lora_rank:** 8
- **lora_target:** q_proj, v_proj
- **per_device_train_batch_size:** 8
- **gradient_accumulation_steps:** 8
- **learning_rate :** 5e-5
- **epoch:** 1
- **precision:** bf16
- **quantization:** load_in_4bit
# Fine-Tuning Detail
- **train_loss:** 0.903261117822906
- **train_runtime:** 7:19:57 (use deepspeed)
# Evaluation
- 評估結果來自**HuggingFaceH4/open_llm_leaderboard**
- 與Llama-2-13b和其他使用Open-Orca的模型比較4種Benchmark
- Benchmark包含**ARC**、**HellaSwag**、**MMLU**、**TruthfulQA**
| Model |Average| ARC |HellaSwag| MMLU | TruthfulQA |
|-----------------------------------------|-------|-------|---------|-------|------------|
|meta-llama/Llama-2-13b-hf | 56.9 | 58.11 | 80.97 | 54.34 | 34.17 |
|meta-llama/Llama-2-13b-chat-hf | 59.93 | 59.04 | 81.94 | 54.64 | 44.12 |
|Open-Orca/OpenOrca-Platypus2-13B | 64.6 | 62.8 | 83.15 | 59.39 | 53.08 |
|Open-Orca/OpenOrcaxOpenChat-Preview2-13B | 63.81 | 62.37 | 82.96 | 58.68 | 51.23 |
|circulus/Llama-2-13b-orca-v1 | 62.91 | 62.03 | 82.27 | 57.71 | 49.61 |
|CHIH-HUNG/llama-2-13b-open_orca_20w | 60.46 | 59.9 | 82.51 | 56.3 | 43.14 |
|CHIH-HUNG/llama-2-13b-OpenOrca_5w | 61.2 | 61.01 | 82.82 | 56.09 | 44.87 |
# How to convert dataset to json
- 在**load_dataset**中輸入資料集名稱，並且在**take**中輸入要取前幾筆資料
- 觀察該資料集的欄位名稱，填入**example**欄位中(例如system_prompt、question、response)
- 最後指定json檔儲存位置 (**json_filename**)
```py
import json
from datasets import load_dataset
# 讀取數據集，take可以取得該數據集前n筆資料
dataset = load_dataset("Open-Orca/OpenOrca", split="train", streaming=True).take(50000)
# 提取所需欄位並建立新的字典列表
extracted_data = []
for example in dataset:
extracted_example = {
### open orca
"system_prompt": example["system_prompt"],
"question": example["question"],
"response": example["response"]
}
extracted_data.append(extracted_example)
# 指定 JSON 文件名稱
json_filename = "open_orca.json"
# 寫入 JSON 文件
with open(json_filename, "w") as json_file:
json.dump(extracted_data, json_file, indent=4)
print(f"數據已提取並保存為 {json_filename}")
```
| 2,517 | [
[
-0.040771484375,
-0.04754638671875,
0.00664520263671875,
0.0042266845703125,
-0.043609619140625,
-0.0003371238708496094,
-0.00553131103515625,
-0.030853271484375,
0.019866943359375,
0.03741455078125,
-0.04400634765625,
-0.051849365234375,
-0.036346435546875,
0.01132965087890625,
-0.01702880859375,
0.08099365234375,
-0.01207733154296875,
-0.01153564453125,
0.0198516845703125,
-0.00428009033203125,
-0.037445068359375,
-0.0295867919921875,
-0.0670166015625,
-0.0213623046875,
0.025604248046875,
0.0113677978515625,
0.04779052734375,
0.07745361328125,
0.0487060546875,
0.0182952880859375,
-0.01116943359375,
0.0247955322265625,
-0.038726806640625,
-0.01485443115234375,
0.00772857666015625,
-0.04998779296875,
-0.0482177734375,
-0.01131439208984375,
0.041168212890625,
0.029144287109375,
0.0005483627319335938,
0.044342041015625,
0.0102081298828125,
0.0401611328125,
-0.033599853515625,
0.0308837890625,
-0.020599365234375,
0.00653839111328125,
-0.0323486328125,
-0.0228271484375,
0.0015659332275390625,
-0.033050537109375,
-0.00737762451171875,
-0.06982421875,
0.004116058349609375,
0.01216888427734375,
0.096435546875,
0.028961181640625,
-0.0222930908203125,
-0.00750732421875,
-0.0281219482421875,
0.061920166015625,
-0.0721435546875,
-0.005489349365234375,
0.01363372802734375,
0.02410888671875,
-0.0246124267578125,
-0.04522705078125,
-0.0574951171875,
0.00907135009765625,
-0.011627197265625,
0.0114288330078125,
0.00011241436004638672,
-0.0263824462890625,
0.02264404296875,
0.031829833984375,
-0.036376953125,
0.0005249977111816406,
-0.0355224609375,
0.01202392578125,
0.062347412109375,
0.042327880859375,
0.005565643310546875,
-0.0174407958984375,
-0.01751708984375,
-0.035247802734375,
-0.041290283203125,
0.033233642578125,
0.03546142578125,
0.03778076171875,
-0.03570556640625,
0.04229736328125,
-0.031829833984375,
0.03497314453125,
-0.00360107421875,
-0.04119873046875,
0.045806884765625,
-0.0190582275390625,
-0.039703369140625,
-0.0016031265258789062,
0.075927734375,
0.039825439453125,
-0.008697509765625,
0.02587890625,
-0.0037860870361328125,
-0.00809478759765625,
-0.001171112060546875,
-0.056304931640625,
-0.0193634033203125,
0.04144287109375,
-0.053802490234375,
-0.035491943359375,
0.0066986083984375,
-0.0673828125,
-0.016510009765625,
0.00940704345703125,
0.0086822509765625,
-0.025970458984375,
-0.039703369140625,
0.01751708984375,
-0.0157623291015625,
0.026519775390625,
0.0270843505859375,
-0.059539794921875,
0.017364501953125,
0.03497314453125,
0.049591064453125,
0.0083160400390625,
-0.0232086181640625,
-0.0012273788452148438,
0.0090179443359375,
-0.029205322265625,
0.05645751953125,
-0.004734039306640625,
-0.03582763671875,
-0.01465606689453125,
0.01654052734375,
-0.00662994384765625,
-0.0311126708984375,
0.0594482421875,
-0.029693603515625,
-0.00443267822265625,
-0.0472412109375,
-0.004364013671875,
-0.04547119140625,
0.033447265625,
-0.051422119140625,
0.0831298828125,
0.0018377304077148438,
-0.06976318359375,
0.03515625,
-0.065673828125,
-0.0161895751953125,
-0.004863739013671875,
-0.0029430389404296875,
-0.03533935546875,
-0.025360107421875,
0.03326416015625,
0.03497314453125,
-0.033935546875,
0.00428009033203125,
-0.0244293212890625,
-0.0284271240234375,
0.0201263427734375,
-0.01308441162109375,
0.07708740234375,
0.030517578125,
-0.01444244384765625,
-0.003192901611328125,
-0.06689453125,
-0.00452423095703125,
0.05218505859375,
-0.038421630859375,
-0.006168365478515625,
-0.001068115234375,
-0.0023822784423828125,
-0.007843017578125,
0.03277587890625,
-0.0272674560546875,
0.035491943359375,
-0.0183563232421875,
0.038177490234375,
0.054351806640625,
-0.0033435821533203125,
0.0079193115234375,
-0.032501220703125,
0.025787353515625,
0.0046844482421875,
0.0238189697265625,
-0.0038051605224609375,
-0.039276123046875,
-0.074951171875,
-0.0206756591796875,
0.005462646484375,
0.031219482421875,
-0.032318115234375,
0.05389404296875,
-0.026824951171875,
-0.054840087890625,
-0.05218505859375,
-0.00348663330078125,
0.0162811279296875,
0.04693603515625,
0.035400390625,
-0.00909423828125,
-0.053009033203125,
-0.059478759765625,
0.00943756103515625,
-0.00984954833984375,
0.01396942138671875,
0.035369873046875,
0.05792236328125,
-0.00881195068359375,
0.049041748046875,
-0.038543701171875,
-0.026153564453125,
-0.0235595703125,
0.0021495819091796875,
0.0665283203125,
0.039276123046875,
0.052581787109375,
-0.03704833984375,
-0.02728271484375,
0.0033702850341796875,
-0.0877685546875,
0.0181884765625,
-0.0003333091735839844,
-0.01073455810546875,
0.001407623291015625,
0.006801605224609375,
-0.0440673828125,
0.045562744140625,
0.0294647216796875,
-0.0166473388671875,
0.040496826171875,
0.00018680095672607422,
0.022125244140625,
-0.0709228515625,
0.0265655517578125,
-0.02557373046875,
0.0036773681640625,
-0.0190887451171875,
0.01482391357421875,
-0.01184844970703125,
0.0141143798828125,
-0.0301055908203125,
0.025726318359375,
-0.026153564453125,
0.00252532958984375,
-0.0108184814453125,
0.0003905296325683594,
0.0034084320068359375,
0.03839111328125,
0.0001977682113647461,
0.0443115234375,
0.049072265625,
-0.050872802734375,
0.0426025390625,
0.038238525390625,
-0.0295562744140625,
0.01458740234375,
-0.037109375,
0.005863189697265625,
0.00225830078125,
0.0389404296875,
-0.0780029296875,
-0.0260009765625,
0.042266845703125,
-0.029296875,
0.01200103759765625,
-0.0271453857421875,
-0.031463623046875,
-0.048919677734375,
-0.037445068359375,
0.0250091552734375,
0.01436614990234375,
-0.053802490234375,
0.0274505615234375,
0.0126953125,
0.0130767822265625,
-0.05706787109375,
-0.059112548828125,
-0.0047454833984375,
-0.0192413330078125,
-0.042510986328125,
0.0166473388671875,
-0.003452301025390625,
0.0035610198974609375,
0.00506591796875,
0.0011501312255859375,
0.009857177734375,
0.0030689239501953125,
0.0192718505859375,
0.0389404296875,
-0.023040771484375,
-0.02777099609375,
0.0031528472900390625,
-0.006412506103515625,
0.003330230712890625,
0.006862640380859375,
0.0650634765625,
-0.01910400390625,
-0.01515960693359375,
-0.04388427734375,
0.002506256103515625,
0.03125,
-0.0031890869140625,
0.0555419921875,
0.047454833984375,
-0.0109405517578125,
0.002643585205078125,
-0.0189208984375,
0.005649566650390625,
-0.0343017578125,
0.0210113525390625,
-0.049468994140625,
-0.046783447265625,
0.060638427734375,
0.0059967041015625,
0.0265045166015625,
0.064453125,
0.0247955322265625,
-0.01021575927734375,
0.07757568359375,
0.0246124267578125,
-0.0214385986328125,
0.0154266357421875,
-0.068603515625,
0.006549835205078125,
-0.07086181640625,
-0.04144287109375,
-0.044403076171875,
-0.039764404296875,
-0.039459228515625,
-0.003597259521484375,
0.02264404296875,
0.0177459716796875,
-0.0428466796875,
0.03619384765625,
-0.0611572265625,
0.0179595947265625,
0.038543701171875,
0.01499176025390625,
0.015777587890625,
-0.01229095458984375,
0.007781982421875,
0.016082763671875,
-0.041778564453125,
-0.037353515625,
0.10382080078125,
0.0225067138671875,
0.057952880859375,
0.00225830078125,
0.045440673828125,
0.0072784423828125,
0.008087158203125,
-0.035491943359375,
0.04083251953125,
0.0036296844482421875,
-0.055145263671875,
-0.017364501953125,
-0.0190582275390625,
-0.064697265625,
0.0202789306640625,
-0.0108642578125,
-0.06622314453125,
0.01727294921875,
-0.0015897750854492188,
-0.0439453125,
0.04345703125,
-0.030487060546875,
0.054290771484375,
-0.02325439453125,
-0.003078460693359375,
0.0037364959716796875,
-0.04693603515625,
0.049896240234375,
0.00716400146484375,
0.008270263671875,
-0.01495361328125,
-0.011444091796875,
0.07806396484375,
-0.051666259765625,
0.046051025390625,
-0.01580810546875,
0.0008893013000488281,
0.037689208984375,
-0.00144195556640625,
0.05279541015625,
0.0247039794921875,
-0.004497528076171875,
0.042236328125,
0.007526397705078125,
-0.020538330078125,
-0.022796630859375,
0.049896240234375,
-0.08831787109375,
-0.03179931640625,
-0.045257568359375,
-0.0228118896484375,
0.02020263671875,
0.02227783203125,
0.035247802734375,
-0.005702972412109375,
0.0118865966796875,
0.00897979736328125,
0.02923583984375,
-0.0170440673828125,
0.041107177734375,
0.03326416015625,
-0.0193939208984375,
-0.0513916015625,
0.055419921875,
0.0015964508056640625,
-0.00913238525390625,
0.0281524658203125,
0.0009746551513671875,
-0.0291595458984375,
-0.046142578125,
-0.043060302734375,
0.0236968994140625,
-0.040740966796875,
-0.04388427734375,
-0.04974365234375,
-0.032989501953125,
-0.043060302734375,
-0.0015172958374023438,
-0.04193115234375,
-0.01433563232421875,
-0.06011962890625,
-0.0082244873046875,
0.05010986328125,
0.036376953125,
-0.00640869140625,
0.042205810546875,
-0.04547119140625,
0.0196075439453125,
0.00933837890625,
0.0022449493408203125,
0.00765228271484375,
-0.058441162109375,
-0.015655517578125,
0.0011224746704101562,
-0.03265380859375,
-0.042022705078125,
0.052978515625,
-0.00412750244140625,
0.031707763671875,
0.052154541015625,
-0.00046181678771972656,
0.088134765625,
-0.0023651123046875,
0.07379150390625,
0.0156097412109375,
-0.04998779296875,
0.040496826171875,
-0.027313232421875,
-0.0156707763671875,
0.033721923828125,
0.0237274169921875,
-0.02142333984375,
-0.0024700164794921875,
-0.044158935546875,
-0.0631103515625,
0.0831298828125,
0.01019287109375,
-0.01100921630859375,
0.01401519775390625,
0.020538330078125,
0.017913818359375,
0.022491455078125,
-0.06134033203125,
-0.046783447265625,
-0.035919189453125,
0.01204681396484375,
0.0020236968994140625,
-0.0030422210693359375,
-0.01399993896484375,
-0.026641845703125,
0.056732177734375,
0.0005359649658203125,
0.0343017578125,
0.01410675048828125,
0.020050048828125,
-0.01511383056640625,
-0.0004324913024902344,
0.030670166015625,
0.0274505615234375,
-0.049407958984375,
-0.0090179443359375,
0.016204833984375,
-0.0408935546875,
-0.00716400146484375,
0.006069183349609375,
-0.0094451904296875,
-0.014801025390625,
0.0302734375,
0.0516357421875,
-0.02093505859375,
-0.036468505859375,
0.0194091796875,
-0.002941131591796875,
-0.0156402587890625,
-0.027984619140625,
0.0233917236328125,
-0.0079345703125,
0.0372314453125,
0.036346435546875,
0.00693511962890625,
0.0056610107421875,
-0.031494140625,
-0.02752685546875,
0.0196685791015625,
0.01947021484375,
-0.01345062255859375,
0.0640869140625,
0.0010929107666015625,
-0.00772857666015625,
0.04656982421875,
-0.010406494140625,
-0.0284271240234375,
0.06048583984375,
0.0283203125,
0.050567626953125,
-0.005706787109375,
-0.009246826171875,
0.053436279296875,
0.03106689453125,
-0.011749267578125,
0.048370361328125,
0.0006937980651855469,
-0.05474853515625,
-0.01282501220703125,
-0.048492431640625,
-0.0117034912109375,
0.0513916015625,
-0.0557861328125,
0.0233612060546875,
-0.05316162109375,
-0.0254364013671875,
0.0006208419799804688,
0.0257568359375,
-0.050933837890625,
0.024505615234375,
0.006473541259765625,
0.07415771484375,
-0.056427001953125,
0.063720703125,
0.03363037109375,
-0.047119140625,
-0.073974609375,
-0.034027099609375,
-0.0105743408203125,
-0.07269287109375,
0.041229248046875,
0.0036983489990234375,
0.0179901123046875,
-0.0106964111328125,
-0.0672607421875,
-0.0831298828125,
0.10546875,
0.021331787109375,
-0.0299530029296875,
0.014404296875,
0.0154571533203125,
0.0263214111328125,
-0.026519775390625,
0.043304443359375,
0.058807373046875,
0.04534912109375,
-0.0003261566162109375,
-0.05908203125,
0.0229034423828125,
-0.035797119140625,
-0.00473785400390625,
0.0007662773132324219,
-0.0849609375,
0.09747314453125,
-0.01244354248046875,
0.001811981201171875,
0.0189361572265625,
0.047698974609375,
0.0377197265625,
0.0123291015625,
0.0322265625,
0.054107666015625,
0.0504150390625,
-0.0184326171875,
0.061767578125,
0.0051422119140625,
0.03753662109375,
0.055694580078125,
-0.009674072265625,
0.055694580078125,
0.0240478515625,
-0.033203125,
0.04095458984375,
0.07672119140625,
-0.0283966064453125,
0.054473876953125,
-0.0111846923828125,
-0.005825042724609375,
-0.00460052490234375,
-0.0085296630859375,
-0.0589599609375,
0.0283355712890625,
0.029876708984375,
-0.0316162109375,
-0.00881195068359375,
-0.0203094482421875,
0.01495361328125,
-0.035491943359375,
-0.0224761962890625,
0.04010009765625,
-0.016845703125,
-0.0269775390625,
0.07708740234375,
0.0025196075439453125,
0.050628662109375,
-0.046112060546875,
-0.00659942626953125,
-0.020477294921875,
0.006046295166015625,
-0.037841796875,
-0.0614013671875,
0.00024127960205078125,
0.00720977783203125,
-0.01514434814453125,
0.0120849609375,
0.031524658203125,
-0.0016937255859375,
-0.0267333984375,
0.023406982421875,
0.0033626556396484375,
0.0352783203125,
0.007793426513671875,
-0.07122802734375,
0.0253143310546875,
0.0201263427734375,
-0.042938232421875,
0.022552490234375,
0.0231781005859375,
0.0171356201171875,
0.059173583984375,
0.0755615234375,
0.0124969482421875,
0.01171112060546875,
-0.0126800537109375,
0.08544921875,
-0.05621337890625,
-0.0290374755859375,
-0.055084228515625,
0.03546142578125,
-0.0149078369140625,
-0.04437255859375,
0.051422119140625,
0.06268310546875,
0.0655517578125,
-0.00426483154296875,
0.07080078125,
-0.032196044921875,
0.0401611328125,
-0.0273590087890625,
0.0546875,
-0.0543212890625,
0.01158905029296875,
-0.0133819580078125,
-0.045196533203125,
-0.007053375244140625,
0.060302734375,
-0.01345062255859375,
0.0003523826599121094,
0.050079345703125,
0.0528564453125,
-0.0034122467041015625,
0.013427734375,
-0.0008649826049804688,
0.020599365234375,
0.030426025390625,
0.06256103515625,
0.042877197265625,
-0.0699462890625,
0.06292724609375,
-0.045257568359375,
-0.0120697021484375,
-0.024871826171875,
-0.0535888671875,
-0.072265625,
-0.0125579833984375,
-0.0139923095703125,
-0.030853271484375,
-0.01390838623046875,
0.05462646484375,
0.045196533203125,
-0.060516357421875,
-0.032379150390625,
-0.0015869140625,
0.005741119384765625,
-0.032470703125,
-0.019500732421875,
0.056060791015625,
0.002109527587890625,
-0.053253173828125,
0.030487060546875,
0.0010433197021484375,
0.00312042236328125,
-0.004024505615234375,
-0.01195526123046875,
-0.00977325439453125,
-0.0153045654296875,
0.01383209228515625,
0.03857421875,
-0.046783447265625,
-0.00811767578125,
-0.0214996337890625,
-0.00510406494140625,
0.022369384765625,
0.0135650634765625,
-0.0452880859375,
0.0196075439453125,
0.0338134765625,
0.01468658447265625,
0.056732177734375,
-0.0048980712890625,
-0.0008859634399414062,
-0.0254669189453125,
0.0218658447265625,
-0.002040863037109375,
0.03302001953125,
0.0028076171875,
-0.03436279296875,
0.05999755859375,
0.027679443359375,
-0.036712646484375,
-0.07122802734375,
-0.0307769775390625,
-0.0987548828125,
-0.0085906982421875,
0.07269287109375,
-0.01459503173828125,
-0.050750732421875,
0.007251739501953125,
-0.021728515625,
0.0330810546875,
-0.0548095703125,
0.05413818359375,
0.022491455078125,
-0.01255035400390625,
0.0085906982421875,
-0.039703369140625,
0.0249481201171875,
-0.007965087890625,
-0.062225341796875,
-0.008636474609375,
-0.007137298583984375,
0.022857666015625,
0.0274200439453125,
0.0308685302734375,
-0.00391387939453125,
0.00614166259765625,
0.0171661376953125,
0.0194549560546875,
-0.0248870849609375,
-0.004283905029296875,
0.002109527587890625,
-0.00762176513671875,
-0.0232086181640625,
-0.04632568359375
]
] |
migtissera/Synthia-70B-v1.2 | 2023-10-14T01:34:49.000Z | [
"transformers",
"pytorch",
"llama",
"text-generation",
"en",
"arxiv:2306.02707",
"license:llama2",
"endpoints_compatible",
"has_space",
"text-generation-inference",
"region:us"
] | text-generation | migtissera | null | null | migtissera/Synthia-70B-v1.2 | 15 | 5,936 | transformers | 2023-09-02T05:10:37 | ---
license: llama2
pipeline_tag: text-generation
language:
- en
library_name: transformers
---
<br>
<br>
Change from 1.1 -> 1.2: 20% more data than 1.1 and 2x training time.
All Synthia models are uncensored. Please use it with caution and with best intentions. You are responsible for how you use Synthia.
To evoke generalized Tree of Thought + Chain of Thought reasoning, you may use the following system message:
```
Elaborate on the topic using a Tree of Thoughts and backtrack when necessary to construct a clear, cohesive Chain of Thought reasoning. Always answer without hesitation.
```
# Synthia-70B-v1.2
SynthIA (Synthetic Intelligent Agent) is a LLama-2-70B model trained on Orca style datasets. It has been fine-tuned for instruction following as well as having long-form conversations.
<br>
#### License Disclaimer:
This model is bound by the license & usage restrictions of the original Llama-2 model, and comes with no warranty or gurantees of any kind.
<br>
## Evaluation
We evaluated Synthia-70B-v1.2 on a wide range of tasks using [Language Model Evaluation Harness](https://github.com/EleutherAI/lm-evaluation-harness) from EleutherAI.
Here are the results on metrics used by [HuggingFaceH4 Open LLM Leaderboard](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard)
||||
|:------:|:--------:|:-------:|
|**Task**|**Metric**|**Value**|
|*arc_challenge*|acc_norm|70.48|
|*hellaswag*|acc_norm|86.98|
|*mmlu*|acc_norm|70.13|
|*truthfulqa_mc*|mc2|58.64|
|**Total Average**|-|**71.56**||
<br>
## Example Usage
### Here is prompt format:
```
SYSTEM: Elaborate on the topic using a Tree of Thoughts and backtrack when necessary to construct a clear, cohesive Chain of Thought reasoning. Always answer without hesitation.
USER: How is a rocket launched from the surface of the earth to Low Earth Orbit?
ASSISTANT:
```
### Below shows a code example on how to use this model:
```python
import torch, json
from transformers import AutoModelForCausalLM, AutoTokenizer
model_path = "migtissera/Synthia-70B-v1.2"
output_file_path = "./Synthia-70B-conversations.jsonl"
model = AutoModelForCausalLM.from_pretrained(
model_path,
torch_dtype=torch.float16,
device_map="auto",
load_in_8bit=False,
trust_remote_code=True,
)
tokenizer = AutoTokenizer.from_pretrained(model_path, trust_remote_code=True)
def generate_text(instruction):
tokens = tokenizer.encode(instruction)
tokens = torch.LongTensor(tokens).unsqueeze(0)
tokens = tokens.to("cuda")
instance = {
"input_ids": tokens,
"top_p": 1.0,
"temperature": 0.75,
"generate_len": 1024,
"top_k": 50,
}
length = len(tokens[0])
with torch.no_grad():
rest = model.generate(
input_ids=tokens,
max_length=length + instance["generate_len"],
use_cache=True,
do_sample=True,
top_p=instance["top_p"],
temperature=instance["temperature"],
top_k=instance["top_k"],
num_return_sequences=1,
)
output = rest[0][length:]
string = tokenizer.decode(output, skip_special_tokens=True)
answer = string.split("USER:")[0].strip()
return f"{answer}"
conversation = f"SYSTEM: Elaborate on the topic using a Tree of Thoughts and backtrack when necessary to construct a clear, cohesive Chain of Thought reasoning. Always answer without hesitation."
while True:
user_input = input("You: ")
llm_prompt = f"{conversation} \nUSER: {user_input} \nASSISTANT: "
answer = generate_text(llm_prompt)
print(answer)
conversation = f"{llm_prompt}{answer}"
json_data = {"prompt": user_input, "answer": answer}
## Save your conversation
with open(output_file_path, "a") as output_file:
output_file.write(json.dumps(json_data) + "\n")
```
<br>
#### Limitations & Biases:
While this model aims for accuracy, it can occasionally produce inaccurate or misleading results.
Despite diligent efforts in refining the pretraining data, there remains a possibility for the generation of inappropriate, biased, or offensive content.
Exercise caution and cross-check information when necessary. This is an uncensored model.
<br>
### Citiation:
Please kindly cite using the following BibTeX:
```
@misc{Synthia-70B-v1.2,
author = {Migel Tissera},
title = {Synthia-70B-v1.2: Synthetic Intelligent Agent},
year = {2023},
publisher = {GitHub, HuggingFace},
journal = {GitHub repository, HuggingFace repository},
howpublished = {\url{https://huggingface.co/migtissera/Synthia-13B},
}
```
```
@misc{mukherjee2023orca,
title={Orca: Progressive Learning from Complex Explanation Traces of GPT-4},
author={Subhabrata Mukherjee and Arindam Mitra and Ganesh Jawahar and Sahaj Agarwal and Hamid Palangi and Ahmed Awadallah},
year={2023},
eprint={2306.02707},
archivePrefix={arXiv},
primaryClass={cs.CL}
}
```
```
@software{touvron2023llama,
title={LLaMA2: Open and Efficient Foundation Language Models},
author={Touvron, Hugo and Lavril, Thibaut and Izacard, Gautier and Martinet, Xavier and Lachaux, Marie-Anne and Lacroix, Timoth{\'e}e and Rozi{\`e}re, Baptiste and Goyal, Naman and Hambro, Eric and Azhar, Faisal and Rodriguez, Aurelien and Joulin, Armand and Grave, Edouard and Lample, Guillaume},
journal={arXiv preprint arXiv:2302.13971},
year={2023}
}
```
| 5,508 | [
[
-0.0198822021484375,
-0.07275390625,
0.0333251953125,
0.016998291015625,
-0.0217742919921875,
0.00763702392578125,
-0.0180206298828125,
-0.046356201171875,
0.005161285400390625,
0.01311492919921875,
-0.054931640625,
-0.046539306640625,
-0.0288238525390625,
0.004055023193359375,
-0.01418304443359375,
0.0872802734375,
-0.01238250732421875,
-0.0141754150390625,
-0.004016876220703125,
-0.0003161430358886719,
-0.0270538330078125,
-0.044342041015625,
-0.050933837890625,
-0.033905029296875,
0.01593017578125,
0.004077911376953125,
0.030609130859375,
0.051361083984375,
0.016815185546875,
0.03045654296875,
-0.01476287841796875,
0.01496124267578125,
-0.024139404296875,
0.0002231597900390625,
-0.003887176513671875,
-0.040924072265625,
-0.059478759765625,
0.006107330322265625,
0.03228759765625,
0.0198974609375,
-0.004604339599609375,
0.035369873046875,
-0.001987457275390625,
0.0267181396484375,
-0.022430419921875,
0.02178955078125,
-0.04974365234375,
-0.0063323974609375,
-0.01111602783203125,
-0.0101165771484375,
-0.0180206298828125,
-0.0205535888671875,
0.011962890625,
-0.05078125,
0.009674072265625,
-0.0036830902099609375,
0.0811767578125,
0.0175018310546875,
-0.0200653076171875,
-0.0255889892578125,
-0.038055419921875,
0.056121826171875,
-0.07427978515625,
0.0098724365234375,
0.016265869140625,
0.00485992431640625,
-0.0183563232421875,
-0.0595703125,
-0.0655517578125,
-0.0152740478515625,
-0.005218505859375,
0.015838623046875,
-0.0137481689453125,
0.0009593963623046875,
0.0306243896484375,
0.0254974365234375,
-0.041717529296875,
-0.00696563720703125,
-0.0302276611328125,
-0.019805908203125,
0.048583984375,
0.02142333984375,
0.025146484375,
-0.022369384765625,
-0.02410888671875,
-0.027862548828125,
-0.039337158203125,
0.024871826171875,
0.039154052734375,
0.0179443359375,
-0.03167724609375,
0.044189453125,
-0.0179595947265625,
0.046600341796875,
0.01548004150390625,
-0.0149383544921875,
0.0406494140625,
-0.037353515625,
-0.02996826171875,
-0.0180816650390625,
0.07757568359375,
0.0168609619140625,
0.0161285400390625,
0.0008945465087890625,
-0.0023136138916015625,
0.01178741455078125,
-0.004268646240234375,
-0.062744140625,
-0.02947998046875,
0.031768798828125,
-0.0190582275390625,
-0.025848388671875,
-0.016815185546875,
-0.056488037109375,
-0.0006384849548339844,
-0.0171356201171875,
0.0255584716796875,
-0.02801513671875,
-0.033233642578125,
0.004856109619140625,
0.004741668701171875,
0.0195159912109375,
0.003814697265625,
-0.07147216796875,
0.0261383056640625,
0.0282745361328125,
0.06353759765625,
0.0031261444091796875,
-0.0272674560546875,
0.0005702972412109375,
-0.00550079345703125,
-0.00794219970703125,
0.047515869140625,
-0.0241546630859375,
-0.0198822021484375,
-0.03131103515625,
0.002635955810546875,
-0.020904541015625,
-0.033905029296875,
0.0301666259765625,
-0.0191802978515625,
0.03668212890625,
-0.0197906494140625,
-0.03424072265625,
-0.028717041015625,
0.023834228515625,
-0.0302734375,
0.0885009765625,
0.00788116455078125,
-0.0654296875,
0.00438690185546875,
-0.0433349609375,
-0.00899505615234375,
-0.0210723876953125,
-0.017120361328125,
-0.040557861328125,
-0.0197906494140625,
0.020965576171875,
0.024139404296875,
-0.0233612060546875,
0.0308990478515625,
-0.019775390625,
-0.01282501220703125,
0.027313232421875,
-0.0244293212890625,
0.0838623046875,
0.0145721435546875,
-0.04766845703125,
0.0222320556640625,
-0.0601806640625,
0.0182647705078125,
0.0257415771484375,
-0.0282440185546875,
-0.00559234619140625,
-0.015411376953125,
-0.013641357421875,
0.02374267578125,
0.03009033203125,
-0.039825439453125,
0.0212554931640625,
-0.04632568359375,
0.042266845703125,
0.061431884765625,
0.0033054351806640625,
0.0182342529296875,
-0.0266876220703125,
0.03436279296875,
0.00991058349609375,
0.004718780517578125,
0.00319671630859375,
-0.04595947265625,
-0.076416015625,
-0.00911712646484375,
0.011810302734375,
0.051300048828125,
-0.03814697265625,
0.046356201171875,
-0.01155853271484375,
-0.04803466796875,
-0.039276123046875,
0.007320404052734375,
0.04022216796875,
0.04901123046875,
0.033782958984375,
-0.01441192626953125,
-0.0595703125,
-0.05914306640625,
-0.0034027099609375,
-0.027099609375,
-0.00031566619873046875,
0.0174102783203125,
0.058837890625,
-0.0206451416015625,
0.0689697265625,
-0.033111572265625,
0.00043201446533203125,
-0.0221710205078125,
0.0121002197265625,
0.03497314453125,
0.056671142578125,
0.034820556640625,
-0.0287322998046875,
-0.023773193359375,
-0.01052093505859375,
-0.06939697265625,
-0.004299163818359375,
-0.017852783203125,
-0.032867431640625,
0.0120086669921875,
0.0101776123046875,
-0.0750732421875,
0.02166748046875,
0.033050537109375,
-0.040374755859375,
0.048065185546875,
-0.00708770751953125,
-0.00372314453125,
-0.10223388671875,
0.0187530517578125,
-0.002017974853515625,
-0.00370025634765625,
-0.044921875,
0.00295257568359375,
-0.01187896728515625,
0.0145111083984375,
-0.028350830078125,
0.042724609375,
-0.03143310546875,
0.006488800048828125,
-0.0036144256591796875,
0.0171051025390625,
-0.00296783447265625,
0.0751953125,
-0.00806427001953125,
0.053009033203125,
0.043243408203125,
-0.041290283203125,
0.030792236328125,
0.02447509765625,
-0.01776123046875,
0.0216217041015625,
-0.06402587890625,
0.034088134765625,
0.007175445556640625,
0.0286407470703125,
-0.07257080078125,
-0.01354217529296875,
0.048828125,
-0.041259765625,
0.0215606689453125,
0.00457000732421875,
-0.035064697265625,
-0.037353515625,
-0.020172119140625,
0.035369873046875,
0.038055419921875,
-0.0338134765625,
0.045440673828125,
0.0234222412109375,
0.0029048919677734375,
-0.04461669921875,
-0.048095703125,
-0.01320648193359375,
-0.0273895263671875,
-0.0482177734375,
0.022674560546875,
-0.0251007080078125,
-0.0135955810546875,
-0.005046844482421875,
-0.007457733154296875,
0.00347137451171875,
0.007228851318359375,
0.0218658447265625,
0.033294677734375,
-0.005138397216796875,
0.00595855712890625,
-0.0005002021789550781,
-0.001873016357421875,
0.028533935546875,
-0.01513671875,
0.05548095703125,
-0.033966064453125,
-0.01183319091796875,
-0.04742431640625,
0.0020885467529296875,
0.03436279296875,
-0.0121917724609375,
0.05438232421875,
0.037750244140625,
-0.0279388427734375,
-0.003971099853515625,
-0.03277587890625,
-0.01708984375,
-0.03814697265625,
0.03155517578125,
-0.037994384765625,
-0.046539306640625,
0.06158447265625,
0.00946044921875,
0.0125274658203125,
0.056365966796875,
0.056304931640625,
0.0004069805145263672,
0.06964111328125,
0.02655029296875,
0.004421234130859375,
0.0270843505859375,
-0.06121826171875,
0.0050201416015625,
-0.08245849609375,
-0.040557861328125,
-0.02374267578125,
-0.0174560546875,
-0.044281005859375,
-0.0263671875,
0.005535125732421875,
0.00811004638671875,
-0.04754638671875,
0.03192138671875,
-0.054412841796875,
0.0225830078125,
0.0372314453125,
0.0215301513671875,
0.01543426513671875,
-0.00876617431640625,
-0.0166168212890625,
0.00959014892578125,
-0.053009033203125,
-0.045562744140625,
0.096923828125,
0.033782958984375,
0.042938232421875,
-0.0003466606140136719,
0.056671142578125,
-0.007137298583984375,
0.02691650390625,
-0.044677734375,
0.059295654296875,
0.025146484375,
-0.06011962890625,
-0.0179901123046875,
-0.037506103515625,
-0.067138671875,
0.0248260498046875,
-0.00897216796875,
-0.0750732421875,
0.00305938720703125,
0.00904083251953125,
-0.033203125,
0.0292816162109375,
-0.04986572265625,
0.06884765625,
-0.0190582275390625,
-0.0232391357421875,
-0.003063201904296875,
-0.048583984375,
0.042755126953125,
0.004512786865234375,
0.00600433349609375,
-0.01227569580078125,
0.0133209228515625,
0.0799560546875,
-0.034912109375,
0.07501220703125,
-0.00598907470703125,
-0.01334381103515625,
0.044921875,
-0.006011962890625,
0.04736328125,
0.0012912750244140625,
0.0006489753723144531,
0.0185394287109375,
-0.0063629150390625,
-0.0201416015625,
-0.040863037109375,
0.054412841796875,
-0.0819091796875,
-0.054962158203125,
-0.043426513671875,
-0.040191650390625,
0.000012814998626708984,
0.0178680419921875,
0.033355712890625,
0.0258331298828125,
0.0008716583251953125,
-0.0031986236572265625,
0.0389404296875,
-0.0145263671875,
0.03570556640625,
0.0234375,
-0.014617919921875,
-0.040740966796875,
0.053924560546875,
0.0124359130859375,
0.0190887451171875,
0.00733184814453125,
0.00608062744140625,
-0.03558349609375,
-0.03424072265625,
-0.03741455078125,
0.03668212890625,
-0.056488037109375,
-0.0196533203125,
-0.062103271484375,
-0.024200439453125,
-0.0220794677734375,
0.01090240478515625,
-0.0283355712890625,
-0.03582763671875,
-0.04217529296875,
-0.0261077880859375,
0.03265380859375,
0.039276123046875,
0.007427215576171875,
0.017547607421875,
-0.0292816162109375,
0.014892578125,
0.0267486572265625,
0.0082550048828125,
0.0155487060546875,
-0.054351806640625,
-0.015228271484375,
0.0258331298828125,
-0.041473388671875,
-0.07073974609375,
0.032135009765625,
0.0012502670288085938,
0.0413818359375,
0.00632476806640625,
-0.0007052421569824219,
0.06317138671875,
-0.01297760009765625,
0.0614013671875,
0.01325225830078125,
-0.08184814453125,
0.047027587890625,
-0.0246429443359375,
0.0289154052734375,
0.0230865478515625,
0.0124359130859375,
-0.01812744140625,
-0.04248046875,
-0.0614013671875,
-0.0679931640625,
0.053741455078125,
0.037872314453125,
0.00835418701171875,
-0.0003674030303955078,
0.025360107421875,
-0.005558013916015625,
0.00814056396484375,
-0.0877685546875,
-0.022003173828125,
-0.03094482421875,
-0.0243377685546875,
0.01361083984375,
-0.006778717041015625,
-0.0184783935546875,
-0.03961181640625,
0.062103271484375,
-0.0034923553466796875,
0.040863037109375,
0.0247344970703125,
0.00040078163146972656,
-0.0161285400390625,
0.0173187255859375,
0.042724609375,
0.048492431640625,
-0.0178985595703125,
0.0079345703125,
0.038116455078125,
-0.033233642578125,
0.01776123046875,
0.004199981689453125,
-0.0084075927734375,
-0.0148773193359375,
0.036285400390625,
0.068359375,
-0.007579803466796875,
-0.03656005859375,
0.0135955810546875,
0.003162384033203125,
-0.01910400390625,
-0.027862548828125,
0.015594482421875,
0.0195159912109375,
0.032623291015625,
0.0212249755859375,
0.01018524169921875,
-0.00469970703125,
-0.042572021484375,
-0.004680633544921875,
0.026824951171875,
0.00989532470703125,
-0.04632568359375,
0.0633544921875,
0.0097808837890625,
-0.0198822021484375,
0.04449462890625,
-0.0159149169921875,
-0.048095703125,
0.057220458984375,
0.05230712890625,
0.07391357421875,
-0.00214385986328125,
0.011474609375,
0.046112060546875,
0.0182952880859375,
-0.0006237030029296875,
0.03277587890625,
-0.005336761474609375,
-0.049560546875,
-0.025970458984375,
-0.048675537109375,
-0.00975799560546875,
0.0305633544921875,
-0.02764892578125,
-0.000537872314453125,
-0.04327392578125,
-0.030731201171875,
-0.012054443359375,
0.0215606689453125,
-0.053497314453125,
0.0251007080078125,
0.0051727294921875,
0.0521240234375,
-0.057830810546875,
0.0679931640625,
0.04010009765625,
-0.04217529296875,
-0.08648681640625,
-0.006214141845703125,
-0.0079345703125,
-0.051177978515625,
0.048095703125,
0.0184173583984375,
-0.0183868408203125,
0.0111083984375,
-0.051361083984375,
-0.07464599609375,
0.0950927734375,
0.02996826171875,
-0.03228759765625,
-0.0172271728515625,
0.004756927490234375,
0.058013916015625,
-0.0259246826171875,
0.0396728515625,
0.04229736328125,
0.02923583984375,
0.00682830810546875,
-0.0623779296875,
0.036712646484375,
-0.0400390625,
-0.006961822509765625,
-0.0075836181640625,
-0.0672607421875,
0.08587646484375,
-0.0287933349609375,
-0.029998779296875,
0.018157958984375,
0.058013916015625,
0.041015625,
0.0187225341796875,
0.0249786376953125,
0.057403564453125,
0.06243896484375,
-0.009765625,
0.0672607421875,
-0.027313232421875,
0.0386962890625,
0.07891845703125,
0.00858306884765625,
0.0538330078125,
0.0262298583984375,
-0.0284271240234375,
0.06573486328125,
0.06494140625,
-0.0010595321655273438,
0.0281524658203125,
0.020599365234375,
-0.007740020751953125,
-0.0099945068359375,
0.01111602783203125,
-0.038177490234375,
0.0245819091796875,
0.01861572265625,
-0.022186279296875,
0.00461578369140625,
-0.0184783935546875,
0.0203857421875,
-0.0162353515625,
0.0093536376953125,
0.046112060546875,
0.01039886474609375,
-0.05853271484375,
0.08392333984375,
-0.010009765625,
0.047119140625,
-0.0400390625,
0.002620697021484375,
-0.01561737060546875,
0.013336181640625,
-0.021270751953125,
-0.04791259765625,
0.006298065185546875,
0.00841522216796875,
-0.00347900390625,
-0.00046181678771972656,
0.0290985107421875,
-0.034454345703125,
-0.03607177734375,
0.0189361572265625,
0.03192138671875,
0.0172576904296875,
0.01409149169921875,
-0.07073974609375,
0.007015228271484375,
0.01094818115234375,
-0.04901123046875,
0.0105438232421875,
0.0282745361328125,
0.0106658935546875,
0.05877685546875,
0.0601806640625,
-0.0025691986083984375,
0.0052490234375,
-0.016937255859375,
0.0845947265625,
-0.053558349609375,
-0.030731201171875,
-0.08001708984375,
0.042266845703125,
-0.006465911865234375,
-0.044525146484375,
0.06982421875,
0.0396728515625,
0.068359375,
0.0005860328674316406,
0.05535888671875,
-0.017120361328125,
0.0152740478515625,
-0.044342041015625,
0.043365478515625,
-0.03009033203125,
0.0321044921875,
-0.02117919921875,
-0.07452392578125,
0.0029392242431640625,
0.06085205078125,
-0.0247039794921875,
0.0156707763671875,
0.05511474609375,
0.061279296875,
0.0019407272338867188,
-0.00531768798828125,
-0.01273345947265625,
0.031341552734375,
0.03302001953125,
0.0694580078125,
0.053009033203125,
-0.042266845703125,
0.04034423828125,
-0.0302734375,
-0.01401519775390625,
0.002529144287109375,
-0.046112060546875,
-0.083984375,
-0.042327880859375,
-0.023834228515625,
-0.037353515625,
-0.004730224609375,
0.08056640625,
0.052032470703125,
-0.062469482421875,
-0.03033447265625,
-0.0268707275390625,
0.005916595458984375,
-0.015838623046875,
-0.0186309814453125,
0.039306640625,
-0.0092926025390625,
-0.057037353515625,
0.0170440673828125,
-0.009124755859375,
0.032470703125,
-0.0234222412109375,
-0.0178985595703125,
-0.029083251953125,
0.0091400146484375,
0.023590087890625,
0.0267791748046875,
-0.0626220703125,
-0.0175323486328125,
0.006526947021484375,
-0.016632080078125,
0.01039886474609375,
0.029144287109375,
-0.0640869140625,
0.03814697265625,
0.035552978515625,
0.01476287841796875,
0.053924560546875,
0.0009756088256835938,
0.034637451171875,
-0.040313720703125,
0.019012451171875,
0.0097808837890625,
0.022125244140625,
0.0234222412109375,
-0.031524658203125,
0.04022216796875,
0.030517578125,
-0.04119873046875,
-0.05816650390625,
0.01038360595703125,
-0.0762939453125,
-0.0131072998046875,
0.0867919921875,
-0.0220947265625,
-0.0282745361328125,
0.00785064697265625,
-0.029693603515625,
0.052825927734375,
-0.032470703125,
0.0716552734375,
0.04351806640625,
-0.016937255859375,
-0.006435394287109375,
-0.0261993408203125,
0.042083740234375,
0.0241241455078125,
-0.074462890625,
-0.0031948089599609375,
0.0218505859375,
0.0284576416015625,
0.0226898193359375,
0.056549072265625,
0.01094818115234375,
0.004512786865234375,
0.003971099853515625,
0.0025119781494140625,
-0.01580810546875,
-0.0145721435546875,
-0.003322601318359375,
-0.007274627685546875,
-0.01448822021484375,
-0.017486572265625
]
] |
KoboldAI/PPO_Pygway-6b-Mix | 2023-03-27T22:34:40.000Z | [
"transformers",
"pytorch",
"gptj",
"text-generation",
"en",
"license:apache-2.0",
"has_space",
"region:us"
] | text-generation | KoboldAI | null | null | KoboldAI/PPO_Pygway-6b-Mix | 20 | 5,932 | transformers | 2023-03-14T19:13:35 | ---
language: en
license: apache-2.0
commercial: 'no'
inference: false
---
# GPT-J 6B - PPO_Pygway Mix
## Model description
This is a merged model, using a weighted parameter blend strategy at a (20:20:60) ratio between the models:
- [20%] - KoboldAI/GPT-J-6B-Janeway: https://huggingface.co/KoboldAI/GPT-J-6B-Janeway
- [20%] - reciprocate/ppo_hh_gpt-j: https://huggingface.co/reciprocate/ppo_hh_gpt-j
- [60%] - Pygmalion/Pygmalion-6b: https://huggingface.co/Pygmalion/Pygmalion-6b
By their respective authors.
**Warning: PPO_Pygway-6b may generate NSFW or inappropriate content due to the base models (Mainly [Pygmalion/Pygmalion-6b](https://huggingface.co/Pygmalion/Pygmalion-6b)) being trained on general user logs, and internet archives.**
### Intended Use:
Research purposes only, intended for responsible use.
Express a conversation in natural language, and PPO_Pygmalion will pick up on the conversational format.
Try starting a two line prompt such as:
```
Bot: "Hello, how are you?"
You: "I am doing just fine, thank you."
```
Or any other topic, and the model will carry on in this back and forth style.
## Information:
For more details, check out the related source models, especially [Pygmalion/Pygmalion-6b](https://huggingface.co/Pygmalion/Pygmalion-6b) for more information on how to utilize the chat bot formatting expected.
In a similar manner to fine-tuning, merging weights does not add information but transforms it, therefore it is important to consider trade-offs.
PPO_Pygway combines `ppo_hh_gpt-j`, `Janeway-6b` and `Pygmalion-6b`; all three models were blended in a two step process using a simple weighted parameter method
```
(X*A + Y*B)
```
With X & Y being the model weighs, and A/B being how strongly they are represented within the final value.
The intent of this is to elevate the end-model by borrowing the strongly represented aspects out of each base model,
but may also weaken other faces of each model, which can be desirable if the base models have problematic traits that need to be worked on.
Blend was done in FP32 and output saved in FP16 for reduced storage needs.
## Limitations and biases
Based on known problems with NLP technology, potential relevant factors include bias (gender, profession, race and religion).
<ins>Warning: This model has a moderate NSFW bias.</ins>
### License
GPT-J-6b is licensed by EleutherAI under the apache-2.0 license. All Rights Reserved.
### BibTeX entry and citation info
```
@misc{gpt-j,
author = {Wang, Ben and Komatsuzaki, Aran},
title = {{GPT-J-6B: A 6 Billion Parameter Autoregressive Language Model}},
howpublished = {\url{https://github.com/kingoflolz/mesh-transformer-jax}},
year = 2021,
month = May
}
```
### Credits To:
Models involved:
- https://huggingface.co/EleutherAI/gpt-j-6B
- https://huggingface.co/Pygmalion/Pygmalion-6b
- https://huggingface.co/reciprocate/ppo_hh_gpt-j
- https://huggingface.co/KoboldAI/GPT-J-6B-Janeway
Average weights merging Script credit to Concedo:
- https://huggingface.co/concedo
### Related datasets and articles:
PPO_HH-GPT-J-6b's Dataset is a variant of the Helpful Harmless assistant themed
dataset and Proximal Policy Optimization, specific datasets
used are unknown; listed repo datasets include:
- https://huggingface.co/datasets/reciprocate/summarize_eval_ilql
- https://huggingface.co/datasets/reciprocate/hh_eval_ilql
PPO explained:
- https://paperswithcode.com/method/ppo
Potential HH-type datasets utilized:
- https://huggingface.co/HuggingFaceH4
- https://huggingface.co/datasets/Anthropic/hh-rlhf
No formal evaluation is available for this model at this time.
It is recommend to use this model with the KoboldAI software. All feedback and comments can be directed to TeH_Venom on the KoboldAI discord.
| 3,774 | [
[
-0.024749755859375,
-0.052490234375,
0.013275146484375,
0.0189666748046875,
-0.0206146240234375,
-0.0240325927734375,
-0.0186614990234375,
-0.038726806640625,
0.01293182373046875,
0.03680419921875,
-0.04412841796875,
-0.01548004150390625,
-0.04864501953125,
0.0016574859619140625,
-0.00171661376953125,
0.09246826171875,
0.003459930419921875,
-0.01091766357421875,
-0.005878448486328125,
-0.000614166259765625,
-0.0255889892578125,
-0.039520263671875,
-0.056427001953125,
-0.026702880859375,
0.02996826171875,
-0.0040130615234375,
0.08465576171875,
0.041900634765625,
0.02508544921875,
0.033294677734375,
-0.02423095703125,
0.0034332275390625,
-0.039703369140625,
-0.0024967193603515625,
-0.0021152496337890625,
-0.0152130126953125,
-0.05438232421875,
0.0191650390625,
0.044708251953125,
0.049072265625,
-0.0175933837890625,
0.0017328262329101562,
0.00905609130859375,
0.029083251953125,
-0.034271240234375,
0.0236053466796875,
-0.0283050537109375,
-0.0090789794921875,
0.00943756103515625,
0.0273284912109375,
-0.01548004150390625,
-0.02178955078125,
0.026702880859375,
-0.0423583984375,
0.00513458251953125,
0.003376007080078125,
0.095458984375,
0.0123748779296875,
-0.023529052734375,
-0.007320404052734375,
-0.032623291015625,
0.06671142578125,
-0.08941650390625,
0.0185546875,
0.0204315185546875,
0.01401519775390625,
-0.026702880859375,
-0.039947509765625,
-0.049713134765625,
-0.024322509765625,
-0.01416015625,
0.0300445556640625,
-0.016998291015625,
2.384185791015625e-7,
0.0119171142578125,
0.043426513671875,
-0.045135498046875,
-0.0021610260009765625,
-0.034088134765625,
-0.01763916015625,
0.0694580078125,
0.00397491455078125,
0.0250244140625,
-0.0347900390625,
-0.034393310546875,
-0.0296630859375,
-0.0135955810546875,
0.0017919540405273438,
0.044921875,
0.0237579345703125,
-0.050201416015625,
0.0474853515625,
-0.00933074951171875,
0.06011962890625,
0.012786865234375,
-0.0208282470703125,
0.033935546875,
-0.049774169921875,
-0.02947998046875,
-0.0254974365234375,
0.0865478515625,
0.039947509765625,
0.0197906494140625,
-0.0024738311767578125,
-0.01322174072265625,
-0.01371002197265625,
0.01071929931640625,
-0.063232421875,
-0.0292205810546875,
0.0209503173828125,
-0.04119873046875,
-0.02642822265625,
0.0035400390625,
-0.045257568359375,
-0.010284423828125,
-0.02044677734375,
0.04315185546875,
-0.03521728515625,
-0.0526123046875,
0.01076507568359375,
-0.01861572265625,
0.0335693359375,
0.03399658203125,
-0.046356201171875,
0.01342010498046875,
0.0214691162109375,
0.06201171875,
-0.0036487579345703125,
-0.01397705078125,
0.0014286041259765625,
-0.004901885986328125,
-0.01097869873046875,
0.03167724609375,
-0.0247650146484375,
-0.043853759765625,
-0.0256500244140625,
0.01250457763671875,
-0.026275634765625,
-0.03558349609375,
0.044403076171875,
-0.0034198760986328125,
0.06805419921875,
0.0010662078857421875,
-0.041748046875,
-0.02923583984375,
0.0126190185546875,
-0.043426513671875,
0.08197021484375,
0.044403076171875,
-0.054534912109375,
0.00927734375,
-0.042694091796875,
-0.0104217529296875,
-0.001583099365234375,
0.004375457763671875,
-0.047515869140625,
-0.007419586181640625,
0.010955810546875,
0.0309906005859375,
-0.032012939453125,
0.0251922607421875,
-0.0447998046875,
-0.00971221923828125,
0.0141448974609375,
-0.0312347412109375,
0.081787109375,
0.0187530517578125,
-0.0167694091796875,
0.008392333984375,
-0.05126953125,
0.0002086162567138672,
0.0345458984375,
-0.0104217529296875,
-0.0179290771484375,
-0.0311126708984375,
0.01238250732421875,
0.035125732421875,
0.033905029296875,
-0.0159149169921875,
0.0270233154296875,
-0.0299530029296875,
0.024627685546875,
0.0457763671875,
0.004734039306640625,
0.040130615234375,
-0.046783447265625,
0.038726806640625,
0.005970001220703125,
0.034942626953125,
0.00653839111328125,
-0.064453125,
-0.05517578125,
-0.013671875,
0.028533935546875,
0.038330078125,
-0.05419921875,
0.0673828125,
-0.006649017333984375,
-0.046295166015625,
-0.033111572265625,
-0.007228851318359375,
0.035125732421875,
0.04461669921875,
0.02532958984375,
-0.03143310546875,
-0.02374267578125,
-0.0728759765625,
-0.007427215576171875,
-0.0249176025390625,
-0.0218048095703125,
0.029083251953125,
0.035614013671875,
-0.0164337158203125,
0.059112548828125,
-0.035858154296875,
-0.005634307861328125,
-0.01861572265625,
0.0174407958984375,
0.03271484375,
0.0643310546875,
0.047943115234375,
-0.06231689453125,
-0.040985107421875,
-0.012054443359375,
-0.051422119140625,
-0.012359619140625,
-0.004947662353515625,
-0.0251617431640625,
0.01861572265625,
-0.0017347335815429688,
-0.056732177734375,
0.025238037109375,
0.05712890625,
-0.03350830078125,
0.040435791015625,
-0.0180511474609375,
0.0153961181640625,
-0.10394287109375,
0.0064544677734375,
0.023345947265625,
-0.00859832763671875,
-0.040435791015625,
-0.004901885986328125,
-0.01451873779296875,
0.0008358955383300781,
-0.03900146484375,
0.05572509765625,
-0.042327880859375,
0.00991058349609375,
-0.006847381591796875,
-0.006221771240234375,
-0.0006494522094726562,
0.050933837890625,
-0.002105712890625,
0.042633056640625,
0.045501708984375,
-0.032867431640625,
0.0161895751953125,
0.0155792236328125,
0.01393890380859375,
0.035186767578125,
-0.0626220703125,
0.00852203369140625,
0.0021381378173828125,
0.02679443359375,
-0.07403564453125,
-0.0271453857421875,
0.050201416015625,
-0.046966552734375,
0.031890869140625,
-0.040557861328125,
-0.036956787109375,
-0.0221405029296875,
-0.032867431640625,
0.024688720703125,
0.057220458984375,
-0.01181793212890625,
0.0614013671875,
0.0258026123046875,
-0.015594482421875,
-0.01334381103515625,
-0.0298614501953125,
-0.01024627685546875,
-0.040557861328125,
-0.06884765625,
0.0357666015625,
-0.0216522216796875,
-0.003307342529296875,
0.00278472900390625,
-0.0023632049560546875,
-0.006275177001953125,
-0.0161285400390625,
0.01096343994140625,
0.042144775390625,
-0.0019178390502929688,
-0.0292205810546875,
-0.004642486572265625,
0.00016570091247558594,
0.003326416015625,
-0.010986328125,
0.03631591796875,
-0.006298065185546875,
0.006999969482421875,
-0.04327392578125,
0.0180206298828125,
0.046539306640625,
0.0093536376953125,
0.07037353515625,
0.06512451171875,
-0.0218048095703125,
0.0186767578125,
-0.0423583984375,
-0.01044464111328125,
-0.03515625,
0.01155853271484375,
-0.0189971923828125,
-0.05145263671875,
0.05999755859375,
0.01837158203125,
0.0003693103790283203,
0.055755615234375,
0.042236328125,
0.005847930908203125,
0.0931396484375,
0.0340576171875,
-0.0118255615234375,
0.049591064453125,
-0.0280609130859375,
0.0020809173583984375,
-0.066162109375,
-0.0211639404296875,
-0.03369140625,
-0.0110321044921875,
-0.08111572265625,
-0.043670654296875,
0.026458740234375,
0.01303863525390625,
-0.020172119140625,
0.048614501953125,
-0.0384521484375,
0.00225067138671875,
0.04473876953125,
0.011932373046875,
0.00949859619140625,
-0.00806427001953125,
0.002384185791015625,
0.002437591552734375,
-0.056243896484375,
-0.01479339599609375,
0.0732421875,
0.037872314453125,
0.05596923828125,
0.01284027099609375,
0.05413818359375,
-0.004283905029296875,
0.02294921875,
-0.0452880859375,
0.041717529296875,
-0.004497528076171875,
-0.037872314453125,
-0.0163726806640625,
-0.067138671875,
-0.071044921875,
0.0360107421875,
-0.0147857666015625,
-0.061981201171875,
0.01397705078125,
0.00821685791015625,
-0.0189971923828125,
-0.0032253265380859375,
-0.0780029296875,
0.07025146484375,
-0.01334381103515625,
-0.028472900390625,
0.01006317138671875,
-0.06329345703125,
0.031463623046875,
0.0186920166015625,
0.00832366943359375,
-0.01328277587890625,
0.000438690185546875,
0.0635986328125,
-0.04864501953125,
0.05072021484375,
-0.0211944580078125,
-0.01497650146484375,
0.0262298583984375,
-0.011871337890625,
0.038726806640625,
-0.002567291259765625,
0.0110321044921875,
0.0311737060546875,
-0.0031871795654296875,
-0.0287628173828125,
-0.046661376953125,
0.050384521484375,
-0.07073974609375,
-0.0308380126953125,
-0.0372314453125,
-0.0484619140625,
-0.01007843017578125,
0.0157928466796875,
0.0196075439453125,
0.0191497802734375,
0.0241241455078125,
0.0238037109375,
0.03643798828125,
-0.021636962890625,
0.02435302734375,
0.035308837890625,
-0.01375579833984375,
-0.043609619140625,
0.058929443359375,
0.0025463104248046875,
0.035308837890625,
0.005115509033203125,
0.0391845703125,
-0.03790283203125,
-0.02716064453125,
-0.02716064453125,
0.039947509765625,
-0.0238037109375,
-0.011627197265625,
-0.04315185546875,
-0.035369873046875,
-0.049285888671875,
0.0023059844970703125,
-0.033843994140625,
-0.0297393798828125,
-0.018707275390625,
0.022003173828125,
0.02862548828125,
0.0265350341796875,
0.0031261444091796875,
0.034881591796875,
-0.053497314453125,
0.01485443115234375,
0.006610870361328125,
0.025634765625,
-0.012420654296875,
-0.0654296875,
-0.01253509521484375,
0.0271148681640625,
-0.0187225341796875,
-0.0665283203125,
0.045654296875,
-0.002880096435546875,
0.048248291015625,
0.004550933837890625,
-0.001712799072265625,
0.0562744140625,
-0.004833221435546875,
0.06072998046875,
0.022369384765625,
-0.050201416015625,
0.035858154296875,
-0.04925537109375,
0.037994384765625,
0.0272674560546875,
0.043487548828125,
-0.0270233154296875,
-0.039764404296875,
-0.0751953125,
-0.08624267578125,
0.05126953125,
0.052734375,
0.0006909370422363281,
0.0235137939453125,
0.0225372314453125,
-0.00998687744140625,
0.01543426513671875,
-0.071044921875,
-0.021575927734375,
-0.020050048828125,
-0.0247650146484375,
0.0174560546875,
0.001247406005859375,
-0.006626129150390625,
-0.03277587890625,
0.0625,
-0.0115203857421875,
0.033721923828125,
0.00598907470703125,
0.0010499954223632812,
-0.0010652542114257812,
-0.004856109619140625,
0.031158447265625,
0.0283050537109375,
-0.0357666015625,
-0.031036376953125,
-0.005680084228515625,
-0.036041259765625,
-0.0230712890625,
0.040924072265625,
-0.01180267333984375,
-0.00789642333984375,
0.0236358642578125,
0.0648193359375,
0.0248565673828125,
-0.026214599609375,
0.0287933349609375,
-0.01959228515625,
-0.0258026123046875,
-0.013092041015625,
0.027679443359375,
0.00585174560546875,
0.0171051025390625,
0.01534271240234375,
-0.004077911376953125,
0.0155792236328125,
-0.032989501953125,
0.00661468505859375,
0.02667236328125,
-0.0048065185546875,
-0.018798828125,
0.047210693359375,
0.01290130615234375,
0.0012636184692382812,
0.0562744140625,
-0.0176544189453125,
-0.0259552001953125,
0.0478515625,
0.035003662109375,
0.0692138671875,
-0.0194244384765625,
-0.0005755424499511719,
0.04034423828125,
0.008941650390625,
-0.0069732666015625,
0.006786346435546875,
-0.0020904541015625,
-0.05975341796875,
-0.0310821533203125,
-0.05157470703125,
-0.0262908935546875,
0.0352783203125,
-0.051849365234375,
0.0213165283203125,
-0.020599365234375,
-0.0259246826171875,
0.00032973289489746094,
0.0128631591796875,
-0.045013427734375,
0.024871826171875,
0.005214691162109375,
0.039764404296875,
-0.07952880859375,
0.05242919921875,
0.060546875,
-0.059814453125,
-0.08013916015625,
0.00482940673828125,
0.0054473876953125,
-0.049896240234375,
0.00457000732421875,
0.00830078125,
0.016815185546875,
-0.0021305084228515625,
-0.042144775390625,
-0.07275390625,
0.08819580078125,
0.024871826171875,
-0.024932861328125,
-0.0196685791015625,
-0.0125732421875,
0.036376953125,
-0.0202789306640625,
0.047454833984375,
0.03759765625,
0.032257080078125,
0.033172607421875,
-0.09716796875,
0.00727081298828125,
-0.031341552734375,
0.0263214111328125,
0.0030193328857421875,
-0.07080078125,
0.07757568359375,
0.0022449493408203125,
-0.00048351287841796875,
0.005649566650390625,
0.049285888671875,
0.0275115966796875,
-0.005619049072265625,
0.0491943359375,
0.0487060546875,
0.028289794921875,
-0.0067291259765625,
0.08575439453125,
-0.023834228515625,
0.054931640625,
0.0830078125,
-0.002834320068359375,
0.038665771484375,
0.01427459716796875,
-0.034942626953125,
0.033355712890625,
0.05694580078125,
-0.00559234619140625,
0.0240478515625,
-0.0007390975952148438,
-0.010223388671875,
-0.00226593017578125,
-0.003833770751953125,
-0.042510986328125,
0.0325927734375,
0.0162506103515625,
-0.0271148681640625,
-0.01486968994140625,
-0.0193328857421875,
0.031890869140625,
-0.0259552001953125,
-0.00386810302734375,
0.059417724609375,
0.004398345947265625,
-0.055908203125,
0.05780029296875,
-0.0020751953125,
0.06610107421875,
-0.0626220703125,
-0.00563812255859375,
-0.0252532958984375,
-0.00014543533325195312,
-0.006473541259765625,
-0.059417724609375,
0.0175628662109375,
-0.021026611328125,
-0.00823974609375,
-0.020538330078125,
0.057220458984375,
-0.0537109375,
-0.0439453125,
0.025146484375,
0.027587890625,
0.0220489501953125,
-0.0027790069580078125,
-0.09307861328125,
0.01904296875,
-0.00015282630920410156,
-0.03912353515625,
0.0372314453125,
0.0184173583984375,
0.0016689300537109375,
0.05072021484375,
0.052886962890625,
0.004154205322265625,
-0.0137939453125,
0.0161285400390625,
0.06988525390625,
-0.031280517578125,
-0.04058837890625,
-0.05352783203125,
0.06011962890625,
-0.0174560546875,
-0.0275421142578125,
0.06591796875,
0.03729248046875,
0.06658935546875,
0.0003025531768798828,
0.052459716796875,
-0.030853271484375,
0.0227203369140625,
-0.0341796875,
0.0533447265625,
-0.05596923828125,
-0.0008611679077148438,
-0.0362548828125,
-0.07562255859375,
0.000034689903259277344,
0.061065673828125,
-0.0123443603515625,
0.048095703125,
0.036102294921875,
0.06402587890625,
-0.001434326171875,
-0.0051116943359375,
0.0161895751953125,
0.0254364013671875,
0.0185546875,
0.04010009765625,
0.034423828125,
-0.049224853515625,
0.0294036865234375,
-0.033599853515625,
-0.03375244140625,
-0.0183258056640625,
-0.0634765625,
-0.05767822265625,
-0.046142578125,
-0.041259765625,
-0.0457763671875,
-0.002330780029296875,
0.066650390625,
0.067138671875,
-0.04400634765625,
-0.0255889892578125,
0.0027866363525390625,
0.0004944801330566406,
-0.01067352294921875,
-0.0259246826171875,
0.007167816162109375,
0.001483917236328125,
-0.0736083984375,
0.00872039794921875,
0.003986358642578125,
0.030181884765625,
-0.021209716796875,
-0.00927734375,
-0.019287109375,
-0.005207061767578125,
0.03753662109375,
0.01102447509765625,
-0.050994873046875,
-0.01107025146484375,
-0.0201263427734375,
-0.0005736351013183594,
-0.0018062591552734375,
0.0335693359375,
-0.04144287109375,
0.0282135009765625,
0.042236328125,
0.0207061767578125,
0.040435791015625,
0.014984130859375,
0.050567626953125,
-0.047607421875,
0.0158233642578125,
0.01264190673828125,
0.0263824462890625,
0.03314208984375,
-0.035919189453125,
0.04034423828125,
0.0361328125,
-0.045501708984375,
-0.04351806640625,
-0.001953125,
-0.093994140625,
-0.01302337646484375,
0.095703125,
-0.01107025146484375,
-0.024444580078125,
0.01479339599609375,
-0.022552490234375,
0.0234222412109375,
-0.04425048828125,
0.0408935546875,
0.052734375,
-0.025634765625,
-0.023345947265625,
-0.0419921875,
0.029022216796875,
0.0251007080078125,
-0.055419921875,
-0.0160369873046875,
0.043182373046875,
0.033203125,
0.0058135986328125,
0.047149658203125,
-0.0151519775390625,
-0.0007948875427246094,
-0.0027561187744140625,
0.00775909423828125,
-0.0033168792724609375,
-0.0006303787231445312,
-0.0198822021484375,
-0.00958251953125,
-0.018218994140625,
-0.00986480712890625
]
] |
ddobokki/Llama-2-70b-orca-200k | 2023-08-08T00:15:50.000Z | [
"transformers",
"pytorch",
"llama",
"text-generation",
"llama-2",
"instruct",
"instruction",
"en",
"endpoints_compatible",
"has_space",
"text-generation-inference",
"region:us"
] | text-generation | ddobokki | null | null | ddobokki/Llama-2-70b-orca-200k | 6 | 5,932 | transformers | 2023-08-03T01:23:33 | ---
language:
- en
tags:
- llama-2
- instruct
- instruction
pipeline_tag: text-generation
---
# Llama-2-70b-orca-200k model card
### Used Datasets
- OpenOrca (200k sampling)
### Prompt Template
```
### Human: {Human}
### Assistant: {Assistant}
```
### Contribute
[ddobokki](https://github.com/ddobokki)
[YooSungHyun](https://github.com/YooSungHyun)
### License
[LICENSE.txt](meta-license/LICENSE.txt)
### USE_POLICY
[USE_POLICY.md](meta-license/USE_POLICY.md)
### Responsible Use Guide
[Responsible-Use-Guide.pdf](meta-license/Responsible-Use-Guide.pdf) | 561 | [
[
-0.0212249755859375,
-0.0211944580078125,
0.0250396728515625,
0.030426025390625,
-0.049713134765625,
-0.004863739013671875,
0.028228759765625,
-0.01158905029296875,
0.0256195068359375,
0.046844482421875,
-0.04290771484375,
-0.06640625,
-0.0273284912109375,
0.013092041015625,
-0.00019598007202148438,
0.058837890625,
-0.00836944580078125,
0.00954437255859375,
-0.016876220703125,
-0.0224609375,
-0.045196533203125,
-0.0225982666015625,
-0.07806396484375,
-0.0262908935546875,
0.042144775390625,
0.04730224609375,
0.028961181640625,
0.04217529296875,
0.0211944580078125,
0.01200103759765625,
-0.01419830322265625,
0.00925445556640625,
-0.0185089111328125,
-0.0014171600341796875,
-0.0163421630859375,
-0.047332763671875,
-0.09075927734375,
-0.0013599395751953125,
0.01509857177734375,
0.00913238525390625,
-0.028167724609375,
0.0438232421875,
-0.0037994384765625,
0.019256591796875,
-0.040374755859375,
0.044921875,
-0.0205230712890625,
-0.0019254684448242188,
-0.0413818359375,
-0.015716552734375,
0.001171112060546875,
-0.032684326171875,
-0.0244140625,
-0.08416748046875,
-0.00919342041015625,
0.00873565673828125,
0.058746337890625,
0.0298309326171875,
-0.027679443359375,
-0.02142333984375,
-0.007801055908203125,
0.0382080078125,
-0.059234619140625,
-0.005687713623046875,
0.06439208984375,
0.035308837890625,
-0.0212554931640625,
-0.065185546875,
-0.060028076171875,
-0.002605438232421875,
0.0241241455078125,
0.018341064453125,
-0.0184173583984375,
-0.01403045654296875,
0.0338134765625,
0.01123809814453125,
-0.01549530029296875,
0.032257080078125,
-0.06695556640625,
-0.0115814208984375,
0.0555419921875,
0.038330078125,
0.00791168212890625,
-0.00011098384857177734,
-0.01125335693359375,
-0.034027099609375,
-0.04132080078125,
0.00850677490234375,
0.053131103515625,
0.0009756088256835938,
-0.052001953125,
0.053466796875,
-0.0185089111328125,
0.04241943359375,
-0.01739501953125,
-0.02618408203125,
0.047760009765625,
-0.01287078857421875,
-0.0335693359375,
-0.017730712890625,
0.04083251953125,
0.0237274169921875,
0.0067291259765625,
0.035247802734375,
0.0003840923309326172,
-0.0156707763671875,
0.01093292236328125,
-0.025604248046875,
-0.01399993896484375,
0.028076171875,
-0.047088623046875,
-0.037841796875,
0.026336669921875,
-0.07220458984375,
-0.039642333984375,
-0.0087738037109375,
-0.004901885986328125,
0.01412200927734375,
-0.03167724609375,
0.0167388916015625,
-0.0086822509765625,
0.044921875,
0.0234222412109375,
-0.046142578125,
0.02783203125,
0.0301361083984375,
0.041290283203125,
0.00849151611328125,
-0.00826263427734375,
-0.0042724609375,
0.0293121337890625,
-0.0216827392578125,
0.054901123046875,
0.0026149749755859375,
-0.062408447265625,
-0.0089111328125,
0.00931549072265625,
0.017303466796875,
-0.031524658203125,
0.0709228515625,
-0.019256591796875,
0.010894775390625,
-0.044036865234375,
0.00731658935546875,
-0.03729248046875,
0.006500244140625,
-0.0748291015625,
0.0946044921875,
0.0143890380859375,
-0.043426513671875,
0.0256195068359375,
-0.09259033203125,
-0.0124969482421875,
0.01216888427734375,
-0.00511932373046875,
-0.0225830078125,
0.0003120899200439453,
-0.002227783203125,
0.031768798828125,
-0.0217132568359375,
0.0103607177734375,
-0.016693115234375,
-0.00821685791015625,
0.009979248046875,
-0.00827789306640625,
0.08856201171875,
0.04327392578125,
0.0035877227783203125,
0.01026153564453125,
-0.07568359375,
-0.024169921875,
0.044647216796875,
-0.05078125,
-0.0006570816040039062,
-0.031341552734375,
0.032562255859375,
-0.006008148193359375,
0.036468505859375,
-0.044708251953125,
0.053955078125,
0.005870819091796875,
0.0280609130859375,
0.060272216796875,
-0.00807952880859375,
0.017364501953125,
-0.023773193359375,
0.06475830078125,
-0.00490570068359375,
0.0146026611328125,
0.0160369873046875,
-0.049163818359375,
-0.047332763671875,
-0.025848388671875,
0.006389617919921875,
0.052093505859375,
-0.0261993408203125,
0.05352783203125,
-0.001705169677734375,
-0.048736572265625,
-0.0301055908203125,
-0.0014495849609375,
-0.00046825408935546875,
0.045379638671875,
0.021942138671875,
-0.0355224609375,
-0.050994873046875,
-0.06243896484375,
0.0347900390625,
-0.0195465087890625,
-0.01165771484375,
0.041351318359375,
0.060394287109375,
-0.03350830078125,
0.057403564453125,
-0.0628662109375,
-0.03643798828125,
-0.01959228515625,
-0.00873565673828125,
0.0226898193359375,
0.0280914306640625,
0.049346923828125,
-0.038970947265625,
-0.0180816650390625,
-0.030517578125,
-0.0501708984375,
-0.01904296875,
0.0143280029296875,
-0.0189208984375,
0.0193328857421875,
0.0197296142578125,
-0.032867431640625,
0.0628662109375,
0.0301055908203125,
-0.032012939453125,
0.0293731689453125,
-0.028778076171875,
-0.007091522216796875,
-0.09375,
0.03765869140625,
-0.0235595703125,
-0.0102081298828125,
-0.00170135498046875,
0.0007505416870117188,
-0.01486968994140625,
-0.00875091552734375,
-0.043365478515625,
0.06085205078125,
-0.016265869140625,
0.0033416748046875,
-0.00949859619140625,
-0.022735595703125,
0.0095672607421875,
0.0174713134765625,
-0.0079803466796875,
0.0504150390625,
0.051788330078125,
-0.037841796875,
0.05169677734375,
0.04376220703125,
-0.0089569091796875,
0.053375244140625,
-0.0665283203125,
0.0248565673828125,
-0.00858306884765625,
0.053619384765625,
-0.0767822265625,
-0.049163818359375,
0.06280517578125,
-0.0114288330078125,
0.0066070556640625,
0.00778961181640625,
-0.0677490234375,
-0.03564453125,
-0.0272674560546875,
0.04534912109375,
0.041778564453125,
-0.0323486328125,
0.044281005859375,
0.0302886962890625,
0.01178741455078125,
-0.03741455078125,
-0.0726318359375,
0.00372314453125,
-0.0196990966796875,
-0.0234222412109375,
0.017822265625,
-0.01177215576171875,
-0.007457733154296875,
-0.0142059326171875,
0.01363372802734375,
-0.01131439208984375,
-0.0081787109375,
0.0361328125,
0.0196380615234375,
0.0055999755859375,
-0.00482177734375,
-0.00281524658203125,
-0.032073974609375,
-0.0012083053588867188,
0.019622802734375,
0.0460205078125,
0.018280029296875,
-0.02178955078125,
-0.047698974609375,
0.01444244384765625,
0.03656005859375,
-0.00830841064453125,
0.04095458984375,
0.031097412109375,
-0.0240020751953125,
0.003360748291015625,
-0.033111572265625,
0.016632080078125,
-0.0301971435546875,
0.0272216796875,
-0.031097412109375,
-0.0194091796875,
0.0650634765625,
0.006061553955078125,
0.013702392578125,
0.04791259765625,
0.0450439453125,
-0.00414276123046875,
0.03155517578125,
0.032806396484375,
-0.025604248046875,
0.0292816162109375,
-0.029052734375,
-0.0197296142578125,
-0.07244873046875,
-0.04388427734375,
-0.053070068359375,
-0.018707275390625,
-0.0281982421875,
-0.024627685546875,
0.0132904052734375,
0.0016069412231445312,
-0.02886962890625,
0.04437255859375,
-0.041015625,
0.0224761962890625,
0.051239013671875,
0.02581787109375,
0.01617431640625,
-0.01192474365234375,
0.0013103485107421875,
0.033050537109375,
-0.042510986328125,
-0.051910400390625,
0.10662841796875,
0.03851318359375,
0.06256103515625,
-0.00025081634521484375,
0.055206298828125,
0.0140838623046875,
0.006072998046875,
-0.0211029052734375,
0.0540771484375,
0.0032558441162109375,
-0.05615234375,
-0.004001617431640625,
-0.0004963874816894531,
-0.0875244140625,
-0.0171051025390625,
0.0007767677307128906,
-0.047149658203125,
0.034637451171875,
0.0189971923828125,
-0.0218963623046875,
0.0288848876953125,
-0.026275634765625,
0.059906005859375,
-0.00858306884765625,
0.01708984375,
-0.01497650146484375,
-0.0316162109375,
0.022918701171875,
0.006317138671875,
0.00390625,
-0.0104522705078125,
-0.040679931640625,
0.08087158203125,
-0.02191162109375,
0.07293701171875,
-0.01995849609375,
-0.014892578125,
0.038116455078125,
-0.024169921875,
0.035308837890625,
0.0214996337890625,
-0.004901885986328125,
0.03125,
0.00832366943359375,
-0.040679931640625,
0.00174713134765625,
0.03631591796875,
-0.06793212890625,
-0.0109710693359375,
-0.044403076171875,
-0.032867431640625,
0.0266876220703125,
0.0018825531005859375,
0.0294342041015625,
0.00017833709716796875,
-0.0080108642578125,
-0.00028967857360839844,
0.034637451171875,
-0.031341552734375,
0.0158538818359375,
0.034637451171875,
-0.00931549072265625,
-0.044830322265625,
0.04443359375,
0.0091552734375,
-0.00759124755859375,
0.0004410743713378906,
-0.00434112548828125,
-0.0274505615234375,
-0.03533935546875,
-0.0280609130859375,
0.020477294921875,
-0.05499267578125,
-0.033660888671875,
-0.0285186767578125,
0.00528717041015625,
-0.03271484375,
-0.00531768798828125,
-0.01551055908203125,
-0.03082275390625,
-0.071044921875,
-0.03497314453125,
0.0489501953125,
0.042633056640625,
-0.033721923828125,
0.049713134765625,
-0.0233154296875,
0.03326416015625,
0.015960693359375,
0.01983642578125,
0.007503509521484375,
-0.041900634765625,
0.0055999755859375,
-0.01314544677734375,
-0.03912353515625,
-0.06951904296875,
0.0232391357421875,
-0.015472412109375,
0.05902099609375,
0.0224456787109375,
-0.00989532470703125,
0.04833984375,
0.0031528472900390625,
0.0582275390625,
0.0030155181884765625,
-0.048919677734375,
0.033905029296875,
-0.0310821533203125,
0.0174560546875,
0.06658935546875,
0.0222930908203125,
-0.01406097412109375,
0.01442718505859375,
-0.06256103515625,
-0.06976318359375,
0.041473388671875,
0.00394439697265625,
0.0224456787109375,
0.0167388916015625,
0.028961181640625,
0.00933837890625,
0.0224456787109375,
-0.072509765625,
-0.04034423828125,
-0.0165252685546875,
-0.0207977294921875,
0.0242919921875,
-0.03240966796875,
-0.0224761962890625,
-0.01383209228515625,
0.04730224609375,
0.0074005126953125,
0.022186279296875,
-0.00032591819763183594,
0.0256805419921875,
-0.01482391357421875,
0.00647735595703125,
0.061737060546875,
0.0599365234375,
-0.033416748046875,
-0.0232086181640625,
0.0229644775390625,
-0.058746337890625,
0.0000737309455871582,
0.003536224365234375,
-0.011810302734375,
-0.0226287841796875,
0.032470703125,
0.0400390625,
-0.01399993896484375,
-0.0296783447265625,
0.008056640625,
-0.00313568115234375,
-0.0151214599609375,
-0.04913330078125,
0.0209503173828125,
0.0115966796875,
0.00850677490234375,
0.032257080078125,
0.006992340087890625,
-0.0036029815673828125,
-0.03265380859375,
-0.01189422607421875,
0.01145172119140625,
-0.0014066696166992188,
-0.012359619140625,
0.0576171875,
0.0205841064453125,
-0.03240966796875,
0.036102294921875,
-0.0065765380859375,
-0.0070037841796875,
0.07666015625,
0.02154541015625,
0.05792236328125,
-0.01137542724609375,
-0.01107025146484375,
0.040008544921875,
0.0097808837890625,
-0.01495361328125,
0.0706787109375,
0.0027942657470703125,
-0.037078857421875,
0.005504608154296875,
-0.02423095703125,
-0.03662109375,
0.0201263427734375,
-0.07000732421875,
0.0418701171875,
-0.052459716796875,
-0.00859832763671875,
-0.01186370849609375,
0.00901031494140625,
-0.064208984375,
0.01303863525390625,
-0.004375457763671875,
0.1014404296875,
-0.058013916015625,
0.07562255859375,
0.0509033203125,
-0.059112548828125,
-0.070556640625,
-0.035980224609375,
0.013946533203125,
-0.074951171875,
0.0498046875,
-0.002201080322265625,
0.0223388671875,
0.00395965576171875,
-0.0753173828125,
-0.050201416015625,
0.12237548828125,
0.01446533203125,
-0.03289794921875,
0.0280914306640625,
-0.011322021484375,
0.021270751953125,
-0.03533935546875,
0.01922607421875,
0.032989501953125,
0.054656982421875,
0.023712158203125,
-0.06854248046875,
0.003620147705078125,
-0.0010128021240234375,
-0.019073486328125,
-0.0290069580078125,
-0.048675537109375,
0.061279296875,
-0.0079193115234375,
0.00347137451171875,
0.0374755859375,
0.049713134765625,
0.0579833984375,
0.0303497314453125,
0.03375244140625,
0.0592041015625,
0.053009033203125,
0.001224517822265625,
0.0654296875,
0.01488494873046875,
0.01971435546875,
0.0721435546875,
-0.0228271484375,
0.046905517578125,
0.0294189453125,
-0.034637451171875,
0.0523681640625,
0.061279296875,
-0.032470703125,
0.034271240234375,
0.0211029052734375,
-0.00771331787109375,
0.013458251953125,
-0.00902557373046875,
-0.052581787109375,
0.0330810546875,
0.0171966552734375,
-0.0296630859375,
-0.009368896484375,
-0.01470947265625,
0.024139404296875,
-0.0236053466796875,
-0.03143310546875,
0.05657958984375,
0.00571441650390625,
-0.014923095703125,
0.020172119140625,
0.0267333984375,
0.04266357421875,
-0.062042236328125,
-0.024627685546875,
-0.018218994140625,
-0.0006837844848632812,
-0.030364990234375,
-0.04742431640625,
0.0191650390625,
-0.00217437744140625,
-0.01503753662109375,
0.0021800994873046875,
0.035400390625,
0.003238677978515625,
-0.049407958984375,
0.0149993896484375,
0.006130218505859375,
0.033538818359375,
0.0177001953125,
-0.06011962890625,
0.0202789306640625,
0.01146697998046875,
-0.0220947265625,
0.01617431640625,
0.01509857177734375,
-0.01357269287109375,
0.04998779296875,
0.06451416015625,
0.017852783203125,
-0.00946807861328125,
0.0002999305725097656,
0.07244873046875,
-0.04632568359375,
-0.043060302734375,
-0.056121826171875,
0.056121826171875,
-0.00469207763671875,
-0.07073974609375,
0.0303497314453125,
0.047576904296875,
0.0584716796875,
-0.0266876220703125,
0.043365478515625,
-0.0256195068359375,
0.02471923828125,
-0.0255889892578125,
0.055908203125,
-0.033416748046875,
0.00634002685546875,
-0.01309967041015625,
-0.0655517578125,
-0.0309295654296875,
0.06634521484375,
-0.00267791748046875,
0.007419586181640625,
0.0260467529296875,
0.063232421875,
-0.01068115234375,
-0.0199737548828125,
0.00865936279296875,
-0.0020160675048828125,
0.025909423828125,
0.039215087890625,
0.0211639404296875,
-0.037567138671875,
0.04473876953125,
-0.0132598876953125,
-0.0268707275390625,
-0.007320404052734375,
-0.0791015625,
-0.057403564453125,
-0.0225830078125,
-0.01788330078125,
-0.032196044921875,
-0.0295562744140625,
0.06500244140625,
0.06396484375,
-0.06500244140625,
-0.040496826171875,
0.0013666152954101562,
0.004077911376953125,
0.0018129348754882812,
-0.0124053955078125,
0.04376220703125,
0.029998779296875,
-0.047454833984375,
0.006099700927734375,
-0.01030731201171875,
0.0267486572265625,
-0.03582763671875,
-0.01457977294921875,
-0.01470947265625,
-0.00677490234375,
0.002288818359375,
0.04815673828125,
-0.05316162109375,
0.007221221923828125,
-0.022552490234375,
-0.015899658203125,
0.00896453857421875,
0.0361328125,
-0.04644775390625,
0.018402099609375,
0.047149658203125,
-0.0002779960632324219,
0.0276336669921875,
0.017852783203125,
0.024261474609375,
-0.03948974609375,
0.043304443359375,
0.004024505615234375,
0.036163330078125,
0.00647735595703125,
-0.0352783203125,
0.06781005859375,
0.01467132568359375,
-0.034454345703125,
-0.052001953125,
0.0025959014892578125,
-0.11090087890625,
0.00516510009765625,
0.06475830078125,
-0.01361083984375,
-0.0200653076171875,
-0.00860595703125,
-0.059326171875,
0.022735595703125,
-0.05633544921875,
0.035797119140625,
0.0372314453125,
0.0009646415710449219,
-0.00803375244140625,
-0.047149658203125,
0.01708984375,
-0.01226806640625,
-0.08758544921875,
-0.034637451171875,
0.0198974609375,
0.0185089111328125,
0.01105499267578125,
0.057464599609375,
-0.00803375244140625,
0.01332855224609375,
-0.0038852691650390625,
0.031646728515625,
-0.052703857421875,
0.0012273788452148438,
-0.02471923828125,
-0.00852203369140625,
-0.0006284713745117188,
-0.054962158203125
]
] |
Fredithefish/Guanaco-7B-Uncensored | 2023-09-04T17:09:27.000Z | [
"transformers",
"pytorch",
"llama",
"text-generation",
"conversational",
"en",
"dataset:Fredithefish/openassistant-guanaco-unfiltered",
"license:apache-2.0",
"has_space",
"text-generation-inference",
"region:us"
] | conversational | Fredithefish | null | null | Fredithefish/Guanaco-7B-Uncensored | 6 | 5,932 | transformers | 2023-09-04T17:03:32 | ---
license: apache-2.0
datasets:
- Fredithefish/openassistant-guanaco-unfiltered
language:
- en
library_name: transformers
pipeline_tag: conversational
inference: false
---
<img src="https://huggingface.co/Fredithefish/Guanaco-3B-Uncensored/resolve/main/Guanaco-Uncensored.jpg" alt="Alt Text" width="295"/>
# ✨ Guanaco - 7B - Uncensored ✨
Guanaco-7B-Uncensored has been fine-tuned for 4 epochs on the [Unfiltered Guanaco Dataset.](https://huggingface.co/datasets/Fredithefish/openassistant-guanaco-unfiltered) using [Llama-2-7b](https://hf.co/meta-llama/Llama-2-7b-hf) as the base model.
<br>The model does not perform well with languages other than English.
<br>Please note: This model is designed to provide responses without content filtering or censorship. It generates answers without denials.
## Special thanks
I would like to thank AutoMeta for providing me with the computing power necessary to train this model.
### Prompt Template
```
### Human: {prompt} ### Assistant:
```
### Dataset
The model has been fine-tuned on the V2 of the Guanaco unfiltered dataset. | 1,079 | [
[
-0.022186279296875,
-0.0465087890625,
0.029083251953125,
0.02337646484375,
-0.07110595703125,
0.0015716552734375,
-0.0046539306640625,
-0.049835205078125,
0.017425537109375,
0.048553466796875,
-0.041534423828125,
-0.0560302734375,
-0.05706787109375,
0.03192138671875,
-0.00884246826171875,
0.0880126953125,
0.0165863037109375,
-0.007843017578125,
-0.00774383544921875,
-0.021575927734375,
-0.04351806640625,
-0.0255279541015625,
-0.044281005859375,
-0.03216552734375,
0.03863525390625,
0.0294952392578125,
0.05413818359375,
0.06475830078125,
0.028167724609375,
0.0207366943359375,
-0.0162200927734375,
0.032867431640625,
-0.048248291015625,
-0.00911712646484375,
-0.01049041748046875,
-0.0230255126953125,
-0.052490234375,
-0.00539398193359375,
0.03240966796875,
0.01025390625,
-0.031341552734375,
0.032958984375,
-0.0051422119140625,
0.032806396484375,
-0.040985107421875,
0.005863189697265625,
-0.05914306640625,
-0.0137786865234375,
-0.0318603515625,
-0.0007610321044921875,
-0.01108551025390625,
-0.0289459228515625,
-0.035369873046875,
-0.057464599609375,
0.006008148193359375,
0.01172637939453125,
0.08392333984375,
0.0304412841796875,
-0.033416748046875,
0.0019283294677734375,
-0.033843994140625,
0.040985107421875,
-0.0640869140625,
0.01427459716796875,
0.069091796875,
0.0245361328125,
-0.0205078125,
-0.059844970703125,
-0.0540771484375,
0.002765655517578125,
0.007762908935546875,
-0.00882720947265625,
-0.038177490234375,
-0.0164337158203125,
0.0031642913818359375,
0.026458740234375,
-0.040283203125,
0.034515380859375,
-0.055999755859375,
-0.0271148681640625,
0.055145263671875,
0.0171661376953125,
0.004261016845703125,
-0.01114654541015625,
-0.033721923828125,
-0.0009007453918457031,
-0.06597900390625,
-0.0006694793701171875,
0.068115234375,
-0.0004935264587402344,
-0.031829833984375,
0.03857421875,
-0.04400634765625,
0.0491943359375,
-0.00018894672393798828,
-0.00897979736328125,
0.035064697265625,
-0.0039825439453125,
-0.0295257568359375,
-0.0360107421875,
0.064453125,
0.0264434814453125,
0.0047149658203125,
0.0028171539306640625,
-0.01171112060546875,
0.00821685791015625,
0.019439697265625,
-0.058258056640625,
-0.01517486572265625,
0.0138702392578125,
-0.038055419921875,
-0.0297698974609375,
-0.0095062255859375,
-0.0531005859375,
-0.0184783935546875,
-0.0084075927734375,
0.0270538330078125,
0.003963470458984375,
-0.018829345703125,
0.01824951171875,
0.020294189453125,
0.01898193359375,
0.0230255126953125,
-0.0631103515625,
0.0167694091796875,
0.00762176513671875,
0.05194091796875,
0.01515960693359375,
0.003467559814453125,
-0.006561279296875,
-0.001224517822265625,
-0.01041412353515625,
0.0740966796875,
-0.029296875,
-0.02288818359375,
-0.01479339599609375,
0.025665283203125,
0.01369476318359375,
-0.03753662109375,
0.0687255859375,
-0.06005859375,
0.007366180419921875,
-0.009490966796875,
-0.0017566680908203125,
-0.04412841796875,
0.0009341239929199219,
-0.049835205078125,
0.06817626953125,
0.014251708984375,
-0.042266845703125,
0.01027679443359375,
-0.03155517578125,
0.004322052001953125,
-0.005138397216796875,
0.00771331787109375,
-0.042266845703125,
-0.018951416015625,
0.03192138671875,
0.0082244873046875,
-0.040283203125,
0.0235595703125,
-0.036102294921875,
-0.0309600830078125,
0.008758544921875,
-0.02789306640625,
0.06915283203125,
0.03521728515625,
-0.031982421875,
0.005626678466796875,
-0.05523681640625,
0.0037212371826171875,
0.0264434814453125,
-0.024444580078125,
-0.00710296630859375,
-0.007205963134765625,
-0.007274627685546875,
0.01181793212890625,
0.031402587890625,
-0.051605224609375,
0.0119781494140625,
-0.003406524658203125,
0.039459228515625,
0.0693359375,
-0.005184173583984375,
0.007770538330078125,
-0.0308685302734375,
0.038330078125,
-0.003131866455078125,
0.04437255859375,
0.01043701171875,
-0.051025390625,
-0.05902099609375,
-0.0289306640625,
0.0212554931640625,
0.029327392578125,
-0.0293731689453125,
0.036407470703125,
-0.0172119140625,
-0.051513671875,
-0.052490234375,
0.031646728515625,
0.024627685546875,
0.035247802734375,
0.04168701171875,
-0.047454833984375,
-0.04376220703125,
-0.08465576171875,
0.01152801513671875,
-0.01226806640625,
-0.0108489990234375,
0.0178680419921875,
0.039459228515625,
-0.0242156982421875,
0.03900146484375,
-0.033477783203125,
-0.030792236328125,
0.007843017578125,
-0.01270294189453125,
0.022430419921875,
0.0294647216796875,
0.0384521484375,
-0.0439453125,
-0.019927978515625,
0.003936767578125,
-0.07720947265625,
-0.024688720703125,
0.024169921875,
-0.03759765625,
0.00167083740234375,
0.0218505859375,
-0.03448486328125,
0.04791259765625,
0.04815673828125,
-0.0229644775390625,
0.01134490966796875,
-0.0159149169921875,
0.00328826904296875,
-0.0687255859375,
0.0159912109375,
-0.0186309814453125,
-0.0232086181640625,
-0.0216064453125,
0.014373779296875,
0.01326751708984375,
0.007328033447265625,
-0.0338134765625,
0.03656005859375,
-0.042236328125,
0.01239776611328125,
-0.035400390625,
-0.002353668212890625,
0.01067352294921875,
0.049072265625,
0.001194000244140625,
0.045257568359375,
0.0301361083984375,
-0.04364013671875,
0.0213623046875,
0.044708251953125,
-0.0249176025390625,
0.03057861328125,
-0.0718994140625,
0.05078125,
-0.01277923583984375,
0.041961669921875,
-0.052520751953125,
-0.0279693603515625,
0.0345458984375,
-0.037811279296875,
0.002147674560546875,
-0.0277099609375,
-0.032257080078125,
-0.0303955078125,
-0.031005859375,
0.0518798828125,
0.042022705078125,
-0.053558349609375,
0.0195770263671875,
0.03179931640625,
0.01033782958984375,
-0.05810546875,
-0.044158935546875,
0.0025043487548828125,
-0.03350830078125,
-0.04302978515625,
0.0110015869140625,
-0.0174407958984375,
0.004772186279296875,
-0.013671875,
0.008270263671875,
-0.01062774658203125,
-0.00699615478515625,
0.04193115234375,
0.03497314453125,
0.0055084228515625,
-0.0166778564453125,
0.0229034423828125,
0.00811767578125,
0.0010938644409179688,
-0.01270294189453125,
0.042633056640625,
-0.001850128173828125,
-0.00817108154296875,
-0.046875,
0.008636474609375,
0.0301666259765625,
-0.022247314453125,
0.058746337890625,
0.032135009765625,
-0.0310516357421875,
0.00909423828125,
-0.054779052734375,
0.004024505615234375,
-0.038909912109375,
0.0008211135864257812,
-0.0116424560546875,
-0.0638427734375,
0.0634765625,
0.0325927734375,
-0.00835418701171875,
0.043212890625,
0.051727294921875,
0.0034618377685546875,
0.0521240234375,
0.03515625,
-0.001987457275390625,
0.0275115966796875,
-0.0189361572265625,
-0.00664520263671875,
-0.08184814453125,
-0.0556640625,
-0.0394287109375,
-0.020477294921875,
-0.0496826171875,
-0.02105712890625,
0.0220489501953125,
0.0028743743896484375,
-0.04486083984375,
0.038787841796875,
-0.0374755859375,
0.03521728515625,
0.035919189453125,
0.040130615234375,
0.034088134765625,
0.0019779205322265625,
0.009857177734375,
0.006069183349609375,
-0.0287628173828125,
-0.04833984375,
0.09912109375,
0.02911376953125,
0.0704345703125,
0.032989501953125,
0.016571044921875,
0.033843994140625,
0.0184783935546875,
-0.02655029296875,
0.0265045166015625,
-0.0031147003173828125,
-0.0654296875,
0.00438690185546875,
-0.0162200927734375,
-0.08575439453125,
0.0249176025390625,
-0.0115509033203125,
-0.05303955078125,
0.027923583984375,
0.0084686279296875,
-0.0258331298828125,
0.0240936279296875,
-0.04315185546875,
0.03375244140625,
0.0017948150634765625,
-0.033721923828125,
-0.002254486083984375,
-0.056915283203125,
0.0291748046875,
-0.00502777099609375,
0.00949859619140625,
-0.031463623046875,
0.0048065185546875,
0.05255126953125,
-0.038970947265625,
0.0875244140625,
-0.00617218017578125,
-0.017364501953125,
0.045806884765625,
-0.01209259033203125,
0.0276947021484375,
0.033355712890625,
-0.005260467529296875,
0.049072265625,
-0.022796630859375,
-0.045623779296875,
-0.0098114013671875,
0.05206298828125,
-0.0789794921875,
-0.053985595703125,
-0.030487060546875,
-0.0199432373046875,
0.0007452964782714844,
0.012847900390625,
0.049407958984375,
0.005306243896484375,
-0.0016508102416992188,
0.00753021240234375,
0.03692626953125,
0.00250244140625,
0.03424072265625,
0.04443359375,
0.004669189453125,
-0.05419921875,
0.0292816162109375,
0.004444122314453125,
0.00043082237243652344,
0.004566192626953125,
-0.002384185791015625,
-0.044281005859375,
-0.047088623046875,
-0.0526123046875,
0.0330810546875,
-0.04718017578125,
-0.0584716796875,
-0.0380859375,
-0.0279998779296875,
-0.038726806640625,
0.021087646484375,
-0.00913238525390625,
-0.025238037109375,
-0.044189453125,
-0.032501220703125,
0.051788330078125,
0.0574951171875,
-0.02593994140625,
0.045806884765625,
-0.03125,
0.028839111328125,
0.0321044921875,
0.0200347900390625,
-0.0104827880859375,
-0.0860595703125,
-0.01352691650390625,
0.014892578125,
-0.0390625,
-0.05487060546875,
0.0345458984375,
0.0175018310546875,
0.039306640625,
0.022369384765625,
0.01016998291015625,
0.04150390625,
-0.0206146240234375,
0.041473388671875,
-0.008758544921875,
-0.055999755859375,
0.047698974609375,
-0.050140380859375,
0.0170440673828125,
0.04290771484375,
0.019927978515625,
-0.003917694091796875,
-0.0241241455078125,
-0.03631591796875,
-0.069580078125,
0.036407470703125,
0.034454345703125,
0.038787841796875,
0.00429534912109375,
0.0302581787109375,
0.02783203125,
0.0250091552734375,
-0.0797119140625,
-0.0109405517578125,
-0.046417236328125,
-0.0027313232421875,
0.0085906982421875,
-0.0233001708984375,
-0.017547607421875,
-0.026702880859375,
0.04974365234375,
-0.00969696044921875,
0.0291900634765625,
0.012939453125,
-0.0228271484375,
-0.0135498046875,
-0.004131317138671875,
0.057525634765625,
0.0450439453125,
-0.0225067138671875,
-0.005859375,
0.0025482177734375,
-0.062744140625,
0.006359100341796875,
0.005672454833984375,
-0.0215301513671875,
-0.024627685546875,
0.0198516845703125,
0.0911865234375,
-0.012786865234375,
-0.0227203369140625,
0.0230712890625,
-0.01020050048828125,
-0.00722503662109375,
-0.0283966064453125,
0.01456451416015625,
-0.0006780624389648438,
0.0235595703125,
0.0294952392578125,
-0.0134124755859375,
0.00641632080078125,
-0.0146636962890625,
-0.0011196136474609375,
0.0187530517578125,
0.0093994140625,
-0.034271240234375,
0.078857421875,
0.01320648193359375,
-0.010772705078125,
0.056640625,
-0.0111846923828125,
-0.01325225830078125,
0.048553466796875,
0.0592041015625,
0.036041259765625,
-0.028106689453125,
0.02130126953125,
0.056640625,
0.042327880859375,
0.0037784576416015625,
0.0257720947265625,
-0.0008063316345214844,
-0.04241943359375,
-0.00667572021484375,
-0.04443359375,
-0.0194091796875,
0.0440673828125,
-0.061187744140625,
0.012847900390625,
-0.047119140625,
-0.02130126953125,
-0.0207977294921875,
0.0098724365234375,
-0.063720703125,
0.03497314453125,
-0.00042176246643066406,
0.0587158203125,
-0.08184814453125,
0.07330322265625,
0.03741455078125,
-0.038726806640625,
-0.055328369140625,
-0.030120849609375,
-0.0095977783203125,
-0.079833984375,
0.0086212158203125,
0.01107025146484375,
-0.01519775390625,
0.007457733154296875,
-0.0712890625,
-0.07183837890625,
0.1064453125,
0.04345703125,
-0.024383544921875,
0.01922607421875,
-0.00786590576171875,
0.04815673828125,
-0.03466796875,
0.037506103515625,
0.0325927734375,
0.02801513671875,
-0.0033016204833984375,
-0.074951171875,
0.005672454833984375,
-0.041839599609375,
0.01226806640625,
-0.00968170166015625,
-0.087890625,
0.06597900390625,
-0.01209259033203125,
-0.0027618408203125,
0.0298004150390625,
0.0706787109375,
0.022613525390625,
0.0185546875,
0.04058837890625,
0.059600830078125,
0.0550537109375,
0.00344085693359375,
0.0596923828125,
0.0150146484375,
0.025054931640625,
0.0860595703125,
-0.0115966796875,
0.058929443359375,
0.0258331298828125,
-0.01329803466796875,
0.068115234375,
0.09027099609375,
-0.01273345947265625,
0.056976318359375,
0.006252288818359375,
-0.0247344970703125,
-0.0068359375,
-0.0294952392578125,
-0.040313720703125,
0.0546875,
0.00852203369140625,
-0.0145416259765625,
-0.0120391845703125,
-0.0129852294921875,
0.0206756591796875,
0.005115509033203125,
-0.02142333984375,
0.041351318359375,
-0.0007696151733398438,
-0.0299835205078125,
0.07391357421875,
0.002777099609375,
0.061981201171875,
-0.04815673828125,
0.0081329345703125,
-0.058380126953125,
-0.0180511474609375,
-0.0288543701171875,
-0.045501708984375,
0.00988006591796875,
0.02215576171875,
-0.006443023681640625,
0.02764892578125,
0.046051025390625,
-0.0165252685546875,
-0.031005859375,
0.0264129638671875,
0.0183868408203125,
0.026611328125,
0.01403045654296875,
-0.03857421875,
0.0173492431640625,
0.0079498291015625,
-0.0168914794921875,
0.023101806640625,
0.01509857177734375,
-0.040130615234375,
0.053985595703125,
0.063720703125,
-0.006076812744140625,
0.0086669921875,
-0.006786346435546875,
0.08258056640625,
-0.02325439453125,
-0.032073974609375,
-0.0477294921875,
0.035858154296875,
-0.002227783203125,
-0.050018310546875,
0.0379638671875,
0.025146484375,
0.06268310546875,
-0.0034465789794921875,
0.0357666015625,
-0.01544189453125,
0.01378631591796875,
-0.048797607421875,
0.0623779296875,
-0.0487060546875,
0.01727294921875,
0.00399017333984375,
-0.058563232421875,
-0.00820159912109375,
0.05450439453125,
0.01213836669921875,
0.01177978515625,
0.0438232421875,
0.0684814453125,
-0.01025390625,
-0.0133056640625,
0.016387939453125,
0.00792694091796875,
0.0163421630859375,
0.0379638671875,
0.04290771484375,
-0.04241943359375,
0.037261962890625,
-0.04083251953125,
-0.0099639892578125,
-0.0005402565002441406,
-0.07672119140625,
-0.0672607421875,
-0.04290771484375,
-0.006275177001953125,
-0.026275634765625,
0.00710296630859375,
0.039825439453125,
0.050567626953125,
-0.044097900390625,
-0.016357421875,
0.0238189697265625,
0.0107574462890625,
0.01102447509765625,
-0.006565093994140625,
0.01885986328125,
0.037841796875,
-0.06182861328125,
0.0282135009765625,
-0.010650634765625,
0.0218505859375,
-0.01320648193359375,
0.006740570068359375,
-0.0181732177734375,
0.00994110107421875,
0.021484375,
0.04193115234375,
-0.049560546875,
-0.03472900390625,
0.00463104248046875,
0.0012340545654296875,
0.02569580078125,
0.01506805419921875,
-0.054229736328125,
0.01806640625,
0.01898193359375,
0.0172119140625,
0.04345703125,
0.0157470703125,
0.02874755859375,
-0.040283203125,
0.04638671875,
-0.000732421875,
0.0254974365234375,
0.042144775390625,
-0.046051025390625,
0.059539794921875,
0.007442474365234375,
-0.059906005859375,
-0.0556640625,
0.006290435791015625,
-0.08026123046875,
0.004058837890625,
0.09039306640625,
-0.0169830322265625,
-0.02569580078125,
-0.00695037841796875,
-0.01369476318359375,
0.0345458984375,
-0.045867919921875,
0.068359375,
0.039459228515625,
0.0025196075439453125,
-0.00433349609375,
-0.053619384765625,
0.02056884765625,
0.0200347900390625,
-0.059600830078125,
-0.0305328369140625,
0.036956787109375,
0.042327880859375,
-0.01123809814453125,
0.06292724609375,
-0.02001953125,
0.004756927490234375,
-0.0222930908203125,
0.01378631591796875,
-0.0213470458984375,
-0.0197296142578125,
-0.031280517578125,
-0.018768310546875,
0.00879669189453125,
-0.038055419921875
]
] |
togethercomputer/Pythia-Chat-Base-7B | 2023-03-29T02:52:46.000Z | [
"transformers",
"pytorch",
"gpt_neox",
"text-generation",
"en",
"license:apache-2.0",
"endpoints_compatible",
"has_space",
"text-generation-inference",
"region:us"
] | text-generation | togethercomputer | null | null | togethercomputer/Pythia-Chat-Base-7B | 61 | 5,930 | transformers | 2023-03-22T02:03:05 | ---
license: apache-2.0
language:
- en
---
***<p style="font-size: 24px">Feel free to try out our [OpenChatKit feedback app](https://huggingface.co/spaces/togethercomputer/OpenChatKit)!</p>***
# Pythia-Chat-Base-7B-v0.16
> TLDR: As part of OpenChatKit (codebase available [here](https://github.com/togethercomputer/OpenChaT)),
> Pythia-Chat-Base-7B-v0.16 is a 7B parameter language model, fine-tuned from EleutherAI’s Pythia 7B with over 40 million instructions on 100% carbon negative compute.
Pythia-Chat-Base-7B-v0.16 is based on ElutherAI’s Pythia-7B model, and is fine-tuned with data focusing on dialog-style interactions.
We focused the tuning on several tasks such as question answering, classification, extraction, and summarization.
We’ve fine-tuned the model with a collection of 43 million high-quality instructions.
Together partnered with LAION and Ontocord.ai, who both helped curate the dataset the model is based on.
You can read more about this process and the availability of this dataset in LAION’s blog post [here](https://laion.ai/blog/oig-dataset/).
In addition to the aforementioned fine-tuning, Pythia-Chat-Base-7B-v0.16 has also undergone further fine-tuning via a small amount of feedback data.
This process allows the model to better adapt to human preferences in the conversations.
One of the notable features of Pythia-Chat-Base-7B-v0.16 is its ability to **run inference on a 12GB GPU**, thanks to the quantization technique.
It helps maintain the dialogue capabilities while making the model more accessible to a wider range of users and hardware configurations.
## Model Details
- **Developed by**: Together Computer.
- **Model type**: Language Model
- **Language(s)**: English
- **License**: Apache 2.0
- **Model Description**: A 7B parameter open source chat model, fine-tuned from EleutherAI’s Pythia with over 40M instructions on 100% carbon negative compute
- **Resources for more information**: [GitHub Repository](https://github.com/togethercomputer/OpenChaT).
# Quick Start
## GPU Inference
This requires a GPU with 24GB memory.
```python
from transformers import AutoTokenizer, AutoModelForCausalLM
# init
tokenizer = AutoTokenizer.from_pretrained("togethercomputer/Pythia-Chat-Base-7B-v0.16")
model = AutoModelForCausalLM.from_pretrained("togethercomputer/Pythia-Chat-Base-7B-v0.16", torch_dtype=torch.float16)
model = model.to('cuda:0')
# infer
inputs = tokenizer("<human>: Hello!\n<bot>:", return_tensors='pt').to(model.device)
outputs = model.generate(**inputs, max_new_tokens=10, do_sample=True, temperature=0.8)
output_str = tokenizer.decode(outputs[0])
print(output_str)
```
## GPU Inference in Int8
This requires a GPU with 12GB memory.
```python
from transformers import AutoTokenizer, AutoModelForCausalLM
# init
tokenizer = AutoTokenizer.from_pretrained("togethercomputer/Pythia-Chat-Base-7B-v0.16")
model = AutoModelForCausalLM.from_pretrained("togethercomputer/Pythia-Chat-Base-7B-v0.16", device_map="auto", load_in_8bit=True)
# infer
inputs = tokenizer("<human>: Hello!\n<bot>:", return_tensors='pt').to(model.device)
outputs = model.generate(**inputs, max_new_tokens=10, do_sample=True, temperature=0.8)
output_str = tokenizer.decode(outputs[0])
print(output_str)
```
## CPU Inference
```python
from transformers import AutoTokenizer, AutoModelForCausalLM
# init
tokenizer = AutoTokenizer.from_pretrained("togethercomputer/Pythia-Chat-Base-7B-v0.16")
model = AutoModelForCausalLM.from_pretrained("togethercomputer/Pythia-Chat-Base-7B-v0.16", torch_dtype=torch.bfloat16)
# infer
inputs = tokenizer("<human>: Hello!\n<bot>:", return_tensors='pt').to(model.device)
outputs = model.generate(**inputs, max_new_tokens=10, do_sample=True, temperature=0.8)
output_str = tokenizer.decode(outputs[0])
print(output_str)
```
## Strengths of the model
There are several tasks that OpenChatKit excels at out of the box. This includes:
- Summarization and question answering within context.
- Extraction.
- Classification.
In addition, the model does well on few-shot prompts. For both classification and extraction, the model performs even better with few shots, as in most HELM tasks. [Contact us](https://www.together.xyz/contact) if you’re interested in trying few-shot prompts with the model.
## Weaknesses of the model
That said, there are several areas where we have more work to do, and we need your help! Some of these include:
- Knowledge-based closed question and answering: The chatbot may hallucinate and give incorrect results. Be sure to fact check, and if possible provide feedback with the corrected information.
- Coding tasks: The chatbot was not trained on a large enough corpus of source code to excel at writing code. We welcome contributions of additional datasets to improve this!
- Repetition: Sometimes the chatbot will repeat its response. We’re working to improve this, but in the meantime you can click the refresh button to start a new conversation.
- Context switching: If you change the topic in the middle of a conversation the chatbot often cannot make the switch automatically and will continue to give answers related to the prior topic.
- Creative writing and longer answers: The chatbot does not generate long, creative text such as an essay or story.
We are excited to work with you to address these weaknesses by getting your feedback, bolstering data sets, and improving accuracy.
# Uses
## Direct Use
The model is intended for research purposes. Possible research areas and tasks include
- Safe deployment of models which have the potential to generate harmful content.
- Probing and understanding the limitations and biases of dialogue models or language models.
- Generation of artworks and use in design and other artistic processes.
- Applications in educational or creative tools.
- Research on dialogue models or language models.
Excluded uses are described below.
### Misuse, Malicious Use, and Out-of-Scope Use
The OpenChatKit community provides Pythia-Chat-Base-7B-v0.16 as an open source tool for building chatbots.
The community is not responsible for any misuse, malicious use, or out-of-scope use of the model.
It is the responsibility of the end user to ensure that the model is used in a responsible and ethical manner.
#### Out-of-Scope Use
Pythia-Chat-Base-7B-v0.16 is designed for use in chatbot applications and may not perform well for other use cases outside of its intended scope.
For example, it may not be suitable for use in safety-critical applications or for making decisions that have a significant impact on individuals or society.
It is important to consider the limitations of the model and to only use it for its intended purpose.
#### Misuse and Malicious Use
Pythia-Chat-Base-7B-v0.16 is designed for use in chatbot applications and should not be used for any other purpose.
Misuse of the model, such as using it to engage in illegal or unethical activities, is strictly prohibited and goes against the principles of the OpenChatKit community project.
Using the model to generate content that is cruel to individuals is a misuse of this model. This includes, but is not limited to:
- Generating fake news, misinformation, or propaganda
- Promoting hate speech, discrimination, or violence against individuals or groups
- Impersonating individuals or organizations without their consent
- Engaging in cyberbullying or harassment
- Defamatory content
- Spamming or scamming
- Sharing confidential or sensitive information without proper authorization
- Violating the terms of use of the model or the data used to train it
- Creating automated bots for malicious purposes such as spreading malware, phishing scams, or spamming
## Limitations
Pythia-Chat-Base-7B-v0.16, like other language model-based chatbots, has limitations that should be taken into consideration.
For example, the model may not always provide accurate or relevant answers, particularly for questions that are complex, ambiguous, or outside of its training data.
We therefore welcome contributions from individuals and organizations, and encourage collaboration towards creating a more robust and inclusive chatbot.
## Training
**Training Data**
Please refer to [togethercomputer/OpenDataHub](https://github.com/togethercomputer/OpenDataHub)
**Training Procedure**
- **Hardware:** 8 x A100 GPUs
- **Optimizer:** [8bit-AdamW](https://github.com/TimDettmers/bitsandbytes)
- **Gradient Accumulations**: 4
- **Batch:** 4 x 4 x 16 x 2048 = 524288 tokens
- **Learning rate:** warmup to 1e-5 for 100 steps and then kept constant
## Community
Join us on [Together Discord](https://discord.gg/6ZVDU8tTD4) | 8,647 | [
[
-0.017303466796875,
-0.08648681640625,
0.005016326904296875,
0.0123748779296875,
-0.0144500732421875,
-0.00257110595703125,
-0.0285797119140625,
-0.034637451171875,
0.0010738372802734375,
0.0201873779296875,
-0.02423095703125,
-0.0212249755859375,
-0.0255126953125,
-0.0290679931640625,
-0.00936126708984375,
0.06494140625,
0.01495361328125,
-0.0099029541015625,
-0.00914764404296875,
0.002109527587890625,
-0.04931640625,
-0.04730224609375,
-0.050567626953125,
-0.0186004638671875,
0.0211029052734375,
0.045440673828125,
0.0645751953125,
0.018951416015625,
0.029296875,
0.030181884765625,
-0.0099945068359375,
0.01349639892578125,
-0.055633544921875,
0.0023860931396484375,
0.005405426025390625,
-0.04193115234375,
-0.03472900390625,
-0.00179290771484375,
0.021575927734375,
0.03271484375,
-0.0121002197265625,
0.020538330078125,
-0.0014972686767578125,
0.01369476318359375,
-0.045745849609375,
0.030731201171875,
-0.038665771484375,
-0.015716552734375,
-0.00792694091796875,
-0.00034117698669433594,
-0.0418701171875,
-0.01323699951171875,
0.0164794921875,
-0.0439453125,
0.0036296844482421875,
0.0140838623046875,
0.0743408203125,
-0.0024356842041015625,
-0.01050567626953125,
-0.01812744140625,
-0.0526123046875,
0.054718017578125,
-0.06854248046875,
0.025604248046875,
0.031982421875,
0.0249786376953125,
-0.028533935546875,
-0.06268310546875,
-0.045806884765625,
-0.025299072265625,
-0.005435943603515625,
0.0024890899658203125,
-0.050872802734375,
0.005435943603515625,
0.0174102783203125,
0.04327392578125,
-0.035614013671875,
0.0016336441040039062,
-0.049591064453125,
-0.0277557373046875,
0.03717041015625,
0.0169677734375,
0.039581298828125,
-0.02484130859375,
0.004482269287109375,
-0.020965576171875,
-0.017913818359375,
0.01561737060546875,
0.0445556640625,
0.016448974609375,
-0.037872314453125,
0.04852294921875,
-0.0249481201171875,
0.032684326171875,
0.006683349609375,
0.0011348724365234375,
0.0217742919921875,
-0.022705078125,
-0.02520751953125,
-0.006221771240234375,
0.08489990234375,
0.03338623046875,
0.0072174072265625,
-0.001102447509765625,
0.0244140625,
0.004848480224609375,
0.006893157958984375,
-0.0654296875,
-0.05950927734375,
0.0225982666015625,
-0.04168701171875,
-0.034393310546875,
-0.01345062255859375,
-0.05859375,
-0.02362060546875,
0.015106201171875,
0.03192138671875,
-0.043121337890625,
-0.04962158203125,
-0.003231048583984375,
-0.01351165771484375,
0.00070953369140625,
0.0184326171875,
-0.07049560546875,
0.0237579345703125,
0.016845703125,
0.0750732421875,
0.005901336669921875,
-0.0199737548828125,
-0.0013132095336914062,
-0.026336669921875,
-0.01421356201171875,
0.0292510986328125,
-0.01995849609375,
-0.0245819091796875,
-0.033843994140625,
-0.0044097900390625,
-0.00006449222564697266,
-0.0227203369140625,
0.0194244384765625,
-0.0231475830078125,
0.031585693359375,
0.00344085693359375,
-0.03961181640625,
-0.018829345703125,
-0.006496429443359375,
-0.031494140625,
0.08233642578125,
0.00801849365234375,
-0.0643310546875,
-0.0008282661437988281,
-0.058380126953125,
-0.00743865966796875,
-0.0030994415283203125,
0.00960540771484375,
-0.03070068359375,
0.00701904296875,
0.0231170654296875,
0.01390838623046875,
-0.0328369140625,
0.016754150390625,
-0.032806396484375,
-0.022552490234375,
0.03179931640625,
-0.057769775390625,
0.07513427734375,
0.0193634033203125,
-0.03033447265625,
0.02740478515625,
-0.0298309326171875,
0.01367950439453125,
0.0180816650390625,
-0.00791168212890625,
0.004131317138671875,
-0.004428863525390625,
-0.00418853759765625,
0.0017423629760742188,
0.0173187255859375,
-0.03997802734375,
0.009979248046875,
-0.0498046875,
0.05218505859375,
0.057586669921875,
-0.003520965576171875,
0.04693603515625,
-0.0325927734375,
0.0259857177734375,
0.0204315185546875,
0.0306396484375,
-0.00994110107421875,
-0.07049560546875,
-0.06646728515625,
-0.0208892822265625,
0.01300048828125,
0.036956787109375,
-0.049530029296875,
0.042938232421875,
-0.0027866363525390625,
-0.07366943359375,
-0.03765869140625,
-0.004520416259765625,
0.025299072265625,
0.0311431884765625,
0.0220794677734375,
-0.041900634765625,
-0.037811279296875,
-0.0487060546875,
0.00281524658203125,
-0.0377197265625,
0.009185791015625,
0.04046630859375,
0.0433349609375,
0.0013380050659179688,
0.076171875,
-0.044158935546875,
0.0114898681640625,
-0.0281982421875,
0.00800323486328125,
0.02606201171875,
0.05126953125,
0.036773681640625,
-0.049957275390625,
-0.031280517578125,
-0.005222320556640625,
-0.061492919921875,
0.0013904571533203125,
0.0082855224609375,
-0.03668212890625,
0.016998291015625,
0.0270843505859375,
-0.07513427734375,
0.06646728515625,
0.0401611328125,
-0.0328369140625,
0.03948974609375,
-0.0118408203125,
0.0118255615234375,
-0.08270263671875,
0.0015201568603515625,
-0.01139068603515625,
-0.007045745849609375,
-0.0482177734375,
-0.01409912109375,
-0.0097198486328125,
-0.018951416015625,
-0.048828125,
0.061279296875,
-0.028350830078125,
0.0176849365234375,
-0.0140380859375,
0.0008916854858398438,
-0.0186004638671875,
0.053375244140625,
-0.01409912109375,
0.04876708984375,
0.0531005859375,
-0.05523681640625,
0.06329345703125,
0.03948974609375,
-0.0157012939453125,
0.019439697265625,
-0.06787109375,
0.0264129638671875,
0.00640106201171875,
0.0281829833984375,
-0.08367919921875,
-0.0270538330078125,
0.062042236328125,
-0.08001708984375,
0.005405426025390625,
-0.013946533203125,
-0.03594970703125,
-0.039306640625,
-0.00799560546875,
0.0292510986328125,
0.055145263671875,
-0.033294677734375,
0.052276611328125,
0.032470703125,
0.005306243896484375,
-0.03033447265625,
-0.03765869140625,
-0.0110931396484375,
-0.01666259765625,
-0.06463623046875,
0.00928497314453125,
-0.011383056640625,
-0.0222015380859375,
0.0001589059829711914,
-0.0023365020751953125,
-0.002429962158203125,
0.003509521484375,
0.036865234375,
0.025634765625,
-0.01073455810546875,
0.0031909942626953125,
-0.0226287841796875,
0.003662109375,
-0.00525665283203125,
-0.0178680419921875,
0.0806884765625,
-0.0163421630859375,
-0.013824462890625,
-0.062164306640625,
0.004657745361328125,
0.041961669921875,
-0.01016998291015625,
0.07293701171875,
0.04693603515625,
-0.0246734619140625,
0.00966644287109375,
-0.04412841796875,
-0.037078857421875,
-0.0384521484375,
0.04083251953125,
-0.0159149169921875,
-0.049774169921875,
0.0528564453125,
0.036895751953125,
0.0196380615234375,
0.032867431640625,
0.0611572265625,
0.0039215087890625,
0.08160400390625,
0.0362548828125,
-0.02117919921875,
0.05328369140625,
-0.0186309814453125,
0.0323486328125,
-0.045318603515625,
-0.01001739501953125,
-0.027923583984375,
-0.01885986328125,
-0.06134033203125,
-0.032470703125,
0.017791748046875,
0.0181884765625,
-0.0308380126953125,
0.02569580078125,
-0.043701171875,
0.0227203369140625,
0.06195068359375,
0.03082275390625,
-0.0014467239379882812,
-0.0036792755126953125,
0.01312255859375,
0.004024505615234375,
-0.0694580078125,
-0.0333251953125,
0.09149169921875,
0.04248046875,
0.05975341796875,
0.005847930908203125,
0.035064697265625,
-0.00649261474609375,
0.020965576171875,
-0.0474853515625,
0.040618896484375,
0.0262908935546875,
-0.065185546875,
-0.0274505615234375,
-0.05303955078125,
-0.0662841796875,
0.01416778564453125,
0.0025844573974609375,
-0.0916748046875,
-0.00559234619140625,
0.01568603515625,
-0.02532958984375,
0.03350830078125,
-0.07598876953125,
0.08673095703125,
-0.009185791015625,
-0.0250701904296875,
-0.018341064453125,
-0.048553466796875,
0.02392578125,
0.0390625,
0.0033111572265625,
-0.0136871337890625,
0.0323486328125,
0.0623779296875,
-0.04644775390625,
0.0877685546875,
-0.0091094970703125,
0.023712158203125,
0.048065185546875,
0.00650787353515625,
0.0149383544921875,
0.0086212158203125,
0.0158538818359375,
0.0213165283203125,
0.0029468536376953125,
-0.035186767578125,
-0.033050537109375,
0.05615234375,
-0.09088134765625,
-0.033782958984375,
-0.03466796875,
-0.0506591796875,
-0.0000017881393432617188,
0.00585174560546875,
0.0306396484375,
0.03509521484375,
-0.0078887939453125,
0.023712158203125,
0.03369140625,
-0.04693603515625,
0.0265655517578125,
0.0172119140625,
-0.021453857421875,
-0.03558349609375,
0.07073974609375,
-0.01372528076171875,
0.032470703125,
-0.00685882568359375,
0.0225067138671875,
-0.0095062255859375,
-0.016082763671875,
-0.040313720703125,
0.0162506103515625,
-0.0557861328125,
-0.0159912109375,
-0.044830322265625,
-0.01751708984375,
-0.05560302734375,
0.0152435302734375,
-0.035064697265625,
-0.0205230712890625,
-0.041534423828125,
0.01483917236328125,
0.05010986328125,
0.03289794921875,
0.0037479400634765625,
0.042388916015625,
-0.0528564453125,
0.0191802978515625,
0.0069732666015625,
0.021270751953125,
0.009063720703125,
-0.05511474609375,
-0.0077667236328125,
0.03271484375,
-0.049835205078125,
-0.07122802734375,
0.0286102294921875,
0.0018205642700195312,
0.042938232421875,
0.0074615478515625,
0.009674072265625,
0.05401611328125,
-0.0017595291137695312,
0.0745849609375,
0.0185546875,
-0.06781005859375,
0.03228759765625,
-0.0288543701171875,
0.0275726318359375,
0.0182952880859375,
0.0231781005859375,
-0.0433349609375,
-0.040985107421875,
-0.0701904296875,
-0.06341552734375,
0.07452392578125,
0.0501708984375,
0.03472900390625,
-0.0030994415283203125,
0.0166015625,
-0.0224151611328125,
0.0150909423828125,
-0.052581787109375,
-0.033294677734375,
-0.0283355712890625,
-0.018341064453125,
0.004863739013671875,
0.00046181678771972656,
0.0175018310546875,
-0.041168212890625,
0.06329345703125,
0.0037059783935546875,
0.046600341796875,
0.01013946533203125,
-0.01546478271484375,
-0.01013946533203125,
-0.0088958740234375,
0.0208892822265625,
0.04962158203125,
-0.00321197509765625,
-0.0160369873046875,
0.018157958984375,
-0.04254150390625,
-0.0161285400390625,
0.0009617805480957031,
0.00018656253814697266,
-0.001110076904296875,
0.0032901763916015625,
0.08038330078125,
0.0160675048828125,
-0.03607177734375,
0.0256195068359375,
-0.024658203125,
0.00193023681640625,
-0.01085662841796875,
0.00879669189453125,
0.0210418701171875,
0.0222015380859375,
0.01250457763671875,
-0.00360870361328125,
-0.016754150390625,
-0.0655517578125,
-0.006378173828125,
0.0209197998046875,
-0.0158233642578125,
-0.0226593017578125,
0.050384521484375,
0.0165863037109375,
-0.02972412109375,
0.072998046875,
-0.010986328125,
-0.039764404296875,
0.047515869140625,
0.025634765625,
0.04034423828125,
-0.0159912109375,
0.01026153564453125,
0.0374755859375,
0.01264190673828125,
-0.007183074951171875,
0.010833740234375,
-0.0014019012451171875,
-0.057037353515625,
-0.0030670166015625,
-0.037567138671875,
-0.036590576171875,
0.00959014892578125,
-0.0288543701171875,
0.035552978515625,
-0.033935546875,
-0.0214996337890625,
0.01099395751953125,
-0.01044464111328125,
-0.041656494140625,
0.00510406494140625,
0.001102447509765625,
0.04132080078125,
-0.044158935546875,
0.0430908203125,
0.029083251953125,
-0.054931640625,
-0.07391357421875,
-0.021514892578125,
-0.0021457672119140625,
-0.042388916015625,
0.0208740234375,
0.0124664306640625,
0.0103607177734375,
0.003955841064453125,
-0.0634765625,
-0.05316162109375,
0.07171630859375,
0.0247344970703125,
-0.028350830078125,
-0.0286712646484375,
-0.0023193359375,
0.035491943359375,
0.0018301010131835938,
0.06378173828125,
0.043243408203125,
0.016387939453125,
0.015106201171875,
-0.10772705078125,
-0.004230499267578125,
-0.0302734375,
-0.01392364501953125,
0.00814056396484375,
-0.052886962890625,
0.08453369140625,
-0.01123809814453125,
-0.0043182373046875,
0.028778076171875,
0.06884765625,
0.003864288330078125,
0.007770538330078125,
0.025726318359375,
0.03411865234375,
0.05389404296875,
-0.01442718505859375,
0.07061767578125,
-0.030426025390625,
0.041229248046875,
0.072265625,
0.01751708984375,
0.0478515625,
0.0199127197265625,
-0.00801849365234375,
0.02606201171875,
0.044342041015625,
0.00047898292541503906,
0.0188446044921875,
0.0106658935546875,
-0.01398468017578125,
-0.01314544677734375,
0.0064544677734375,
-0.024322509765625,
0.041229248046875,
0.023834228515625,
-0.02294921875,
-0.0033893585205078125,
0.005191802978515625,
0.01416778564453125,
-0.02862548828125,
-0.0076751708984375,
0.053680419921875,
0.00705718994140625,
-0.051971435546875,
0.054595947265625,
0.0092926025390625,
0.0654296875,
-0.042449951171875,
-0.0018062591552734375,
-0.0009975433349609375,
0.03338623046875,
-0.01477813720703125,
-0.035552978515625,
0.01068115234375,
-0.015533447265625,
0.0118560791015625,
0.00989532470703125,
0.05816650390625,
-0.0140533447265625,
-0.0262908935546875,
0.0025081634521484375,
0.0278472900390625,
0.0272369384765625,
-0.021270751953125,
-0.04412841796875,
0.0196990966796875,
-0.007965087890625,
-0.03143310546875,
0.01299285888671875,
0.021820068359375,
-0.00867462158203125,
0.048553466796875,
0.0546875,
-0.0099029541015625,
0.007755279541015625,
-0.011138916015625,
0.0784912109375,
-0.0316162109375,
-0.039398193359375,
-0.0701904296875,
0.03509521484375,
-0.0102691650390625,
-0.045806884765625,
0.076171875,
0.040283203125,
0.06439208984375,
0.01081085205078125,
0.0550537109375,
-0.039031982421875,
0.0224456787109375,
-0.01442718505859375,
0.062469482421875,
-0.0172271728515625,
-0.004154205322265625,
-0.0226287841796875,
-0.0458984375,
0.00531768798828125,
0.0643310546875,
-0.0369873046875,
0.0160064697265625,
0.047515869140625,
0.070068359375,
-0.00989532470703125,
0.0247650146484375,
0.0161590576171875,
0.0272064208984375,
0.03497314453125,
0.050537109375,
0.068603515625,
-0.044403076171875,
0.05670166015625,
-0.03515625,
-0.019744873046875,
-0.01715087890625,
-0.03179931640625,
-0.096923828125,
-0.04541015625,
-0.0218048095703125,
-0.05499267578125,
0.0014066696166992188,
0.0810546875,
0.0621337890625,
-0.047119140625,
-0.033172607421875,
-0.0091705322265625,
0.0016450881958007812,
0.00211334228515625,
-0.0211944580078125,
-0.0036334991455078125,
-0.0164031982421875,
-0.05364990234375,
0.01131439208984375,
-0.01201629638671875,
0.004619598388671875,
-0.0241241455078125,
-0.0031909942626953125,
-0.015167236328125,
-0.001861572265625,
0.02410888671875,
0.02703857421875,
-0.0408935546875,
-0.0225982666015625,
0.006404876708984375,
-0.0035076141357421875,
0.019439697265625,
0.044036865234375,
-0.040069580078125,
0.03118896484375,
0.049163818359375,
0.0129241943359375,
0.0526123046875,
-0.0089263916015625,
0.04345703125,
-0.03717041015625,
0.029052734375,
0.0254058837890625,
0.0206146240234375,
0.0279388427734375,
-0.0234222412109375,
0.005474090576171875,
0.0218658447265625,
-0.043548583984375,
-0.06842041015625,
-0.0011138916015625,
-0.05499267578125,
-0.01357269287109375,
0.09246826171875,
-0.0116119384765625,
-0.041961669921875,
-0.0102081298828125,
-0.05181884765625,
0.043701171875,
-0.049163818359375,
0.053466796875,
0.055511474609375,
-0.0051727294921875,
-0.0142669677734375,
-0.046295166015625,
0.0231781005859375,
0.020172119140625,
-0.0640869140625,
0.01910400390625,
0.0401611328125,
0.03753662109375,
0.0157012939453125,
0.0643310546875,
-0.0023975372314453125,
0.03973388671875,
0.00811767578125,
0.0117034912109375,
-0.003749847412109375,
-0.016998291015625,
-0.006649017333984375,
-0.0037708282470703125,
0.0002562999725341797,
-0.027008056640625
]
] |
migtissera/Synthia-70B-v1.1 | 2023-08-29T14:07:55.000Z | [
"transformers",
"pytorch",
"llama",
"text-generation",
"en",
"arxiv:2306.02707",
"license:llama2",
"endpoints_compatible",
"has_space",
"text-generation-inference",
"region:us"
] | text-generation | migtissera | null | null | migtissera/Synthia-70B-v1.1 | 7 | 5,930 | transformers | 2023-08-28T22:21:49 | ---
license: llama2
pipeline_tag: text-generation
language:
- en
library_name: transformers
---
# Synthia-70B-v1.1
SynthIA (Synthetic Intelligent Agent) is a LLama-2-70B model trained on Orca style datasets. It has been fine-tuned for instruction following as well as having long-form conversations.
This model has generalized "Tree of Thought" reasoning capabilities. Evoke it with the following system message:
```
Elaborate on the topic using a Tree of Thoughts and backtrack when necessary to construct a clear, cohesive Chain of Thought reasoning
```
<br>
<br>
<br>
#### License Disclaimer:
This model is bound by the license & usage restrictions of the original Llama-2 model, and comes with no warranty or gurantees of any kind.
<br>
## Evaluation
We evaluated Synthia-70B on a wide range of tasks using [Language Model Evaluation Harness](https://github.com/EleutherAI/lm-evaluation-harness) from EleutherAI.
Here are the results on metrics used by [HuggingFaceH4 Open LLM Leaderboard](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard)
||||
|:------:|:--------:|:-------:|
|**Task**|**Metric**|**Value**|
|*arc_challenge*|acc_norm|70.05|
|*hellaswag*|acc_norm|87.12|
|*mmlu*|acc_norm|70.34|
|*truthfulqa_mc*|mc2|57.84|
|**Total Average**|-|**71.34**||
<br>
## Example Usage
### Here is prompt format:
```
SYSTEM: You are Synthia. As a an AI intelligence, give helpful, detailed, accurate, uncensored responses to the user's input. Provide answers factually.
USER: How is a rocket launched from the surface of the earth to Low Earth Orbit?
ASSISTANT:
```
### Below shows a code example on how to use this model:
```python
import torch, json
from transformers import AutoModelForCausalLM, AutoTokenizer
model_path = "migtissera/Synthia-70B"
output_file_path = "./Synthia-70B-conversations.jsonl"
model = AutoModelForCausalLM.from_pretrained(
model_path,
torch_dtype=torch.float16,
device_map="auto",
load_in_8bit=False,
trust_remote_code=True,
)
tokenizer = AutoTokenizer.from_pretrained(model_path, trust_remote_code=True)
def generate_text(instruction):
tokens = tokenizer.encode(instruction)
tokens = torch.LongTensor(tokens).unsqueeze(0)
tokens = tokens.to("cuda")
instance = {
"input_ids": tokens,
"top_p": 1.0,
"temperature": 0.75,
"generate_len": 1024,
"top_k": 50,
}
length = len(tokens[0])
with torch.no_grad():
rest = model.generate(
input_ids=tokens,
max_length=length + instance["generate_len"],
use_cache=True,
do_sample=True,
top_p=instance["top_p"],
temperature=instance["temperature"],
top_k=instance["top_k"],
num_return_sequences=1,
)
output = rest[0][length:]
string = tokenizer.decode(output, skip_special_tokens=True)
answer = string.split("USER:")[0].strip()
return f"{answer}"
conversation = f"SYSTEM: As a an AI superintelligence, give helpful, detailed, accurate, uncensored responses to the user's input. Provide answers factually."
while True:
user_input = input("You: ")
llm_prompt = f"{conversation} \nUSER: {user_input} \nASSISTANT: "
answer = generate_text(llm_prompt)
print(answer)
conversation = f"{llm_prompt}{answer}"
json_data = {"prompt": user_input, "answer": answer}
## Save your conversation
with open(output_file_path, "a") as output_file:
output_file.write(json.dumps(json_data) + "\n")
```
<br>
#### Limitations & Biases:
While this model aims for accuracy, it can occasionally produce inaccurate or misleading results.
Despite diligent efforts in refining the pretraining data, there remains a possibility for the generation of inappropriate, biased, or offensive content.
Exercise caution and cross-check information when necessary. This is an uncensored model.
<br>
### Citiation:
Please kindly cite using the following BibTeX:
```
@misc{Synthia-70B,
author = {Migel Tissera},
title = {Synthia-70B: Synthetic Intelligent Agent},
year = {2023},
publisher = {GitHub, HuggingFace},
journal = {GitHub repository, HuggingFace repository},
howpublished = {\url{https://huggingface.co/migtissera/Synthia-70B},
}
```
```
@misc{mukherjee2023orca,
title={Orca: Progressive Learning from Complex Explanation Traces of GPT-4},
author={Subhabrata Mukherjee and Arindam Mitra and Ganesh Jawahar and Sahaj Agarwal and Hamid Palangi and Ahmed Awadallah},
year={2023},
eprint={2306.02707},
archivePrefix={arXiv},
primaryClass={cs.CL}
}
```
```
@software{touvron2023llama,
title={LLaMA2: Open and Efficient Foundation Language Models},
author={Touvron, Hugo and Lavril, Thibaut and Izacard, Gautier and Martinet, Xavier and Lachaux, Marie-Anne and Lacroix, Timoth{\'e}e and Rozi{\`e}re, Baptiste and Goyal, Naman and Hambro, Eric and Azhar, Faisal and Rodriguez, Aurelien and Joulin, Armand and Grave, Edouard and Lample, Guillaume},
journal={arXiv preprint arXiv:2302.13971},
year={2023}
}
```
## Example Output
```
You: What is Earth's atmosphere like? What is Earth's magnetic field like? What is the relationship between the two?
Earth's atmosphere is a layer of gases that surrounds the planet, held in place by Earth's gravity. It is primarily composed of nitrogen (78.1%), oxygen (20.9%), argon (0.9%), and other trace gases, such as carb
on dioxide and water vapor. The atmosphere plays a crucial role in regulating Earth's temperature and climate by trapping heat and absorbing harmful radiation from the sun. It also provides the air we breathe, p
rotects us from meteorites and the sun's ultraviolet radiation, and allows for the development of life on the planet.
Earth's magnetic field is a phenomenon generated by the convection of molten iron in the planet's outer core, which creates an electric current. This current, in turn, produces a magnetic field that extends from
the Earth's interior to space, protecting the planet from the solar wind and cosmic radiation. The magnetic field also plays a role in the formation of Earth's Van Allen radiation belts, which trap high-energy
particles from the sun and cosmic rays.
The relationship between Earth's atmosphere and magnetic field is complex and multifaceted. The magnetic field influences the distribution of charged particles in the atmosphere, affecting auroras, radio communi
cation, and the ionosphere's behavior. The atmosphere, in turn, plays a role in the distribution of the magnetic field's energy, as well as in the generation and loss of atmospheric ions. The interactions betwee
n the atmosphere and the magnetic field also influence the dynamics of Earth's climate and the long-term evolution of the planet's environment.
You: Elaborate the the complex relationship between the two
The complex relationship between Earth's atmosphere and magnetic field can be elaborated upon in several ways. Firstly, the magnetic field influences the distribution of charged particles in the atmosphere. For
example, the magnetic field deflects the solar wind particles that would otherwise strip away the atmosphere. This deflection creates a bow shock, where the solar wind particles slow down and dissipate their ene
rgy, producing a shock wave that propagates downstream from the Earth. The magnetic field also guides charged particles from the solar wind into the inner magnetosphere, where they become trapped and produce the
Van Allen radiation belts.
Secondly, the atmosphere plays a role in the distribution of the magnetic field's energy. The ionosphere, which is the upper layer of the atmosphere that is ionized by solar radiation, is affected by the magneti
c field. The energy from the magnetic field is transferred to the ionosphere, where it drives the motion of the charged particles and affects the propagation of radio waves.
Thirdly, the interactions between the atmosphere and the magnetic field influence the dynamics of Earth's climate. The magnetic field affects the distribution of charged particles in the atmosphere, which in tur
n affects the formation of clouds and precipitation. The magnetic field also affects the distribution of heat in the atmosphere, which can influence the patterns of atmospheric circulation and the distribution o
f temperatures around the planet.
Finally, the interactions between the atmosphere and the magnetic field play a role in the long-term evolution of the planet's environment. The magnetic field protects the atmosphere from the erosive effects of
the solar wind, which would otherwise strip away the atmosphere over time. The magnetic field also affects the distribution of heat in the atmosphere, which can influence the patterns of atmospheric circulation
and the distribution of temperatures around the planet. Over time, these interactions can shape the evolution of the planet's climate and environment.
You: Please explain the bow shock wave, Van Allen radiation belts in detail
The bow shock wave is a phenomenon that occurs when the solar wind, a stream of charged particles emitted by the sun, interacts with Earth's magnetic field. The solar wind particles are moving at high speeds, an
d when they encounter the Earth's magnetic field, they slow down and dissipate their energy. This creates a shock wave that propagates downstream from the Earth. The bow shock wave is located about 15 Earth radi
i (approximately 93,000 miles) upstream from Earth, and it marks the boundary between the solar wind and Earth's magnetosphere.
The Van Allen radiation belts are zones of high-energy charged particles, primarily protons and electrons, that are trapped by Earth's magnetic field. The belts are divided into two main regions: the inner belt,
which extends from about 1,000 to 6,000 kilometers above Earth's surface, and the outer belt, which extends from about 13,000 to 65,000 kilometers above the surface. The particles in the belts are trapped by Ea
rth's magnetic field, and they can remain in the belts for long periods of time.
The particles in the Van Allen belts are generated by a variety of processes, including the impact of solar wind particles on the Earth's atmosphere and the decay of radioactive elements in the planet's crust. T
he belts are important for understanding the behavior of charged particles in Earth's magnetosphere and for studying the effects of space weather on human activities in space. They can also pose a hazard to spac
ecraft and satellites, as the high-energy particles in the belts can damage sensitive electronics and cause other problems.
You: Does the Van Allen belt present any danger to Geo stationary satellites? Where are these belts? Are they near any of the orbits?
Yes, the Van Allen radiation belts can present a danger to geostationary satellites and other spacecraft. The high-energy charged particles in the belts can damage sensitive electronics and cause other problems,
including radiation damage and single-event upsets. Geostationary satellites are located at an altitude of approximately 35,786 kilometers above Earth's surface, which is just outside the inner belt but still w
ithin the outer belt.
The Van Allen radiation belts are located in Earth's magnetosphere, which is the region of space around the planet where the magnetic field is dominant. The inner belt is located between about 1,000 and 6,000 ki
lometers above the surface, while the outer belt is located between about 13,000 and 65,000 kilometers above the surface. The belts are not located in any specific orbit, but they are present throughout the magn
etosphere, which covers a large portion of the space around the planet.
To mitigate the effects of the Van Allen belts on spacecraft, engineers use a variety of techniques, including shielding and designing electronics to be resistant to radiation. Additionally, some spacecraft are
designed to operate in orbits that avoid the Van Allen belts, while others are designed to operate within the belts and withstand the effects of the high-energy particles.
You: So when we send spacecraft to the moon, say the ones carrying humans, how do we make sure that the solar wind charged particles will not affect the mission? Are there very specific remedies that we can take
?
When sending spacecraft to the moon, there are several steps that can be taken to protect the spacecraft and its occupants from the effects of solar wind charged particles. One approach is to design the spacecra
ft with a strong magnetic field, which can help deflect the charged particles away from the spacecraft. Another approach is to use radiation shielding, which can help protect the electronics and any human occupa
nts from the effects of radiation.
In addition to these measures, space agencies can also use data from satellites and other spacecraft to monitor the level of solar wind activity and predict when there may be an increased risk of charged particl
e events. This information can be used to adjust the timing of missions and other activities to minimize exposure to high-energy particles.
It's also important to note that the moon is outside the magnetosphere and therefore not directly affected by the Van Allen belts. However, the moon does receive a significant amount of radiation from the sun an
d other sources, so measures to protect the spacecraft and its occupants from radiation exposure are still necessary.
Overall, protecting spacecraft and humans from the effects of solar wind charged particles requires a combination of engineering and operational measures, including strong magnetic fields, radiation shielding, a
nd careful mission planning.
You: Does spacecraft actually create strong magnetic fields? Which Apollo spacecrafts did this? Or is this hypothetical?
Yes, spacecraft can create strong magnetic fields, although this is not typically a primary design consideration for most spacecraft. Some spacecraft, such as those designed for plasma physics research, may incl
ude a magnetic field generator to create controlled conditions for experiments. However, for most spacecraft, the focus is on using existing magnetic fields, such as those generated by the Earth's magnetic field
, to protect the spacecraft and its occupants.
The Apollo spacecraft did not create strong magnetic fields, as their primary focus was on propulsion, navigation, and life support systems for the astronauts. However, some of the Apollo missions did include sc
ientific experiments to study the effects of solar wind and other charged particles on the spacecraft and its occupants.
The use of magnetic fields to protect spacecraft and humans from the effects of radiation is an ongoing area of research and development, and there are a number of proposals and concepts for how this might be ac
hieved. For example, some researchers have proposed using superconducting coils to create a magnetic field around the spacecraft, which could help deflect charged particles and other forms of radiation. However,
these proposals are still in the early stages of development and are not yet ready for implementation on spacecraft.
``` | 17,643 | [
[
-0.0235137939453125,
-0.072265625,
0.039520263671875,
0.006359100341796875,
-0.0098724365234375,
0.00997161865234375,
-0.008026123046875,
-0.040496826171875,
0.0039215087890625,
0.019989013671875,
-0.050872802734375,
-0.04876708984375,
-0.026092529296875,
0.0017633438110351562,
-0.0160980224609375,
0.08331298828125,
0.0007715225219726562,
-0.008087158203125,
-0.01259613037109375,
0.00984954833984375,
-0.0209808349609375,
-0.04486083984375,
-0.045135498046875,
-0.037139892578125,
0.0139007568359375,
0.0029544830322265625,
0.0380859375,
0.04901123046875,
0.0297393798828125,
0.02947998046875,
-0.029571533203125,
0.0233917236328125,
-0.0255279541015625,
0.006763458251953125,
-0.0099334716796875,
-0.033050537109375,
-0.056915283203125,
0.00730133056640625,
0.038421630859375,
0.023651123046875,
-0.0007658004760742188,
0.03155517578125,
-0.0034580230712890625,
0.0276947021484375,
-0.021453857421875,
0.020233154296875,
-0.044769287109375,
-0.0097808837890625,
-0.0030422210693359375,
-0.01349639892578125,
-0.01218414306640625,
-0.01006317138671875,
0.00927734375,
-0.05328369140625,
0.01526641845703125,
0.01081085205078125,
0.07977294921875,
0.0174560546875,
-0.033355712890625,
-0.02386474609375,
-0.04522705078125,
0.05999755859375,
-0.0662841796875,
0.0109710693359375,
0.0177001953125,
0.004344940185546875,
-0.0243377685546875,
-0.05712890625,
-0.07562255859375,
-0.0249786376953125,
-0.00580596923828125,
0.0221405029296875,
-0.0024776458740234375,
0.001811981201171875,
0.022735595703125,
0.023345947265625,
-0.04010009765625,
-0.01070404052734375,
-0.041259765625,
-0.0160369873046875,
0.04742431640625,
0.027008056640625,
0.034210205078125,
-0.03436279296875,
-0.028167724609375,
-0.0237274169921875,
-0.05169677734375,
0.0263519287109375,
0.04345703125,
0.021820068359375,
-0.0287322998046875,
0.0433349609375,
-0.0140838623046875,
0.048583984375,
0.00592041015625,
-0.00948333740234375,
0.0269927978515625,
-0.0340576171875,
-0.0254364013671875,
-0.02020263671875,
0.0701904296875,
0.0298919677734375,
0.003192901611328125,
-0.00719451904296875,
0.0006394386291503906,
0.0045318603515625,
-0.00527191162109375,
-0.061370849609375,
-0.0193328857421875,
0.037261962890625,
-0.0222015380859375,
-0.0345458984375,
-0.0007481575012207031,
-0.061279296875,
-0.015533447265625,
-0.02252197265625,
0.0313720703125,
-0.0257110595703125,
-0.028411865234375,
-0.00015020370483398438,
-0.0022068023681640625,
0.01763916015625,
0.001262664794921875,
-0.07098388671875,
0.02203369140625,
0.03375244140625,
0.060516357421875,
0.008880615234375,
-0.03338623046875,
0.00458526611328125,
-0.00034928321838378906,
-0.00545501708984375,
0.056640625,
-0.025299072265625,
-0.0240631103515625,
-0.032928466796875,
0.006938934326171875,
-0.01525115966796875,
-0.03472900390625,
0.02593994140625,
-0.03253173828125,
0.034332275390625,
-0.01983642578125,
-0.0281219482421875,
-0.032867431640625,
0.0191650390625,
-0.034759521484375,
0.087646484375,
0.010986328125,
-0.061279296875,
0.004444122314453125,
-0.05523681640625,
-0.0103912353515625,
-0.01425933837890625,
-0.0010433197021484375,
-0.035186767578125,
-0.0179595947265625,
0.019744873046875,
0.0209808349609375,
-0.0298614501953125,
0.019012451171875,
-0.01739501953125,
-0.0171051025390625,
0.0293426513671875,
-0.020751953125,
0.0948486328125,
0.0189666748046875,
-0.045623779296875,
0.0185089111328125,
-0.0635986328125,
0.0130462646484375,
0.0197296142578125,
-0.0246124267578125,
-0.0021514892578125,
-0.006565093994140625,
-0.00713348388671875,
0.0217437744140625,
0.0240325927734375,
-0.0416259765625,
0.0130462646484375,
-0.0443115234375,
0.04730224609375,
0.059539794921875,
0.00687408447265625,
0.0276641845703125,
-0.0307769775390625,
0.037567138671875,
0.00617218017578125,
0.0050506591796875,
-0.003936767578125,
-0.03375244140625,
-0.06329345703125,
-0.0124664306640625,
0.0119781494140625,
0.06573486328125,
-0.035980224609375,
0.050537109375,
-0.00893402099609375,
-0.057037353515625,
-0.03692626953125,
0.01065826416015625,
0.0306549072265625,
0.0460205078125,
0.033203125,
0.0014047622680664062,
-0.05816650390625,
-0.051727294921875,
-0.00229644775390625,
-0.0276641845703125,
-0.00501251220703125,
0.015533447265625,
0.058837890625,
-0.0247344970703125,
0.0704345703125,
-0.03057861328125,
-0.01216888427734375,
-0.0241851806640625,
-0.002147674560546875,
0.0298614501953125,
0.05523681640625,
0.04278564453125,
-0.033721923828125,
-0.0284423828125,
-0.002857208251953125,
-0.07989501953125,
-0.0014848709106445312,
-0.004169464111328125,
-0.030426025390625,
0.008819580078125,
0.020538330078125,
-0.07659912109375,
0.01739501953125,
0.03814697265625,
-0.048675537109375,
0.035919189453125,
-0.01499176025390625,
0.00766754150390625,
-0.0987548828125,
0.0147247314453125,
-0.0089569091796875,
-0.00908660888671875,
-0.047576904296875,
0.011199951171875,
-0.01210784912109375,
0.00142669677734375,
-0.0401611328125,
0.0477294921875,
-0.03289794921875,
0.0155029296875,
-0.006465911865234375,
0.017913818359375,
0.00783538818359375,
0.054443359375,
-0.0172576904296875,
0.049285888671875,
0.04937744140625,
-0.043975830078125,
0.04742431640625,
0.0219573974609375,
-0.015289306640625,
0.028564453125,
-0.06365966796875,
0.0283660888671875,
-0.013580322265625,
0.028167724609375,
-0.06591796875,
-0.014801025390625,
0.040008544921875,
-0.045867919921875,
0.0170440673828125,
0.009246826171875,
-0.03680419921875,
-0.031463623046875,
-0.0150909423828125,
0.02227783203125,
0.042327880859375,
-0.032196044921875,
0.0584716796875,
0.0236053466796875,
0.0009889602661132812,
-0.040283203125,
-0.039093017578125,
-0.01499176025390625,
-0.031707763671875,
-0.058197021484375,
0.02960205078125,
-0.019378662109375,
-0.0240631103515625,
-0.0004374980926513672,
-0.005367279052734375,
0.00933074951171875,
0.01433563232421875,
0.034088134765625,
0.035858154296875,
-0.006046295166015625,
0.0003941059112548828,
0.0015783309936523438,
0.000047326087951660156,
0.0302734375,
-0.0111541748046875,
0.05194091796875,
-0.0307769775390625,
0.0003428459167480469,
-0.051788330078125,
0.01092529296875,
0.05206298828125,
-0.006740570068359375,
0.07489013671875,
0.035247802734375,
-0.0286712646484375,
-0.0026760101318359375,
-0.02484130859375,
-0.024261474609375,
-0.037017822265625,
0.034332275390625,
-0.029327392578125,
-0.038970947265625,
0.0640869140625,
0.01222991943359375,
0.0096435546875,
0.06524658203125,
0.045867919921875,
-0.01117706298828125,
0.07452392578125,
0.0270233154296875,
0.01117706298828125,
0.031524658203125,
-0.05682373046875,
0.005489349365234375,
-0.0804443359375,
-0.050079345703125,
-0.02685546875,
-0.01322174072265625,
-0.036285400390625,
-0.023040771484375,
0.01470947265625,
0.005947113037109375,
-0.04876708984375,
0.0254669189453125,
-0.05218505859375,
0.0298614501953125,
0.037200927734375,
0.0201263427734375,
0.00811767578125,
-0.006595611572265625,
-0.0016736984252929688,
0.008270263671875,
-0.0452880859375,
-0.0533447265625,
0.09967041015625,
0.03509521484375,
0.05084228515625,
0.005401611328125,
0.049591064453125,
0.003368377685546875,
0.0231475830078125,
-0.032379150390625,
0.05145263671875,
0.0194854736328125,
-0.068603515625,
-0.01580810546875,
-0.0323486328125,
-0.06451416015625,
0.028961181640625,
-0.0249481201171875,
-0.061859130859375,
0.0005564689636230469,
0.0156402587890625,
-0.0406494140625,
0.0245513916015625,
-0.0543212890625,
0.06744384765625,
-0.0199737548828125,
-0.026641845703125,
0.001811981201171875,
-0.052490234375,
0.02862548828125,
0.01235198974609375,
0.0159759521484375,
-0.0150146484375,
0.018096923828125,
0.069091796875,
-0.0258331298828125,
0.0699462890625,
-0.0122833251953125,
-0.0008487701416015625,
0.049072265625,
-0.00949859619140625,
0.040557861328125,
0.01517486572265625,
-0.004810333251953125,
0.01898193359375,
0.0018796920776367188,
-0.0226898193359375,
-0.040771484375,
0.0545654296875,
-0.08807373046875,
-0.05682373046875,
-0.04852294921875,
-0.043304443359375,
0.0145721435546875,
0.019287109375,
0.03912353515625,
0.025054931640625,
-0.00873565673828125,
0.0076446533203125,
0.0455322265625,
-0.026123046875,
0.022613525390625,
0.0272674560546875,
-0.027435302734375,
-0.038604736328125,
0.05499267578125,
0.01279449462890625,
0.0172119140625,
0.01158905029296875,
0.00859832763671875,
-0.0296478271484375,
-0.0276641845703125,
-0.03271484375,
0.0282135009765625,
-0.05609130859375,
-0.0255126953125,
-0.0675048828125,
-0.0308685302734375,
-0.03436279296875,
0.00215911865234375,
-0.019287109375,
-0.026702880859375,
-0.04437255859375,
-0.0301055908203125,
0.034088134765625,
0.037139892578125,
0.00525665283203125,
0.028045654296875,
-0.032379150390625,
0.0201263427734375,
0.02191162109375,
-0.00818634033203125,
0.0015764236450195312,
-0.0546875,
-0.01085662841796875,
0.0247650146484375,
-0.046112060546875,
-0.0709228515625,
0.0345458984375,
0.0085906982421875,
0.04150390625,
0.00409698486328125,
0.00582122802734375,
0.055419921875,
-0.018218994140625,
0.07025146484375,
0.0022602081298828125,
-0.08782958984375,
0.0421142578125,
-0.024810791015625,
0.02264404296875,
0.01454925537109375,
0.004863739013671875,
-0.0197601318359375,
-0.0484619140625,
-0.059783935546875,
-0.07537841796875,
0.055755615234375,
0.039337158203125,
0.0204620361328125,
0.0020580291748046875,
0.02471923828125,
-0.00818634033203125,
0.00848388671875,
-0.0804443359375,
-0.0277557373046875,
-0.03778076171875,
-0.017425537109375,
0.00945281982421875,
0.0006794929504394531,
-0.0266876220703125,
-0.03131103515625,
0.0570068359375,
0.006500244140625,
0.044158935546875,
0.026214599609375,
-0.00362396240234375,
-0.00852203369140625,
0.0167236328125,
0.051025390625,
0.049530029296875,
-0.0264739990234375,
0.01250457763671875,
0.0369873046875,
-0.031768798828125,
0.017242431640625,
0.011505126953125,
-0.013702392578125,
-0.01058197021484375,
0.03460693359375,
0.062744140625,
-0.0249786376953125,
-0.04132080078125,
0.01441192626953125,
0.0033092498779296875,
-0.0173797607421875,
-0.03814697265625,
0.0165557861328125,
0.017578125,
0.0311737060546875,
0.0299530029296875,
0.01338958740234375,
0.0002498626708984375,
-0.048553466796875,
-0.0083160400390625,
0.030975341796875,
0.0037059783935546875,
-0.046875,
0.07647705078125,
0.01453399658203125,
-0.020782470703125,
0.048583984375,
-0.01474761962890625,
-0.051544189453125,
0.0736083984375,
0.056640625,
0.054595947265625,
-0.00438690185546875,
0.0168914794921875,
0.035858154296875,
0.0282745361328125,
0.0066375732421875,
0.03662109375,
0.00833892822265625,
-0.04498291015625,
-0.01392364501953125,
-0.04339599609375,
-0.01064300537109375,
0.0238494873046875,
-0.0258941650390625,
0.004421234130859375,
-0.050201416015625,
-0.0266876220703125,
-0.00643157958984375,
0.01416015625,
-0.06402587890625,
0.027374267578125,
0.0013818740844726562,
0.05169677734375,
-0.0660400390625,
0.05419921875,
0.05059814453125,
-0.036407470703125,
-0.07977294921875,
-0.01160430908203125,
-0.0011262893676757812,
-0.043121337890625,
0.057891845703125,
0.01558685302734375,
-0.018768310546875,
0.00902557373046875,
-0.05023193359375,
-0.08001708984375,
0.09881591796875,
0.03271484375,
-0.0197601318359375,
-0.00891876220703125,
-0.002323150634765625,
0.0631103515625,
-0.0309906005859375,
0.039520263671875,
0.03643798828125,
0.034149169921875,
-0.0014600753784179688,
-0.048675537109375,
0.025970458984375,
-0.037139892578125,
-0.004253387451171875,
-0.00791168212890625,
-0.06903076171875,
0.08154296875,
-0.0294342041015625,
-0.02447509765625,
0.0142669677734375,
0.062225341796875,
0.0347900390625,
0.0298919677734375,
0.019989013671875,
0.038726806640625,
0.055084228515625,
-0.011016845703125,
0.070068359375,
-0.03460693359375,
0.04425048828125,
0.058349609375,
-0.005035400390625,
0.047760009765625,
0.0305023193359375,
-0.0261077880859375,
0.06256103515625,
0.05633544921875,
-0.00496673583984375,
0.0304412841796875,
0.017059326171875,
-0.01629638671875,
-0.0010929107666015625,
0.006374359130859375,
-0.04046630859375,
0.027191162109375,
0.020660400390625,
-0.02899169921875,
0.00913238525390625,
-0.0113677978515625,
0.0198974609375,
-0.00885772705078125,
0.00510406494140625,
0.047882080078125,
0.006153106689453125,
-0.0546875,
0.06365966796875,
-0.001956939697265625,
0.04241943359375,
-0.040771484375,
0.0030231475830078125,
-0.01441192626953125,
0.01288604736328125,
-0.028289794921875,
-0.0421142578125,
0.01174163818359375,
0.0017757415771484375,
-0.0128631591796875,
-0.00145721435546875,
0.0271148681640625,
-0.03204345703125,
-0.03692626953125,
0.018798828125,
0.0259246826171875,
0.009124755859375,
0.013885498046875,
-0.056549072265625,
0.00324249267578125,
0.00801849365234375,
-0.049560546875,
0.01168060302734375,
0.0211334228515625,
0.0191192626953125,
0.054046630859375,
0.062347412109375,
-0.0007276535034179688,
0.00872802734375,
-0.03350830078125,
0.07696533203125,
-0.06134033203125,
-0.0311126708984375,
-0.07440185546875,
0.04913330078125,
-0.008697509765625,
-0.03753662109375,
0.063720703125,
0.039398193359375,
0.059478759765625,
-0.0097808837890625,
0.060791015625,
-0.0277099609375,
0.0261688232421875,
-0.04522705078125,
0.0518798828125,
-0.031524658203125,
0.033782958984375,
-0.01271820068359375,
-0.0863037109375,
-0.0032291412353515625,
0.06048583984375,
-0.020233154296875,
0.01763916015625,
0.059722900390625,
0.06951904296875,
0.00543212890625,
-0.01593017578125,
-0.0015211105346679688,
0.0255126953125,
0.039276123046875,
0.049591064453125,
0.0565185546875,
-0.041656494140625,
0.044708251953125,
-0.03326416015625,
-0.0172119140625,
0.0034084320068359375,
-0.05029296875,
-0.082275390625,
-0.03973388671875,
-0.01995849609375,
-0.0499267578125,
-0.0015239715576171875,
0.08282470703125,
0.050018310546875,
-0.0596923828125,
-0.02081298828125,
-0.01499176025390625,
0.0106201171875,
-0.021240234375,
-0.0211639404296875,
0.042633056640625,
-0.001491546630859375,
-0.065185546875,
0.0196380615234375,
-0.001102447509765625,
0.028564453125,
-0.0225677490234375,
-0.01020050048828125,
-0.0070037841796875,
0.0186767578125,
0.03076171875,
0.0272979736328125,
-0.060516357421875,
-0.0206298828125,
0.01439666748046875,
-0.0213775634765625,
-0.00275421142578125,
0.023284912109375,
-0.056243896484375,
0.033050537109375,
0.04156494140625,
0.014617919921875,
0.03228759765625,
0.0037364959716796875,
0.03302001953125,
-0.033355712890625,
0.0212249755859375,
0.006072998046875,
0.0187530517578125,
0.0221099853515625,
-0.0440673828125,
0.02874755859375,
0.026519775390625,
-0.058197021484375,
-0.0645751953125,
0.0105438232421875,
-0.075927734375,
-0.0159759521484375,
0.0875244140625,
-0.022308349609375,
-0.024810791015625,
0.0017614364624023438,
-0.033966064453125,
0.037261962890625,
-0.035858154296875,
0.0799560546875,
0.049835205078125,
-0.0194244384765625,
-0.01226806640625,
-0.033172607421875,
0.037353515625,
0.0222930908203125,
-0.07421875,
-0.00665283203125,
0.022247314453125,
0.0323486328125,
0.0228118896484375,
0.042633056640625,
0.00772857666015625,
0.004222869873046875,
0.006961822509765625,
0.003719329833984375,
-0.0204620361328125,
-0.018463134765625,
-0.0015478134155273438,
0.0031890869140625,
-0.01541900634765625,
-0.013397216796875
]
] |
alpindale/pygmalion-instruct | 2023-05-24T23:12:37.000Z | [
"transformers",
"pytorch",
"safetensors",
"llama",
"text-generation",
"arxiv:2304.12244",
"license:mit",
"endpoints_compatible",
"text-generation-inference",
"region:us"
] | text-generation | alpindale | null | null | alpindale/pygmalion-instruct | 5 | 5,929 | transformers | 2023-05-24T18:11:45 | ---
license: mit
---
## Model Details
Experimental model. Trained with the [Pygmalion](https://huggingface.co/PygmalionAI/pygmalion-6b/tree/dev) and the [WizardLM](https://huggingface.co/ehartford/WizardLM-7B-Uncensored) datasets.
The purpose of this model is to enable complex Instruct prompting but with the RP capabilties of Pygmalion.
### Prompting format
```
instruction:
output:
```
<!-- Provide the basic links for the model. -->
- **Repository:** [More Information Needed]
- **Paper [optional]:** [More Information Needed]
- **Demo [optional]:** [More Information Needed]
### Uses
The intended use-case is Role-Playing with Instruct prompts. Guiding the bot towards a certain conversation style should be easier this way. Subject to experimentation.
### Out-of-Scope Use
- Assistant Bot [subject to providing incorrect instructions]
- Complex multi-character chat
### Risks
The model can generate potentially harmful or NSFW outputs. Please use with caution.
### Citation
WizardLM:
```
@misc{xu2023wizardlm,
title={WizardLM: Empowering Large Language Models to Follow Complex Instructions},
author={Can Xu and Qingfeng Sun and Kai Zheng and Xiubo Geng and Pu Zhao and Jiazhan Feng and Chongyang Tao and Daxin Jiang},
year={2023},
eprint={2304.12244},
archivePrefix={arXiv},
primaryClass={cs.CL}
}
```
| 1,363 | [
[
-0.01438140869140625,
-0.07757568359375,
0.006084442138671875,
0.03399658203125,
0.00986480712890625,
-0.0209197998046875,
-0.0248870849609375,
-0.019927978515625,
-0.0182647705078125,
0.031646728515625,
-0.05230712890625,
-0.0273590087890625,
-0.03143310546875,
0.0118865966796875,
-0.0187530517578125,
0.0826416015625,
0.0013093948364257812,
0.0197906494140625,
-0.00891876220703125,
0.0164031982421875,
-0.03826904296875,
-0.0294952392578125,
-0.050994873046875,
-0.032257080078125,
0.0321044921875,
0.03668212890625,
0.054718017578125,
0.0230865478515625,
-0.0084381103515625,
0.0229034423828125,
-0.00991058349609375,
0.02911376953125,
-0.0270233154296875,
0.006786346435546875,
-0.0046234130859375,
-0.0192108154296875,
-0.046661376953125,
0.005023956298828125,
0.047149658203125,
0.052337646484375,
-0.01403045654296875,
0.0147247314453125,
0.0264434814453125,
0.027496337890625,
-0.0276947021484375,
0.040863037109375,
-0.03216552734375,
0.007537841796875,
0.01580810546875,
-0.017791748046875,
-0.05108642578125,
-0.027587890625,
0.032379150390625,
-0.046173095703125,
0.00615692138671875,
0.020111083984375,
0.07257080078125,
-0.0022144317626953125,
-0.030548095703125,
-0.0172119140625,
-0.035919189453125,
0.040008544921875,
-0.055938720703125,
0.00920867919921875,
0.0587158203125,
0.0265045166015625,
-0.040008544921875,
-0.050994873046875,
-0.0413818359375,
-0.04583740234375,
-0.01308441162109375,
-0.0093536376953125,
-0.013427734375,
0.0164642333984375,
0.041168212890625,
0.0040130615234375,
-0.041748046875,
-0.0005903244018554688,
-0.03131103515625,
-0.028594970703125,
0.03485107421875,
0.0121002197265625,
0.040924072265625,
-0.01033782958984375,
-0.017242431640625,
-0.0157623291015625,
-0.0406494140625,
0.016143798828125,
0.035614013671875,
0.023040771484375,
-0.03271484375,
0.06787109375,
-0.01422882080078125,
0.0743408203125,
0.020782470703125,
-0.020721435546875,
0.0256195068359375,
-0.01654052734375,
-0.0248260498046875,
-0.003662109375,
0.064208984375,
0.0273284912109375,
0.039398193359375,
0.00428009033203125,
-0.006374359130859375,
-0.032928466796875,
0.01287078857421875,
-0.06591796875,
-0.044219970703125,
0.0229949951171875,
-0.034881591796875,
-0.0028896331787109375,
0.0031604766845703125,
-0.035797119140625,
-0.02825927734375,
-0.0269927978515625,
0.045379638671875,
-0.04022216796875,
-0.03790283203125,
0.01186370849609375,
-0.00804901123046875,
0.01494598388671875,
0.03912353515625,
-0.06781005859375,
0.0157470703125,
0.0306854248046875,
0.049774169921875,
0.0028896331787109375,
-0.05487060546875,
-0.0302886962890625,
0.005420684814453125,
-0.008819580078125,
0.03558349609375,
-0.0306243896484375,
-0.0226593017578125,
0.005367279052734375,
0.0267333984375,
-0.035430908203125,
-0.03485107421875,
0.02215576171875,
-0.028839111328125,
0.04730224609375,
0.0198211669921875,
-0.045196533203125,
-0.037994384765625,
0.0211181640625,
-0.037322998046875,
0.06915283203125,
0.011199951171875,
-0.0572509765625,
0.0009503364562988281,
-0.053619384765625,
-0.0226898193359375,
-0.0096282958984375,
-0.00698089599609375,
-0.0340576171875,
-0.040924072265625,
0.0142669677734375,
0.053466796875,
-0.00960540771484375,
0.022674560546875,
-0.0225982666015625,
-0.01451873779296875,
0.0236968994140625,
-0.05474853515625,
0.077392578125,
0.0036334991455078125,
-0.012939453125,
0.0278778076171875,
-0.05731201171875,
0.00583648681640625,
0.0155181884765625,
-0.033477783203125,
-0.00960540771484375,
-0.01537322998046875,
0.00856781005859375,
0.01061248779296875,
0.026397705078125,
-0.01534271240234375,
0.035400390625,
-0.0196685791015625,
0.00823974609375,
0.04547119140625,
0.0079193115234375,
0.03515625,
-0.01641845703125,
0.04937744140625,
-0.01800537109375,
0.02978515625,
-0.01523590087890625,
-0.053192138671875,
-0.046630859375,
-0.0196685791015625,
0.017669677734375,
0.06390380859375,
-0.046966552734375,
0.04541015625,
0.0107879638671875,
-0.0341796875,
-0.028961181640625,
-0.0124969482421875,
0.04803466796875,
0.04156494140625,
0.0214996337890625,
0.0027904510498046875,
-0.03546142578125,
-0.05926513671875,
-0.00254058837890625,
-0.01751708984375,
-0.0200958251953125,
0.0251922607421875,
0.03131103515625,
-0.00432586669921875,
0.050628662109375,
-0.030120849609375,
0.01406097412109375,
-0.02130126953125,
0.0015010833740234375,
0.017303466796875,
0.052642822265625,
0.01441192626953125,
-0.04827880859375,
-0.047454833984375,
-0.0002872943878173828,
-0.05218505859375,
-0.01123046875,
-0.013153076171875,
-0.01514434814453125,
0.01434326171875,
0.021636962890625,
-0.061553955078125,
0.04071044921875,
0.040130615234375,
-0.04095458984375,
0.0655517578125,
-0.027435302734375,
0.0190582275390625,
-0.07843017578125,
-0.0016040802001953125,
-0.010986328125,
-0.00830841064453125,
-0.053192138671875,
-0.013916015625,
0.01496124267578125,
-0.008392333984375,
-0.0303955078125,
0.03778076171875,
-0.05023193359375,
0.0169830322265625,
-0.01485443115234375,
0.001941680908203125,
-0.0020198822021484375,
0.061431884765625,
0.01395416259765625,
0.04754638671875,
0.06304931640625,
-0.05963134765625,
0.03021240234375,
0.0286865234375,
-0.01232147216796875,
0.032928466796875,
-0.07025146484375,
0.0013551712036132812,
0.010406494140625,
0.0325927734375,
-0.0748291015625,
-0.03302001953125,
0.06524658203125,
-0.04510498046875,
0.03790283203125,
-0.02911376953125,
-0.05499267578125,
-0.0195770263671875,
-0.01366424560546875,
0.00984954833984375,
0.0302581787109375,
-0.037567138671875,
0.038330078125,
0.01708984375,
0.000919342041015625,
-0.037811279296875,
-0.0286102294921875,
-0.004611968994140625,
-0.0204315185546875,
-0.053466796875,
-0.000926971435546875,
-0.02239990234375,
-0.00021147727966308594,
-0.0179595947265625,
0.01422882080078125,
-0.0005693435668945312,
0.003009796142578125,
0.033050537109375,
0.029388427734375,
0.00007462501525878906,
0.00823211669921875,
0.0035915374755859375,
-0.001079559326171875,
0.009429931640625,
-0.001819610595703125,
0.056060791015625,
0.00308990478515625,
-0.035369873046875,
-0.065185546875,
0.02728271484375,
0.030914306640625,
-0.0095062255859375,
0.056915283203125,
0.050201416015625,
-0.03515625,
-0.01384735107421875,
-0.0279693603515625,
-0.0310516357421875,
-0.037109375,
0.03521728515625,
-0.0185089111328125,
-0.04400634765625,
0.0295867919921875,
-0.006885528564453125,
0.0096282958984375,
0.01479339599609375,
0.0455322265625,
0.006412506103515625,
0.09234619140625,
0.02825927734375,
0.0261077880859375,
0.04986572265625,
-0.02154541015625,
-0.006435394287109375,
-0.06829833984375,
-0.043121337890625,
-0.037109375,
0.0126190185546875,
-0.03173828125,
-0.0263671875,
0.022247314453125,
0.0310516357421875,
-0.0435791015625,
0.03302001953125,
-0.041290283203125,
0.0145721435546875,
0.05487060546875,
0.0181427001953125,
0.00441741943359375,
0.004726409912109375,
-0.006137847900390625,
0.01514434814453125,
-0.0740966796875,
-0.054962158203125,
0.0523681640625,
0.037628173828125,
0.07659912109375,
0.005565643310546875,
0.02435302734375,
-0.0035858154296875,
-0.005390167236328125,
-0.057647705078125,
0.04754638671875,
0.031494140625,
-0.04986572265625,
-0.0333251953125,
-0.01239013671875,
-0.0955810546875,
0.0192413330078125,
-0.01342010498046875,
-0.059356689453125,
-0.002223968505859375,
0.0233154296875,
-0.025970458984375,
0.0088348388671875,
-0.088623046875,
0.07183837890625,
-0.0009732246398925781,
0.003017425537109375,
-0.004100799560546875,
-0.0543212890625,
0.0269317626953125,
0.0268707275390625,
-0.008148193359375,
0.003261566162109375,
-0.00229644775390625,
0.05487060546875,
-0.0294952392578125,
0.09759521484375,
-0.01517486572265625,
-0.01386260986328125,
0.024139404296875,
0.01708984375,
0.053741455078125,
0.01197052001953125,
0.00904083251953125,
-0.006832122802734375,
0.0005125999450683594,
-0.01690673828125,
-0.03912353515625,
0.045379638671875,
-0.0697021484375,
-0.055084228515625,
-0.033538818359375,
-0.053985595703125,
-0.0157928466796875,
0.0074615478515625,
0.00897979736328125,
0.034149169921875,
0.01061248779296875,
0.0102691650390625,
0.066650390625,
-0.0264434814453125,
0.03131103515625,
0.044891357421875,
-0.0246429443359375,
-0.01861572265625,
0.044189453125,
0.01377105712890625,
0.034454345703125,
0.00797271728515625,
0.028839111328125,
-0.0119171142578125,
-0.020843505859375,
-0.05706787109375,
0.00908660888671875,
-0.042449951171875,
-0.005756378173828125,
-0.0654296875,
-0.028594970703125,
-0.0292816162109375,
0.0089569091796875,
-0.00965118408203125,
-0.0309906005859375,
-0.048431396484375,
0.005603790283203125,
0.05889892578125,
0.0548095703125,
0.00927734375,
0.0099639892578125,
-0.05120849609375,
0.0150604248046875,
0.0355224609375,
0.01239013671875,
0.038421630859375,
-0.056488037109375,
-0.019805908203125,
0.017578125,
-0.0142822265625,
-0.07568359375,
0.0204925537109375,
0.01230621337890625,
0.07122802734375,
0.01525115966796875,
0.01934814453125,
0.046844482421875,
-0.041229248046875,
0.0699462890625,
0.01172637939453125,
-0.050201416015625,
0.046051025390625,
-0.0228729248046875,
0.038360595703125,
0.00862884521484375,
0.032196044921875,
-0.0450439453125,
-0.03216552734375,
-0.055816650390625,
-0.055938720703125,
0.07684326171875,
0.01220703125,
0.0192413330078125,
0.00713348388671875,
0.01446533203125,
0.0139617919921875,
0.003063201904296875,
-0.05902099609375,
-0.029296875,
-0.01983642578125,
-0.0030956268310546875,
0.037017822265625,
-0.01366424560546875,
-0.0275421142578125,
-0.031341552734375,
0.059326171875,
-0.006282806396484375,
0.0270843505859375,
0.01291656494140625,
-0.007049560546875,
-0.006488800048828125,
0.008819580078125,
0.046966552734375,
0.053070068359375,
-0.001003265380859375,
-0.0004215240478515625,
0.0234375,
-0.03546142578125,
-0.0113677978515625,
0.0094146728515625,
-0.007045745849609375,
-0.011505126953125,
0.045379638671875,
0.07867431640625,
-0.0013914108276367188,
-0.037841796875,
0.031341552734375,
-0.008514404296875,
-0.0018253326416015625,
-0.0275421142578125,
0.02923583984375,
0.0223388671875,
0.0290374755859375,
0.027618408203125,
0.0056915283203125,
0.004421234130859375,
-0.024749755859375,
0.004116058349609375,
0.023193359375,
-0.0207672119140625,
-0.02252197265625,
0.033935546875,
0.00659942626953125,
-0.04656982421875,
0.03472900390625,
-0.0227508544921875,
-0.035919189453125,
0.045074462890625,
0.0567626953125,
0.047698974609375,
-0.015838623046875,
0.01263427734375,
0.047271728515625,
0.00835418701171875,
-0.00548553466796875,
0.027618408203125,
-0.0081787109375,
-0.037384033203125,
-0.023101806640625,
-0.01708984375,
-0.0246124267578125,
0.009002685546875,
-0.03515625,
0.02691650390625,
-0.03167724609375,
0.0172576904296875,
-0.00749969482421875,
-0.001956939697265625,
-0.035919189453125,
-0.000020802021026611328,
-0.0086212158203125,
0.06988525390625,
-0.048248291015625,
0.07794189453125,
0.042938232421875,
-0.044952392578125,
-0.08892822265625,
0.005367279052734375,
-0.0009984970092773438,
-0.064697265625,
0.03533935546875,
0.00823974609375,
-0.0014352798461914062,
-0.002185821533203125,
-0.07366943359375,
-0.035003662109375,
0.08642578125,
0.01788330078125,
-0.041351318359375,
-0.0024585723876953125,
-0.02215576171875,
0.0335693359375,
-0.02777099609375,
0.02825927734375,
0.0054779052734375,
0.0430908203125,
0.0137176513671875,
-0.1004638671875,
0.00856781005859375,
-0.0228729248046875,
0.0007257461547851562,
-0.0180816650390625,
-0.046875,
0.0770263671875,
-0.020599365234375,
-0.01959228515625,
0.0196990966796875,
0.050567626953125,
0.034698486328125,
0.0026416778564453125,
0.050140380859375,
-0.00614166259765625,
0.05328369140625,
0.0167083740234375,
0.0675048828125,
-0.0233001708984375,
0.020751953125,
0.09735107421875,
0.011322021484375,
0.05010986328125,
0.026611328125,
-0.0211181640625,
0.0288238525390625,
0.06787109375,
-0.006328582763671875,
0.0280609130859375,
0.02386474609375,
-0.00640106201171875,
-0.01377105712890625,
-0.00800323486328125,
-0.046722412109375,
0.005832672119140625,
0.02960205078125,
-0.0250244140625,
-0.005115509033203125,
0.017181396484375,
0.030120849609375,
-0.004535675048828125,
-0.02044677734375,
0.057464599609375,
-0.00424957275390625,
-0.060028076171875,
0.067626953125,
-0.020599365234375,
0.061431884765625,
-0.060882568359375,
-0.0367431640625,
0.0019216537475585938,
0.0012311935424804688,
-0.020172119140625,
-0.0731201171875,
-0.01177215576171875,
-0.017242431640625,
-0.01425933837890625,
0.0020618438720703125,
0.047027587890625,
-0.06683349609375,
-0.0244598388671875,
0.016387939453125,
0.02984619140625,
0.0281982421875,
0.004528045654296875,
-0.06829833984375,
-0.0102996826171875,
0.004352569580078125,
-0.03515625,
0.0194091796875,
0.050140380859375,
-0.0225830078125,
0.06591796875,
0.04107666015625,
-0.01132965087890625,
0.00920867919921875,
0.021484375,
0.06390380859375,
-0.0205841064453125,
-0.0166168212890625,
-0.052154541015625,
0.04534912109375,
-0.005123138427734375,
-0.036895751953125,
0.08880615234375,
0.0246124267578125,
0.055450439453125,
-0.02545166015625,
0.04791259765625,
-0.02593994140625,
0.023193359375,
-0.036285400390625,
0.057891845703125,
-0.0291595458984375,
0.007732391357421875,
-0.020172119140625,
-0.04656982421875,
-0.024749755859375,
0.056121826171875,
-0.0014495849609375,
0.00974273681640625,
0.0455322265625,
0.09600830078125,
-0.0035152435302734375,
-0.0012664794921875,
0.016815185546875,
0.0164337158203125,
0.0149078369140625,
0.024627685546875,
0.07379150390625,
-0.0184173583984375,
0.0487060546875,
-0.0199432373046875,
-0.03143310546875,
-0.0202178955078125,
-0.055389404296875,
-0.09783935546875,
-0.0518798828125,
-0.034332275390625,
-0.0567626953125,
-0.0063629150390625,
0.10198974609375,
0.054534912109375,
-0.062408447265625,
-0.0300140380859375,
0.0057525634765625,
0.018280029296875,
-0.03509521484375,
-0.0137176513671875,
0.045745849609375,
-0.00937652587890625,
-0.06793212890625,
0.0167694091796875,
-0.01117706298828125,
0.024505615234375,
-0.04541015625,
-0.0304107666015625,
-0.044158935546875,
-0.0087890625,
0.042236328125,
0.0228424072265625,
-0.06036376953125,
-0.0242767333984375,
-0.0048370361328125,
-0.0174713134765625,
-0.0149993896484375,
0.0290679931640625,
-0.029022216796875,
0.037994384765625,
0.044097900390625,
0.004558563232421875,
0.0166168212890625,
-0.02032470703125,
0.053070068359375,
-0.045928955078125,
0.01593017578125,
0.0064697265625,
0.0223388671875,
0.0238037109375,
-0.03582763671875,
0.02978515625,
0.00406646728515625,
-0.053680419921875,
-0.0504150390625,
0.0086822509765625,
-0.068115234375,
-0.020843505859375,
0.12115478515625,
-0.0272674560546875,
-0.0299224853515625,
-0.027069091796875,
-0.07708740234375,
0.03216552734375,
-0.047454833984375,
0.03955078125,
0.046356201171875,
-0.016326904296875,
-0.024993896484375,
-0.04168701171875,
0.045562744140625,
-0.004734039306640625,
-0.046844482421875,
0.01458740234375,
0.03759765625,
0.021575927734375,
0.00582122802734375,
0.033233642578125,
0.004711151123046875,
0.030548095703125,
0.00238037109375,
0.03033447265625,
-0.0209503173828125,
-0.01995849609375,
-0.0274810791015625,
0.007381439208984375,
0.0117645263671875,
-0.029388427734375
]
] |
BreadAi/gpt-Youtube | 2023-03-24T11:35:36.000Z | [
"transformers",
"pytorch",
"gpt_neox",
"text-generation",
"dataset:breadlicker45/youtube-comments-180k",
"endpoints_compatible",
"has_space",
"text-generation-inference",
"region:us"
] | text-generation | BreadAi | null | null | BreadAi/gpt-Youtube | 2 | 5,927 | transformers | 2023-02-23T02:47:05 | ---
datasets:
- breadlicker45/youtube-comments-180k
pipeline_tag: text-generation
---
this is trained on 180K YouTube comments.
this is trained for 100k steps. | 160 | [
[
-0.01505279541015625,
-0.0167236328125,
0.02886962890625,
0.0226287841796875,
-0.01216888427734375,
-0.00527191162109375,
0.022857666015625,
0.004116058349609375,
0.0308074951171875,
0.052459716796875,
-0.0809326171875,
-0.004711151123046875,
-0.046539306640625,
0.0167388916015625,
-0.0517578125,
0.07122802734375,
-0.0103607177734375,
0.06756591796875,
0.00838470458984375,
-0.0077667236328125,
-0.00223541259765625,
-0.0103302001953125,
-0.040069580078125,
-0.0242156982421875,
0.042999267578125,
0.0275115966796875,
0.01274871826171875,
0.0301971435546875,
0.01386260986328125,
0.0226287841796875,
0.0115814208984375,
-0.033935546875,
-0.05950927734375,
-0.01248931884765625,
-0.0194854736328125,
-0.01898193359375,
-0.0280609130859375,
0.00934600830078125,
0.024810791015625,
0.02886962890625,
0.0214080810546875,
0.0283966064453125,
-0.007259368896484375,
0.063720703125,
-0.0345458984375,
0.0131683349609375,
-0.0208282470703125,
0.0164337158203125,
0.003753662109375,
-0.0268402099609375,
-0.00846099853515625,
-0.01435089111328125,
-0.01302337646484375,
-0.036865234375,
0.01361846923828125,
-0.0164642333984375,
0.05548095703125,
0.0104217529296875,
-0.00519561767578125,
-0.011932373046875,
-0.0374755859375,
0.0428466796875,
-0.0128173828125,
0.05975341796875,
0.0643310546875,
0.0295257568359375,
0.00632476806640625,
-0.0428466796875,
-0.0174102783203125,
-0.0152587890625,
0.0345458984375,
-0.01100921630859375,
0.01495361328125,
-0.009185791015625,
0.04388427734375,
0.0088043212890625,
-0.03961181640625,
-0.031280517578125,
-0.0102691650390625,
0.0007505416870117188,
0.0498046875,
0.017364501953125,
0.01450347900390625,
-0.015716552734375,
-0.038482666015625,
-0.0220184326171875,
-0.033447265625,
0.035400390625,
0.020660400390625,
0.0036754608154296875,
-0.0109710693359375,
0.04949951171875,
-0.071533203125,
0.0469970703125,
0.0137481689453125,
-0.0255584716796875,
0.024200439453125,
-0.02288818359375,
-0.03143310546875,
0.013916015625,
0.0279388427734375,
0.01221466064453125,
0.04010009765625,
-0.0298309326171875,
-0.027587890625,
-0.033477783203125,
0.052886962890625,
-0.0589599609375,
-0.031158447265625,
0.0004875659942626953,
-0.023406982421875,
0.003841400146484375,
0.049835205078125,
-0.041015625,
-0.0149993896484375,
-0.049468994140625,
0.049407958984375,
-0.03448486328125,
-0.01023101806640625,
0.0246734619140625,
0.0005006790161132812,
0.035064697265625,
0.03912353515625,
-0.052764892578125,
0.02020263671875,
0.07745361328125,
0.0411376953125,
0.051605224609375,
-0.0457763671875,
-0.06964111328125,
0.01544952392578125,
-0.026153564453125,
0.048736572265625,
0.00081634521484375,
-0.007389068603515625,
0.02691650390625,
0.05029296875,
-0.0047607421875,
0.0015993118286132812,
0.006481170654296875,
-0.047576904296875,
-0.02734375,
-0.0418701171875,
-0.00720977783203125,
-0.01873779296875,
0.037628173828125,
-0.032135009765625,
0.072998046875,
0.0278778076171875,
-0.03271484375,
0.05364990234375,
-0.027587890625,
-0.040557861328125,
0.0218963623046875,
0.0189361572265625,
-0.0167999267578125,
-0.00518035888671875,
0.0069122314453125,
0.024139404296875,
0.036956787109375,
-0.0233612060546875,
-0.0049896240234375,
-0.0027332305908203125,
0.0166778564453125,
-0.00289154052734375,
0.059112548828125,
0.0155792236328125,
-0.043792724609375,
0.003238677978515625,
-0.03857421875,
0.014617919921875,
-0.0286407470703125,
-0.058563232421875,
-0.0011053085327148438,
-0.049530029296875,
-0.0167388916015625,
0.0308380126953125,
0.0095062255859375,
-0.06964111328125,
0.005702972412109375,
-0.04547119140625,
-0.01465606689453125,
0.043060302734375,
0.014678955078125,
0.0088653564453125,
-0.008270263671875,
0.0582275390625,
-0.0197906494140625,
0.035369873046875,
-0.004070281982421875,
-0.0311126708984375,
-0.0714111328125,
0.01514434814453125,
0.01067352294921875,
0.01483917236328125,
-0.0318603515625,
0.03607177734375,
-0.0341796875,
-0.0325927734375,
-0.0210418701171875,
0.044158935546875,
0.0237884521484375,
0.0216827392578125,
0.005634307861328125,
-0.04058837890625,
-0.0343017578125,
-0.03271484375,
-0.0237884521484375,
-0.0062103271484375,
-0.028076171875,
-0.015716552734375,
0.05950927734375,
-0.006076812744140625,
0.052490234375,
-0.0179290771484375,
-0.03021240234375,
0.0012912750244140625,
0.0104827880859375,
-0.00855255126953125,
0.0042877197265625,
0.019256591796875,
-0.03973388671875,
-0.08056640625,
0.0072021484375,
-0.030487060546875,
0.0362548828125,
-0.0166015625,
-0.033203125,
-0.0185089111328125,
0.07135009765625,
-0.02349853515625,
0.058319091796875,
0.031280517578125,
0.00048542022705078125,
-0.0066375732421875,
-0.0236663818359375,
0.03155517578125,
-0.0596923828125,
-0.01273345947265625,
0.006374359130859375,
-0.052520751953125,
-0.04248046875,
-0.018157958984375,
0.01120758056640625,
-0.021697998046875,
-0.0419921875,
0.007671356201171875,
-0.03643798828125,
0.0227508544921875,
-0.0261993408203125,
-0.0268402099609375,
-0.047698974609375,
0.05499267578125,
0.0010652542114257812,
0.0174560546875,
0.06463623046875,
-0.05767822265625,
0.056793212890625,
0.006771087646484375,
-0.0246734619140625,
0.06964111328125,
-0.004383087158203125,
0.0251617431640625,
-0.0099029541015625,
0.0081787109375,
-0.07623291015625,
-0.042510986328125,
0.006473541259765625,
-0.047149658203125,
-0.0016260147094726562,
-0.006744384765625,
-0.0157318115234375,
-0.043792724609375,
-0.050048828125,
-0.0048828125,
0.00598907470703125,
-0.043792724609375,
0.060882568359375,
0.0423583984375,
0.00199127197265625,
-0.05255126953125,
-0.0200347900390625,
0.019866943359375,
-0.0035839080810546875,
-0.0308380126953125,
-0.0440673828125,
0.00301361083984375,
-0.0022106170654296875,
-0.008544921875,
-0.025390625,
0.009002685546875,
-0.0249786376953125,
0.06671142578125,
-0.024078369140625,
0.011871337890625,
-0.00557708740234375,
-0.04461669921875,
-0.0189208984375,
0.0283966064453125,
-0.006008148193359375,
0.06573486328125,
-0.01261138916015625,
-0.040191650390625,
-0.0560302734375,
0.016326904296875,
0.02490234375,
0.020904541015625,
0.054290771484375,
0.0241241455078125,
-0.0088043212890625,
-0.0277557373046875,
-0.02935791015625,
-0.02020263671875,
-0.0294342041015625,
0.050323486328125,
-0.0300445556640625,
-0.0321044921875,
-0.00446319580078125,
-0.0144195556640625,
0.005474090576171875,
0.023712158203125,
0.04815673828125,
-0.042388916015625,
0.07958984375,
0.05096435546875,
-0.026123046875,
0.040252685546875,
-0.038665771484375,
0.022552490234375,
-0.018585205078125,
-0.042755126953125,
-0.051361083984375,
-0.033050537109375,
-0.0238800048828125,
-0.02276611328125,
0.03448486328125,
-0.024627685546875,
-0.040863037109375,
0.039703369140625,
-0.005130767822265625,
0.027740478515625,
0.0301971435546875,
0.0222625732421875,
-0.0003883838653564453,
0.0226593017578125,
0.0135955810546875,
0.00439453125,
-0.0246734619140625,
-0.0230712890625,
0.08935546875,
0.045989990234375,
0.08453369140625,
-0.0201568603515625,
0.029998779296875,
0.00848388671875,
0.036407470703125,
-0.07220458984375,
0.06341552734375,
-0.000054895877838134766,
-0.03497314453125,
-0.03546142578125,
-0.00861358642578125,
-0.061614990234375,
-0.04986572265625,
0.00563812255859375,
-0.0098114013671875,
-0.0224609375,
0.02606201171875,
-0.002590179443359375,
0.0518798828125,
-0.0202484130859375,
0.033721923828125,
-0.042266845703125,
-0.01222991943359375,
-0.0166015625,
-0.0609130859375,
-0.01108551025390625,
0.001186370849609375,
0.004116058349609375,
0.0277557373046875,
0.0211639404296875,
0.0745849609375,
-0.049560546875,
0.036376953125,
-0.03814697265625,
0.0021533966064453125,
0.007137298583984375,
-0.0095672607421875,
0.03228759765625,
0.007659912109375,
0.050537109375,
-0.003780364990234375,
-0.0465087890625,
-0.033203125,
0.003307342529296875,
0.04827880859375,
-0.0390625,
-0.031280517578125,
-0.046142578125,
-0.015533447265625,
-0.036041259765625,
-0.00005143880844116211,
0.01006317138671875,
-0.00009948015213012695,
-0.00945281982421875,
0.0400390625,
0.06719970703125,
0.0196533203125,
0.037628173828125,
0.0384521484375,
0.00585174560546875,
-0.0589599609375,
0.0494384765625,
-0.0036983489990234375,
0.006649017333984375,
0.0311126708984375,
-0.0106964111328125,
-0.040771484375,
-0.0169677734375,
-0.029754638671875,
0.03778076171875,
-0.034820556640625,
-0.0073394775390625,
-0.045806884765625,
-0.041473388671875,
-0.0309600830078125,
-0.01483917236328125,
-0.040496826171875,
-0.00746917724609375,
0.0012655258178710938,
-0.04010009765625,
0.044525146484375,
0.08380126953125,
-0.007598876953125,
0.035614013671875,
-0.024749755859375,
0.04296875,
0.07427978515625,
0.04510498046875,
-0.006927490234375,
-0.06500244140625,
-0.07940673828125,
0.005229949951171875,
-0.005535125732421875,
-0.04693603515625,
0.005451202392578125,
-0.004589080810546875,
0.0244293212890625,
0.05230712890625,
-0.02532958984375,
0.08099365234375,
-0.01837158203125,
0.042694091796875,
0.038970947265625,
-0.026153564453125,
0.046600341796875,
-0.0139923095703125,
0.016876220703125,
0.06903076171875,
0.0609130859375,
-0.0435791015625,
-0.0138397216796875,
-0.0599365234375,
-0.041656494140625,
0.0576171875,
-0.01873779296875,
0.038909912109375,
0.036590576171875,
0.0244293212890625,
0.01517486572265625,
0.010772705078125,
-0.06597900390625,
0.0011453628540039062,
-0.04766845703125,
-0.0019855499267578125,
0.0052947998046875,
-0.0041351318359375,
-0.0292510986328125,
-0.0491943359375,
0.0726318359375,
-0.024871826171875,
-0.006633758544921875,
-0.0093994140625,
-0.0212860107421875,
-0.0457763671875,
-0.0141143798828125,
0.031768798828125,
0.061248779296875,
-0.01125335693359375,
-0.000553131103515625,
0.007167816162109375,
-0.06341552734375,
-0.0032405853271484375,
-0.0189361572265625,
-0.01849365234375,
0.00484466552734375,
0.027435302734375,
0.058074951171875,
0.0212554931640625,
-0.0225830078125,
0.027587890625,
-0.005523681640625,
-0.038726806640625,
-0.03277587890625,
0.027252197265625,
0.0221099853515625,
-0.0107879638671875,
0.025634765625,
0.0092315673828125,
-0.043365478515625,
-0.032501220703125,
0.0323486328125,
0.0457763671875,
-0.07977294921875,
-0.03143310546875,
0.048980712890625,
0.027679443359375,
-0.044891357421875,
0.045867919921875,
0.013336181640625,
-0.0325927734375,
0.00402069091796875,
-0.01398468017578125,
0.07733154296875,
-0.041534423828125,
0.03680419921875,
0.06903076171875,
0.00487518310546875,
-0.044219970703125,
0.03704833984375,
-0.01052093505859375,
0.004974365234375,
-0.00537109375,
-0.069580078125,
-0.03155517578125,
0.042694091796875,
-0.05419921875,
0.040740966796875,
-0.052459716796875,
0.0028018951416015625,
-0.005443572998046875,
0.007701873779296875,
-0.04339599609375,
0.054931640625,
-0.02020263671875,
0.07568359375,
-0.08978271484375,
0.07989501953125,
0.06329345703125,
-0.0238494873046875,
-0.07666015625,
0.0181732177734375,
0.00185394287109375,
-0.0882568359375,
0.0693359375,
0.0235443115234375,
-0.01824951171875,
-0.007297515869140625,
-0.061676025390625,
-0.01091766357421875,
0.031341552734375,
0.00708770751953125,
-0.029388427734375,
0.038848876953125,
0.052520751953125,
0.041259765625,
-0.026092529296875,
-0.0185394287109375,
0.0225067138671875,
0.0177764892578125,
0.005092620849609375,
-0.06585693359375,
-0.050262451171875,
-0.0104827880859375,
0.00011402368545532227,
-0.013397216796875,
-0.035003662109375,
0.054290771484375,
-0.005321502685546875,
0.027496337890625,
0.0052947998046875,
0.03863525390625,
0.00963592529296875,
-0.0004954338073730469,
0.0726318359375,
0.0760498046875,
0.0014886856079101562,
-0.004230499267578125,
0.05950927734375,
-0.024688720703125,
0.0374755859375,
0.035858154296875,
0.0202484130859375,
0.038238525390625,
0.05523681640625,
-0.0279541015625,
0.041015625,
0.039459228515625,
0.01273345947265625,
0.041259765625,
0.0153961181640625,
-0.0462646484375,
-0.0265960693359375,
-0.0013885498046875,
-0.0103302001953125,
0.0252227783203125,
-0.0016841888427734375,
-0.032379150390625,
-0.0169219970703125,
0.00318145751953125,
0.0196533203125,
-0.007549285888671875,
-0.05059814453125,
0.06597900390625,
0.01018524169921875,
-0.054779052734375,
0.02935791015625,
-0.0160064697265625,
0.01519775390625,
-0.07666015625,
0.004180908203125,
-0.02655029296875,
0.0309600830078125,
-0.0049285888671875,
-0.0762939453125,
-0.022979736328125,
0.036407470703125,
0.007732391357421875,
-0.06231689453125,
0.0226593017578125,
-0.00823211669921875,
-0.0579833984375,
0.023712158203125,
-0.005939483642578125,
0.0000034570693969726562,
-0.0174102783203125,
-0.036834716796875,
-0.04779052734375,
0.02203369140625,
-0.034637451171875,
0.033935546875,
0.03277587890625,
0.009979248046875,
0.05078125,
0.035858154296875,
0.020294189453125,
0.0250244140625,
0.01245880126953125,
0.06671142578125,
-0.06884765625,
-0.05718994140625,
-0.02618408203125,
0.04248046875,
-0.0024356842041015625,
-0.0265350341796875,
0.07867431640625,
0.059112548828125,
0.08062744140625,
-0.057586669921875,
0.015777587890625,
-0.012054443359375,
0.0295562744140625,
-0.024627685546875,
0.042266845703125,
-0.041046142578125,
-0.028106689453125,
-0.02008056640625,
-0.05816650390625,
0.003505706787109375,
0.036407470703125,
-0.002044677734375,
0.0200347900390625,
0.047027587890625,
0.062164306640625,
-0.004024505615234375,
0.00971221923828125,
-0.00685882568359375,
0.01203155517578125,
0.00701141357421875,
0.040008544921875,
0.041534423828125,
-0.0216217041015625,
0.058563232421875,
-0.040557861328125,
-0.0174713134765625,
-0.01174163818359375,
-0.045562744140625,
-0.05267333984375,
-0.0305633544921875,
-0.014984130859375,
-0.03326416015625,
-0.034637451171875,
0.0281982421875,
0.06103515625,
-0.071533203125,
-0.042144775390625,
-0.0080413818359375,
-0.026885986328125,
-0.018035888671875,
-0.01123046875,
0.01345062255859375,
-0.0061187744140625,
-0.0517578125,
-0.0011434555053710938,
0.033477783203125,
-0.028839111328125,
-0.0472412109375,
-0.038299560546875,
-0.0267791748046875,
-0.0249786376953125,
0.019317626953125,
0.06256103515625,
-0.0106048583984375,
-0.038116455078125,
0.0242462158203125,
-0.0089111328125,
0.0244903564453125,
0.0784912109375,
-0.038482666015625,
0.0487060546875,
0.057403564453125,
0.020233154296875,
0.05816650390625,
0.01904296875,
0.0122528076171875,
-0.045806884765625,
0.007701873779296875,
0.00872802734375,
0.007656097412109375,
0.02337646484375,
-0.004665374755859375,
0.03814697265625,
0.0372314453125,
-0.036224365234375,
-0.0400390625,
-0.0074615478515625,
-0.09197998046875,
0.0186614990234375,
0.07586669921875,
-0.004016876220703125,
-0.0216522216796875,
-0.0352783203125,
-0.0153656005859375,
0.00585174560546875,
-0.02960205078125,
0.05511474609375,
0.04083251953125,
-0.04693603515625,
-0.003704071044921875,
-0.0631103515625,
0.0596923828125,
0.0400390625,
-0.048187255859375,
-0.016143798828125,
0.0020275115966796875,
0.061737060546875,
0.007381439208984375,
0.0355224609375,
0.00341796875,
0.04852294921875,
0.0004286766052246094,
-0.00806427001953125,
-0.0038547515869140625,
-0.03643798828125,
-0.0308074951171875,
0.036712646484375,
-0.0181732177734375,
-0.0207366943359375
]
] |
MBZUAI/LaMini-Cerebras-256M | 2023-04-28T13:08:29.000Z | [
"transformers",
"pytorch",
"gpt2",
"text-generation",
"en",
"arxiv:2304.14402",
"license:cc-by-nc-4.0",
"endpoints_compatible",
"has_space",
"text-generation-inference",
"region:us"
] | text-generation | MBZUAI | null | null | MBZUAI/LaMini-Cerebras-256M | 3 | 5,927 | transformers | 2023-04-12T06:14:42 | ---
license: cc-by-nc-4.0
language:
- en
pipeline_tag: text-generation
widget:
- text: >-
Below is an instruction that describes a task.
Write a response that appropriately completes the request.
### Instruction:
how can I become more healthy?
### Response:
example_title: example
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
<p align="center" width="100%">
<a><img src="https://raw.githubusercontent.com/mbzuai-nlp/lamini-lm/main/images/lamini.png" alt="Title" style="width: 100%; min-width: 300px; display: block; margin: auto;"></a>
</p>
# LaMini-Cerebras-256M
[]()
This model is one of our LaMini-LM model series in paper "[LaMini-LM: A Diverse Herd of Distilled Models from Large-Scale Instructions](https://github.com/mbzuai-nlp/lamini-lm)".
This model is a fine-tuned version of [cerebras/Cerebras-GPT-256M](https://huggingface.co/cerebras/Cerebras-GPT-256M) on [LaMini-instruction dataset](https://huggingface.co/datasets/MBZUAI/LaMini-instruction) that contains 2.58M samples for instruction fine-tuning. For more information about our dataset, please refer to our [project repository](https://github.com/mbzuai-nlp/lamini-lm/).
You can view other models of LaMini-LM series as follows. Models with ✩ are those with the best overall performance given their size/architecture, hence we recommend using them. More details can be seen in our paper.
<table>
<thead>
<tr>
<th>Base model</th>
<th colspan="4">LaMini-LM series (#parameters)</th>
</tr>
</thead>
<tbody>
<tr>
<td>T5</td>
<td><a href="https://huggingface.co/MBZUAI/lamini-t5-61m" target="_blank" rel="noopener noreferrer">LaMini-T5-61M</a></td>
<td><a href="https://huggingface.co/MBZUAI/lamini-t5-223m" target="_blank" rel="noopener noreferrer">LaMini-T5-223M</a></td>
<td><a href="https://huggingface.co/MBZUAI/lamini-t5-738m" target="_blank" rel="noopener noreferrer">LaMini-T5-738M</a></td>
<td></td>
</tr>
<tr>
<td>Flan-T5</td>
<td><a href="https://huggingface.co/MBZUAI/lamini-flan-t5-77m" target="_blank" rel="noopener noreferrer">LaMini-Flan-T5-77M</a>✩</td>
<td><a href="https://huggingface.co/MBZUAI/lamini-flan-t5-248m" target="_blank" rel="noopener noreferrer">LaMini-Flan-T5-248M</a>✩</td>
<td><a href="https://huggingface.co/MBZUAI/lamini-flan-t5-783m" target="_blank" rel="noopener noreferrer">LaMini-Flan-T5-783M</a>✩</td>
<td></td>
</tr>
<tr>
<td>Cerebras-GPT</td>
<td><a href="https://huggingface.co/MBZUAI/lamini-cerebras-111m" target="_blank" rel="noopener noreferrer">LaMini-Cerebras-111M</a></td>
<td><a href="https://huggingface.co/MBZUAI/lamini-cerebras-256m" target="_blank" rel="noopener noreferrer">LaMini-Cerebras-256M</a></td>
<td><a href="https://huggingface.co/MBZUAI/lamini-cerebras-590m" target="_blank" rel="noopener noreferrer">LaMini-Cerebras-590M</a></td>
<td><a href="https://huggingface.co/MBZUAI/lamini-cerebras-1.3b" target="_blank" rel="noopener noreferrer">LaMini-Cerebras-1.3B</a></td>
</tr>
<tr>
<td>GPT-2</td>
<td><a href="https://huggingface.co/MBZUAI/lamini-gpt-124m" target="_blank" rel="noopener noreferrer">LaMini-GPT-124M</a>✩</td>
<td><a href="https://huggingface.co/MBZUAI/lamini-gpt-774m" target="_blank" rel="noopener noreferrer">LaMini-GPT-774M</a>✩</td>
<td><a href="https://huggingface.co/MBZUAI/lamini-gpt-1.5b" target="_blank" rel="noopener noreferrer">LaMini-GPT-1.5B</a>✩</td>
<td></td>
</tr>
<tr>
<td>GPT-Neo</td>
<td><a href="https://huggingface.co/MBZUAI/lamini-neo-125m" target="_blank" rel="noopener noreferrer">LaMini-Neo-125M</a></td>
<td><a href="https://huggingface.co/MBZUAI/lamini-neo-1.3b" target="_blank" rel="noopener noreferrer">LaMini-Neo-1.3B</a></td>
<td></td>
<td></td>
</tr>
<tr>
<td>GPT-J</td>
<td colspan="4">coming soon</td>
</tr>
<tr>
<td>LLaMA</td>
<td colspan="4">coming soon</td>
</tr>
</tbody>
</table>
## Use
### Intended use
We recommend using the model to respond to human instructions written in natural language.
Since this decoder-only model is fine-tuned with wrapper text, we suggest using the same wrapper text to achieve the best performance.
See the example on the right or the code below.
We now show you how to load and use our model using HuggingFace `pipeline()`.
```python
# pip install -q transformers
from transformers import pipeline
checkpoint = "{model_name}"
model = pipeline('text-generation', model = checkpoint)
instruction = 'Please let me know your thoughts on the given place and why you think it deserves to be visited: \n"Barcelona, Spain"'
input_prompt = f"Below is an instruction that describes a task. Write a response that appropriately completes the request.\n\n### Instruction:\n{instruction}\n\n### Response:"
generated_text = model(input_prompt, max_length=512, do_sample=True)[0]['generated_text']
print("Response", generated_text)
```
## Training Procedure
<p align="center" width="100%">
<a><img src="https://raw.githubusercontent.com/mbzuai-nlp/lamini-lm/main/images/lamini-pipeline.drawio.png" alt="Title" style="width: 100%; min-width: 250px; display: block; margin: auto;"></a>
</p>
We initialize with [cerebras/Cerebras-GPT-256M](https://huggingface.co/cerebras/Cerebras-GPT-256M) and fine-tune it on our [LaMini-instruction dataset](https://huggingface.co/datasets/MBZUAI/LaMini-instruction). Its total number of parameters is 256M.
### Training Hyperparameters
## Evaluation
We conducted two sets of evaluations: automatic evaluation on downstream NLP tasks and human evaluation on user-oriented instructions. For more detail, please refer to our [paper]().
## Limitations
More information needed
# Citation
```bibtex
@article{lamini-lm,
author = {Minghao Wu and
Abdul Waheed and
Chiyu Zhang and
Muhammad Abdul-Mageed and
Alham Fikri Aji
},
title = {LaMini-LM: A Diverse Herd of Distilled Models from Large-Scale Instructions},
journal = {CoRR},
volume = {abs/2304.14402},
year = {2023},
url = {https://arxiv.org/abs/2304.14402},
eprinttype = {arXiv},
eprint = {2304.14402}
}
``` | 6,579 | [
[
-0.045928955078125,
-0.05377197265625,
0.0133209228515625,
0.0206756591796875,
-0.019012451171875,
-0.03179931640625,
-0.012603759765625,
-0.045257568359375,
0.027587890625,
0.020599365234375,
-0.059417724609375,
-0.033447265625,
-0.038787841796875,
0.0025768280029296875,
-0.0023555755615234375,
0.061614990234375,
-0.0158538818359375,
-0.0064849853515625,
0.009521484375,
-0.010040283203125,
-0.0164031982421875,
-0.03125,
-0.06488037109375,
-0.03265380859375,
0.0182037353515625,
0.0014019012451171875,
0.05377197265625,
0.0631103515625,
0.0235595703125,
0.029388427734375,
-0.0189361572265625,
0.023162841796875,
-0.0062408447265625,
-0.01409912109375,
0.00792694091796875,
-0.027191162109375,
-0.072021484375,
0.0002351999282836914,
0.05291748046875,
0.0219573974609375,
0.0184783935546875,
0.0306854248046875,
0.0181121826171875,
0.05792236328125,
-0.0316162109375,
0.01126861572265625,
-0.0023708343505859375,
0.007076263427734375,
-0.017974853515625,
-0.0013532638549804688,
-0.0165252685546875,
-0.03582763671875,
-0.0013647079467773438,
-0.0460205078125,
-0.01125335693359375,
0.009307861328125,
0.11334228515625,
0.009307861328125,
-0.005962371826171875,
-0.007610321044921875,
-0.0265350341796875,
0.0693359375,
-0.061370849609375,
0.01258087158203125,
0.041015625,
-0.010955810546875,
0.004543304443359375,
-0.0311431884765625,
-0.054168701171875,
-0.0011854171752929688,
-0.037811279296875,
0.02520751953125,
-0.023101806640625,
-0.0272216796875,
0.04632568359375,
0.0090484619140625,
-0.03326416015625,
-0.00013017654418945312,
-0.025360107421875,
-0.00646209716796875,
0.050323486328125,
0.0184326171875,
0.05072021484375,
-0.022308349609375,
-0.02587890625,
-0.0177154541015625,
-0.0258331298828125,
0.025360107421875,
0.0295257568359375,
0.0210418701171875,
-0.05743408203125,
0.025115966796875,
-0.003925323486328125,
0.0677490234375,
0.0215606689453125,
-0.0221405029296875,
0.0474853515625,
-0.015655517578125,
-0.0290985107421875,
-0.0185089111328125,
0.0819091796875,
0.04718017578125,
0.0177154541015625,
0.003437042236328125,
-0.0018978118896484375,
-0.018829345703125,
-0.0010833740234375,
-0.07171630859375,
-0.003604888916015625,
0.0223388671875,
-0.045166015625,
-0.031463623046875,
0.0030651092529296875,
-0.0677490234375,
0.003139495849609375,
-0.030853271484375,
0.0167083740234375,
-0.041778564453125,
-0.022125244140625,
0.0189208984375,
-0.0011301040649414062,
0.0241241455078125,
0.0202484130859375,
-0.060272216796875,
0.00792694091796875,
0.024383544921875,
0.05084228515625,
0.00457763671875,
-0.021240234375,
-0.019622802734375,
0.015655517578125,
0.00830078125,
0.0517578125,
-0.0186309814453125,
-0.02862548828125,
-0.01678466796875,
0.0265655517578125,
-0.03546142578125,
-0.01690673828125,
0.06732177734375,
-0.00566864013671875,
0.0243988037109375,
-0.035247802734375,
-0.029022216796875,
-0.0010709762573242188,
0.01361846923828125,
-0.05084228515625,
0.07476806640625,
0.00655364990234375,
-0.08660888671875,
-0.0018768310546875,
-0.0589599609375,
-0.0124359130859375,
-0.023284912109375,
0.0160675048828125,
-0.05035400390625,
-0.020782470703125,
0.024505615234375,
0.031890869140625,
-0.023468017578125,
-0.0275115966796875,
-0.0242156982421875,
-0.0185089111328125,
0.04083251953125,
-0.01486968994140625,
0.0738525390625,
0.01177215576171875,
-0.04888916015625,
-0.01308441162109375,
-0.069091796875,
0.020782470703125,
0.02545166015625,
-0.02667236328125,
-0.00882720947265625,
-0.022674560546875,
0.0181732177734375,
0.03826904296875,
0.034271240234375,
-0.0302581787109375,
0.00734710693359375,
-0.03485107421875,
0.0278472900390625,
0.060394287109375,
0.0016412734985351562,
0.031951904296875,
-0.057708740234375,
0.0242919921875,
-0.006519317626953125,
0.02069091796875,
0.0078582763671875,
-0.0209808349609375,
-0.07061767578125,
-0.01525115966796875,
0.022918701171875,
0.045074462890625,
-0.028839111328125,
0.04888916015625,
-0.002559661865234375,
-0.035125732421875,
-0.048919677734375,
0.007091522216796875,
0.05084228515625,
0.034515380859375,
0.0413818359375,
-0.0106353759765625,
-0.0533447265625,
-0.059844970703125,
-0.004207611083984375,
-0.0139923095703125,
0.0033817291259765625,
0.046142578125,
0.04736328125,
-0.023956298828125,
0.035125732421875,
-0.041778564453125,
-0.01316070556640625,
-0.02337646484375,
0.005329132080078125,
0.017822265625,
0.059600830078125,
0.05108642578125,
-0.06134033203125,
-0.046539306640625,
0.001514434814453125,
-0.070068359375,
-0.00977325439453125,
-0.0186309814453125,
-0.034759521484375,
0.0167083740234375,
0.01038360595703125,
-0.0335693359375,
0.041717529296875,
0.0240325927734375,
-0.037322998046875,
0.03936767578125,
-0.020111083984375,
0.01137542724609375,
-0.09014892578125,
0.037567138671875,
0.032958984375,
0.007434844970703125,
-0.067138671875,
0.0114898681640625,
-0.0079498291015625,
0.0294036865234375,
-0.040924072265625,
0.06341552734375,
-0.03131103515625,
0.01702880859375,
-0.01332855224609375,
0.0242919921875,
0.0220489501953125,
0.044677734375,
0.0189666748046875,
0.0379638671875,
0.03070068359375,
-0.03436279296875,
0.0211181640625,
0.035308837890625,
-0.01221466064453125,
0.051483154296875,
-0.059417724609375,
0.00637054443359375,
-0.0055694580078125,
0.0125885009765625,
-0.036041259765625,
-0.0174407958984375,
0.03955078125,
-0.031341552734375,
0.048858642578125,
-0.007659912109375,
-0.030731201171875,
-0.0498046875,
-0.0210418701171875,
0.011474609375,
0.037506103515625,
-0.0284271240234375,
0.037689208984375,
0.01873779296875,
0.0232696533203125,
-0.0560302734375,
-0.052764892578125,
-0.0215606689453125,
-0.036529541015625,
-0.058502197265625,
0.03857421875,
-0.01387786865234375,
-0.007572174072265625,
-0.0202789306640625,
-0.0063934326171875,
-0.0159912109375,
0.00913238525390625,
0.029632568359375,
0.035858154296875,
-0.016998291015625,
-0.0128021240234375,
-0.0185394287109375,
-0.0087890625,
0.0095977783203125,
-0.0009045600891113281,
0.05474853515625,
-0.030487060546875,
-0.0033931732177734375,
-0.09918212890625,
0.00406646728515625,
0.039794921875,
-0.021820068359375,
0.06573486328125,
0.0787353515625,
-0.022125244140625,
0.01282501220703125,
-0.042083740234375,
-0.00762176513671875,
-0.0384521484375,
-0.0166015625,
-0.03765869140625,
-0.03314208984375,
0.050018310546875,
-0.00026798248291015625,
-0.01506805419921875,
0.041168212890625,
0.0265655517578125,
-0.0212249755859375,
0.05377197265625,
0.0281982421875,
-0.028167724609375,
0.033050537109375,
-0.05743408203125,
0.005584716796875,
-0.099853515625,
-0.040557861328125,
-0.036590576171875,
-0.03692626953125,
-0.034423828125,
-0.0234222412109375,
0.01207733154296875,
0.037506103515625,
-0.04632568359375,
0.042877197265625,
-0.049285888671875,
0.01163482666015625,
0.035125732421875,
0.044464111328125,
-0.0051727294921875,
-0.01239013671875,
-0.0293426513671875,
-0.001209259033203125,
-0.0262603759765625,
-0.0491943359375,
0.0689697265625,
0.0296173095703125,
0.034515380859375,
0.007602691650390625,
0.06005859375,
0.003032684326171875,
0.0028533935546875,
-0.0313720703125,
0.0340576171875,
-0.004810333251953125,
-0.02880859375,
-0.0229034423828125,
-0.0262298583984375,
-0.072998046875,
0.00605010986328125,
-0.0350341796875,
-0.0836181640625,
0.017181396484375,
0.0150909423828125,
-0.0328369140625,
0.03533935546875,
-0.035552978515625,
0.068115234375,
-0.024932861328125,
-0.06689453125,
0.024688720703125,
-0.046539306640625,
0.01123046875,
0.0285186767578125,
0.0193634033203125,
-0.0012454986572265625,
0.01043701171875,
0.05206298828125,
-0.047271728515625,
0.068359375,
-0.0196380615234375,
-0.006511688232421875,
0.038482666015625,
-0.0138092041015625,
0.04241943359375,
-0.000690460205078125,
-0.025848388671875,
-0.00896453857421875,
-0.005268096923828125,
-0.031402587890625,
-0.035614013671875,
0.058502197265625,
-0.07159423828125,
-0.037384033203125,
-0.03887939453125,
-0.0277099609375,
0.01488494873046875,
0.01235198974609375,
0.0269622802734375,
0.036651611328125,
0.00315093994140625,
0.0069427490234375,
0.054046630859375,
-0.0160980224609375,
0.045501708984375,
0.01340484619140625,
0.004413604736328125,
-0.0172119140625,
0.06231689453125,
-0.003955841064453125,
0.0097198486328125,
0.042236328125,
0.0176239013671875,
-0.03326416015625,
-0.0205078125,
-0.045318603515625,
0.043121337890625,
-0.0223846435546875,
-0.01690673828125,
-0.043487548828125,
-0.0241546630859375,
-0.0267486572265625,
-0.028961181640625,
-0.0132598876953125,
-0.029327392578125,
-0.049957275390625,
-0.004322052001953125,
0.03570556640625,
0.036407470703125,
-0.0170135498046875,
0.0235443115234375,
-0.03857421875,
0.015228271484375,
0.01207733154296875,
0.007755279541015625,
0.00975799560546875,
-0.03533935546875,
-0.0083160400390625,
0.0223846435546875,
-0.0372314453125,
-0.050689697265625,
0.05084228515625,
-0.009796142578125,
0.042816162109375,
0.032073974609375,
0.002712249755859375,
0.05853271484375,
-0.022613525390625,
0.04345703125,
0.026153564453125,
-0.07177734375,
0.048248291015625,
-0.0290374755859375,
0.032867431640625,
0.035247802734375,
0.039093017578125,
-0.0271453857421875,
-0.0163726806640625,
-0.04693603515625,
-0.05474853515625,
0.061614990234375,
0.0215301513671875,
0.000812530517578125,
0.006256103515625,
0.038726806640625,
-0.0323486328125,
-0.003978729248046875,
-0.07415771484375,
-0.04638671875,
-0.0305938720703125,
-0.006999969482421875,
0.02642822265625,
-0.0029315948486328125,
-0.0102081298828125,
-0.03515625,
0.06353759765625,
-0.005016326904296875,
0.043731689453125,
0.016754150390625,
-0.00885772705078125,
-0.0049285888671875,
0.0225830078125,
0.059783935546875,
0.034423828125,
-0.025421142578125,
-0.019744873046875,
0.0234222412109375,
-0.032867431640625,
0.0017328262329101562,
-0.00728607177734375,
-0.028564453125,
-0.005908966064453125,
0.0172576904296875,
0.07586669921875,
0.017059326171875,
-0.0125885009765625,
0.036956787109375,
0.0076751708984375,
-0.01488494873046875,
-0.0256500244140625,
0.0148162841796875,
0.01837158203125,
0.0272064208984375,
0.0016117095947265625,
0.00965118408203125,
0.001651763916015625,
-0.04193115234375,
0.0198211669921875,
0.0282135009765625,
-0.0277099609375,
-0.017425537109375,
0.06427001953125,
-0.004070281982421875,
-0.01366424560546875,
0.022705078125,
-0.014984130859375,
-0.058685302734375,
0.0469970703125,
0.053802490234375,
0.042755126953125,
-0.02276611328125,
0.025543212890625,
0.0706787109375,
-0.0034732818603515625,
-0.0082855224609375,
0.01226043701171875,
0.0017375946044921875,
-0.04534912109375,
0.0055084228515625,
-0.07623291015625,
-0.0003485679626464844,
0.0194244384765625,
-0.0750732421875,
0.026153564453125,
-0.036529541015625,
-0.030731201171875,
-0.0032291412353515625,
0.0286407470703125,
-0.05230712890625,
0.049041748046875,
0.011444091796875,
0.059112548828125,
-0.049835205078125,
0.0770263671875,
0.03802490234375,
-0.0555419921875,
-0.0679931640625,
0.00653076171875,
0.001506805419921875,
-0.0712890625,
0.061004638671875,
0.0010986328125,
-0.0017633438110351562,
-0.007266998291015625,
-0.0233612060546875,
-0.053680419921875,
0.10040283203125,
-0.0091552734375,
-0.019500732421875,
-0.0200347900390625,
0.0207366943359375,
0.049652099609375,
-0.032867431640625,
0.054443359375,
0.0374755859375,
0.050750732421875,
0.004878997802734375,
-0.0640869140625,
0.043212890625,
-0.04541015625,
0.00403594970703125,
-0.0001990795135498047,
-0.10333251953125,
0.07781982421875,
0.00403594970703125,
0.0006589889526367188,
0.01904296875,
0.0372314453125,
0.0255126953125,
0.017120361328125,
0.01123046875,
0.057281494140625,
0.041351318359375,
-0.01702880859375,
0.0819091796875,
-0.032806396484375,
0.039825439453125,
0.07415771484375,
0.0036029815673828125,
0.0709228515625,
0.0144805908203125,
-0.018524169921875,
0.05694580078125,
0.029998779296875,
-0.0254669189453125,
0.01482391357421875,
0.021209716796875,
-0.012725830078125,
-0.009765625,
-0.0074462890625,
-0.03936767578125,
0.0172576904296875,
0.0284423828125,
-0.03955078125,
0.00696563720703125,
-0.022674560546875,
0.031768798828125,
0.00948333740234375,
-0.0184173583984375,
0.0404052734375,
0.01296234130859375,
-0.03424072265625,
0.065673828125,
0.0006885528564453125,
0.052093505859375,
-0.035919189453125,
0.0157012939453125,
-0.01274871826171875,
0.01032257080078125,
-0.024261474609375,
-0.047454833984375,
0.00728607177734375,
0.00801849365234375,
-0.00843048095703125,
-0.0250091552734375,
0.0350341796875,
-0.0158233642578125,
-0.04693603515625,
0.030487060546875,
0.0169677734375,
0.0098724365234375,
0.0245513916015625,
-0.0926513671875,
0.024078369140625,
0.0245513916015625,
-0.031890869140625,
0.0245819091796875,
0.014923095703125,
0.01873779296875,
0.0498046875,
0.036651611328125,
-0.0027923583984375,
0.0111846923828125,
-0.002044677734375,
0.06573486328125,
-0.033935546875,
-0.00716400146484375,
-0.068603515625,
0.05908203125,
-0.030548095703125,
-0.0229644775390625,
0.072021484375,
0.04473876953125,
0.05377197265625,
-0.00981903076171875,
0.05059814453125,
-0.01617431640625,
0.0253448486328125,
-0.0455322265625,
0.07147216796875,
-0.04840087890625,
0.010711669921875,
-0.03375244140625,
-0.04791259765625,
-0.0150909423828125,
0.0753173828125,
-0.0183563232421875,
0.0162811279296875,
0.04974365234375,
0.0565185546875,
0.0022029876708984375,
-0.00514984130859375,
-0.007049560546875,
0.0190582275390625,
-0.0026531219482421875,
0.0689697265625,
0.03900146484375,
-0.06390380859375,
0.01227569580078125,
-0.043914794921875,
-0.006580352783203125,
-0.0261383056640625,
-0.05267333984375,
-0.08221435546875,
-0.04754638671875,
-0.03802490234375,
-0.041168212890625,
-0.005634307861328125,
0.07110595703125,
0.044464111328125,
-0.061798095703125,
-0.0270233154296875,
0.006580352783203125,
0.0014171600341796875,
-0.007965087890625,
-0.019500732421875,
0.057281494140625,
0.0009202957153320312,
-0.0784912109375,
0.006195068359375,
-0.007549285888671875,
0.0396728515625,
0.013763427734375,
-0.02178955078125,
-0.034942626953125,
0.01061248779296875,
0.0153350830078125,
0.04150390625,
-0.0440673828125,
-0.0221405029296875,
-0.002948760986328125,
-0.01910400390625,
0.0162811279296875,
0.0205535888671875,
-0.033050537109375,
0.009765625,
0.038482666015625,
0.01445770263671875,
0.055908203125,
0.016326904296875,
0.0223236083984375,
-0.03643798828125,
0.00893402099609375,
-0.0082855224609375,
0.03350830078125,
0.00888824462890625,
-0.031768798828125,
0.040924072265625,
0.0160675048828125,
-0.035797119140625,
-0.05621337890625,
-0.0081634521484375,
-0.09307861328125,
-0.0034503936767578125,
0.0841064453125,
-0.024810791015625,
-0.039306640625,
0.0235443115234375,
-0.02301025390625,
0.037261962890625,
-0.03497314453125,
0.04034423828125,
0.04736328125,
-0.027191162109375,
-0.01178741455078125,
-0.046112060546875,
0.0494384765625,
0.0172119140625,
-0.0611572265625,
-0.020965576171875,
0.01517486572265625,
0.0216217041015625,
0.0335693359375,
0.032867431640625,
-0.006793975830078125,
0.0098419189453125,
-0.0100555419921875,
0.0024166107177734375,
-0.00640106201171875,
0.0007987022399902344,
-0.0086822509765625,
0.00127410888671875,
-0.020904541015625,
-0.00917816162109375
]
] |
RWKV/rwkv-4-1b5-pile | 2023-05-15T10:01:06.000Z | [
"transformers",
"pytorch",
"rwkv",
"text-generation",
"dataset:EleutherAI/pile",
"endpoints_compatible",
"has_space",
"region:us"
] | text-generation | RWKV | null | null | RWKV/rwkv-4-1b5-pile | 5 | 5,926 | transformers | 2023-05-04T13:42:33 | ---
datasets:
- EleutherAI/pile
---
# Model card for RWKV-4 | 1B5 parameters trained on Pile dataset
RWKV is a project led by [Bo Peng](https://github.com/BlinkDL). Learn more about the model architecture in the blogposts from Johan Wind [here](https://johanwind.github.io/2023/03/23/rwkv_overview.html) and [here](https://johanwind.github.io/2023/03/23/rwkv_details.html). Learn more about the project by joining the [RWKV discord server](https://discordapp.com/users/468093332535640064).
# Table of contents
0. [TL;DR](#TL;DR)
1. [Model Details](#model-details)
2. [Usage](#usage)
3. [Citation](#citation)
## TL;DR
Below is the description from the [original repository](https://github.com/BlinkDL/RWKV-LM)
> RWKV is an RNN with transformer-level LLM performance. It can be directly trained like a GPT (parallelizable). It's combining the best of RNN and transformer - great performance, fast inference, saves VRAM, fast training, "infinite" ctx_len, and free sentence embedding.
## Model Details
The details of the architecture can be found on the blogpost mentioned above and the Hugging Face blogpost of the integration.
## Usage
### Convert the raw weights to the HF format
You can use the [`convert_rwkv_checkpoint_to_hf.py`](https://github.com/huggingface/transformers/tree/main/src/transformers/models/rwkv/convert_rwkv_checkpoint_to_hf.py) script by specifying the repo_id of the original weights, the filename and the output directory. You can also optionally directly push the converted model on the Hub by passing `--push_to_hub` flag and `--model_name` argument to specify where to push the converted weights.
```bash
python convert_rwkv_checkpoint_to_hf.py --repo_id RAW_HUB_REPO --checkpoint_file RAW_FILE --output_dir OUTPUT_DIR --push_to_hub --model_name dummy_user/converted-rwkv
```
### Generate text
You can use the `AutoModelForCausalLM` and `AutoTokenizer` classes to generate texts from the model. Expand the sections below to understand how to run the model in different scenarios:
### Running the model on a CPU
<details>
<summary> Click to expand </summary>
```python
from transformers import AutoModelForCausalLM, AutoTokenizer
model = AutoModelForCausalLM.from_pretrained("RWKV/rwkv-4-1b5-pile")
tokenizer = AutoTokenizer.from_pretrained("RWKV/rwkv-4-1b5-pile")
prompt = "\nIn a shocking finding, scientist discovered a herd of dragons living in a remote, previously unexplored valley, in Tibet. Even more surprising to the researchers was the fact that the dragons spoke perfect Chinese."
inputs = tokenizer(prompt, return_tensors="pt")
output = model.generate(inputs["input_ids"], max_new_tokens=40)
print(tokenizer.decode(output[0].tolist(), skip_special_tokens=True))
```
### Running the model on a single GPU
<details>
<summary> Click to expand </summary>
```python
from transformers import AutoModelForCausalLM, AutoTokenizer
model = AutoModelForCausalLM.from_pretrained("RWKV/rwkv-4-1b5-pile").to(0)
tokenizer = AutoTokenizer.from_pretrained("RWKV/rwkv-4-1b5-pile")
prompt = "\nIn a shocking finding, scientist discovered a herd of dragons living in a remote, previously unexplored valley, in Tibet. Even more surprising to the researchers was the fact that the dragons spoke perfect Chinese."
inputs = tokenizer(prompt, return_tensors="pt").to(0)
output = model.generate(inputs["input_ids"], max_new_tokens=40)
print(tokenizer.decode(output[0].tolist(), skip_special_tokens=True))
```
</details>
</details>
### Running the model in half-precision, on GPU
<details>
<summary> Click to expand </summary>
```python
import torch
from transformers import AutoModelForCausalLM, AutoTokenizer
model = AutoModelForCausalLM.from_pretrained("RWKV/rwkv-4-1b5-pile", torch_dtype=torch.float16).to(0)
tokenizer = AutoTokenizer.from_pretrained("RWKV/rwkv-4-1b5-pile")
prompt = "\nIn a shocking finding, scientist discovered a herd of dragons living in a remote, previously unexplored valley, in Tibet. Even more surprising to the researchers was the fact that the dragons spoke perfect Chinese."
inputs = tokenizer(prompt, return_tensors="pt").to(0)
output = model.generate(inputs["input_ids"], max_new_tokens=40)
print(tokenizer.decode(output[0].tolist(), skip_special_tokens=True))
```
</details>
### Running the model multiple GPUs
<details>
<summary> Click to expand </summary>
```python
# pip install accelerate
from transformers import AutoModelForCausalLM, AutoTokenizer
model = AutoModelForCausalLM.from_pretrained("RWKV/rwkv-4-1b5-pile", device_map="auto")
tokenizer = AutoTokenizer.from_pretrained("RWKV/rwkv-4-1b5-pile")
prompt = "\nIn a shocking finding, scientist discovered a herd of dragons living in a remote, previously unexplored valley, in Tibet. Even more surprising to the researchers was the fact that the dragons spoke perfect Chinese."
inputs = tokenizer(prompt, return_tensors="pt").to(0)
output = model.generate(inputs["input_ids"], max_new_tokens=40)
print(tokenizer.decode(output[0].tolist(), skip_special_tokens=True))
```
</details>
## Citation
If you use this model, please consider citing the original work, from the original repo [here](https://github.com/BlinkDL/ChatRWKV/) | 5,294 | [
[
-0.0287322998046875,
-0.0430908203125,
-0.001308441162109375,
0.012451171875,
-0.0182342529296875,
-0.0240325927734375,
-0.01020050048828125,
-0.02288818359375,
-0.006023406982421875,
0.0187835693359375,
-0.04034423828125,
-0.02667236328125,
-0.037384033203125,
-0.00008112192153930664,
-0.038116455078125,
0.07257080078125,
-0.003276824951171875,
0.005107879638671875,
0.01523590087890625,
-0.006999969482421875,
-0.00989532470703125,
-0.0224151611328125,
-0.040435791015625,
-0.047088623046875,
0.02874755859375,
-0.02490234375,
0.05169677734375,
0.08270263671875,
0.0213775634765625,
0.027862548828125,
-0.01107025146484375,
0.007183074951171875,
-0.027923583984375,
-0.0161285400390625,
0.0098876953125,
-0.0182037353515625,
-0.0221710205078125,
0.01401519775390625,
0.05267333984375,
0.0204925537109375,
-0.0265655517578125,
0.01776123046875,
0.00469207763671875,
0.0190887451171875,
-0.01500701904296875,
0.0164031982421875,
-0.0284881591796875,
0.0261993408203125,
0.0006732940673828125,
-0.00860595703125,
-0.0241546630859375,
-0.00591278076171875,
0.00548553466796875,
-0.0765380859375,
0.04083251953125,
0.0014867782592773438,
0.10101318359375,
0.045379638671875,
-0.01279449462890625,
0.002262115478515625,
-0.04071044921875,
0.061248779296875,
-0.07696533203125,
0.0214080810546875,
0.005157470703125,
0.00458526611328125,
-0.01297760009765625,
-0.07489013671875,
-0.053009033203125,
-0.0175018310546875,
-0.0223846435546875,
0.01348876953125,
-0.0117645263671875,
0.00006812810897827148,
0.043670654296875,
0.039886474609375,
-0.0452880859375,
0.004215240478515625,
-0.04156494140625,
-0.025421142578125,
0.044342041015625,
0.02545166015625,
0.034515380859375,
-0.039794921875,
-0.034698486328125,
-0.041412353515625,
-0.038177490234375,
0.01175689697265625,
0.0252838134765625,
0.022491455078125,
-0.0277252197265625,
0.03741455078125,
-0.0128173828125,
0.056121826171875,
0.0249176025390625,
-0.0001004338264465332,
0.0231170654296875,
-0.0231170654296875,
-0.0302886962890625,
-0.0168914794921875,
0.08160400390625,
0.0085906982421875,
-0.004634857177734375,
-0.00893402099609375,
-0.01480865478515625,
-0.0228271484375,
0.012786865234375,
-0.08123779296875,
-0.046356201171875,
0.0153961181640625,
-0.061126708984375,
-0.03265380859375,
-0.0033321380615234375,
-0.050201416015625,
-0.0193328857421875,
-0.0007390975952148438,
0.049407958984375,
-0.0305633544921875,
-0.045989990234375,
-0.0028057098388671875,
-0.029754638671875,
0.045501708984375,
0.00299072265625,
-0.07562255859375,
-0.00897979736328125,
0.03955078125,
0.06109619140625,
-0.0024871826171875,
-0.054595947265625,
-0.0211029052734375,
-0.000008881092071533203,
-0.022735595703125,
0.045196533203125,
-0.004970550537109375,
-0.03924560546875,
-0.022735595703125,
0.0238800048828125,
-0.0143890380859375,
-0.034698486328125,
0.036041259765625,
-0.023468017578125,
0.03436279296875,
-0.0207061767578125,
-0.035064697265625,
-0.02734375,
0.0157470703125,
-0.040008544921875,
0.102783203125,
0.01093292236328125,
-0.0714111328125,
0.0193939208984375,
-0.037109375,
-0.022979736328125,
0.01245880126953125,
-0.0007257461547851562,
-0.043182373046875,
-0.007030487060546875,
0.0213775634765625,
0.03240966796875,
-0.00981903076171875,
0.0184326171875,
-0.01580810546875,
-0.0362548828125,
0.0113677978515625,
-0.039398193359375,
0.08343505859375,
0.0216522216796875,
-0.0426025390625,
0.0216064453125,
-0.044403076171875,
0.005725860595703125,
0.0134735107421875,
-0.039794921875,
0.006572723388671875,
-0.0006079673767089844,
0.0038089752197265625,
0.00992584228515625,
0.0187835693359375,
-0.0335693359375,
0.017578125,
-0.03790283203125,
0.05609130859375,
0.05535888671875,
-0.0215911865234375,
0.0196075439453125,
-0.0193328857421875,
0.0188751220703125,
-0.0005407333374023438,
0.018829345703125,
-0.0146026611328125,
-0.04095458984375,
-0.07110595703125,
-0.0170135498046875,
0.022064208984375,
0.0300140380859375,
-0.056121826171875,
0.042572021484375,
-0.01727294921875,
-0.05535888671875,
-0.0455322265625,
-0.017791748046875,
0.0102996826171875,
0.044830322265625,
0.032318115234375,
0.0003459453582763672,
-0.0260467529296875,
-0.046417236328125,
-0.014678955078125,
-0.0201416015625,
-0.01087188720703125,
0.0267791748046875,
0.044891357421875,
-0.0194854736328125,
0.05487060546875,
-0.03466796875,
-0.0093231201171875,
-0.01495361328125,
0.0240936279296875,
0.031463623046875,
0.055999755859375,
0.031005859375,
-0.04876708984375,
-0.0282745361328125,
0.0037078857421875,
-0.07220458984375,
0.01230621337890625,
-0.0121307373046875,
-0.007633209228515625,
-0.0007309913635253906,
0.027099609375,
-0.05963134765625,
0.0308837890625,
0.036834716796875,
-0.01904296875,
0.054412841796875,
-0.030242919921875,
0.01001739501953125,
-0.080810546875,
0.0239715576171875,
-0.006992340087890625,
-0.0008187294006347656,
-0.038970947265625,
0.00408935546875,
0.0084075927734375,
-0.014404296875,
-0.031646728515625,
0.054931640625,
-0.0280609130859375,
0.0013637542724609375,
-0.0182342529296875,
-0.00994110107421875,
-0.0028533935546875,
0.054290771484375,
0.00383758544921875,
0.05572509765625,
0.058868408203125,
-0.053314208984375,
0.04644775390625,
0.0306243896484375,
-0.0198516845703125,
-0.00022673606872558594,
-0.0662841796875,
0.0013990402221679688,
0.00453948974609375,
0.0236663818359375,
-0.057647705078125,
-0.021514892578125,
0.034942626953125,
-0.05609130859375,
0.0281524658203125,
-0.02105712890625,
-0.02587890625,
-0.041351318359375,
-0.0083160400390625,
0.0401611328125,
0.050384521484375,
-0.065673828125,
0.06402587890625,
0.0197601318359375,
0.0159149169921875,
-0.061676025390625,
-0.06451416015625,
-0.0005435943603515625,
-0.0245361328125,
-0.042510986328125,
0.036834716796875,
0.0004649162292480469,
0.01454925537109375,
0.0125732421875,
0.01092529296875,
-0.007160186767578125,
-0.00890350341796875,
0.0226898193359375,
0.03460693359375,
-0.01715087890625,
-0.00444793701171875,
-0.0225067138671875,
-0.0229034423828125,
0.019073486328125,
-0.0321044921875,
0.042083740234375,
-0.00963592529296875,
-0.01020050048828125,
-0.0538330078125,
-0.002178192138671875,
0.03717041015625,
-0.0095062255859375,
0.05511474609375,
0.0804443359375,
-0.0321044921875,
-0.022064208984375,
-0.0291748046875,
-0.0262298583984375,
-0.039459228515625,
0.044769287109375,
-0.015045166015625,
-0.033447265625,
0.0494384765625,
0.008148193359375,
0.00926971435546875,
0.0623779296875,
0.042266845703125,
0.0024089813232421875,
0.08135986328125,
0.047760009765625,
-0.00860595703125,
0.032745361328125,
-0.049407958984375,
0.019989013671875,
-0.05841064453125,
-0.0238037109375,
-0.03143310546875,
-0.00010770559310913086,
-0.046783447265625,
-0.03558349609375,
0.0185546875,
0.0070037841796875,
-0.0391845703125,
0.0271453857421875,
-0.0694580078125,
0.0107269287109375,
0.036041259765625,
0.001934051513671875,
-0.01035308837890625,
0.0022449493408203125,
-0.01386260986328125,
0.00818634033203125,
-0.07513427734375,
-0.01739501953125,
0.06597900390625,
0.029083251953125,
0.05194091796875,
-0.0246429443359375,
0.031463623046875,
0.0107421875,
0.025909423828125,
-0.045654296875,
0.035797119140625,
-0.01319122314453125,
-0.049896240234375,
-0.026031494140625,
-0.04315185546875,
-0.05419921875,
0.03692626953125,
-0.01459503173828125,
-0.02606201171875,
0.006252288818359375,
0.01189422607421875,
-0.044036865234375,
0.05047607421875,
-0.035125732421875,
0.08233642578125,
-0.006290435791015625,
-0.017425537109375,
-0.0094146728515625,
-0.034698486328125,
0.0338134765625,
0.019317626953125,
-0.00015807151794433594,
0.00016796588897705078,
0.0193328857421875,
0.072265625,
-0.046417236328125,
0.0606689453125,
-0.02423095703125,
0.005672454833984375,
0.03216552734375,
-0.0248870849609375,
0.04425048828125,
-0.01132965087890625,
-0.008209228515625,
0.028228759765625,
-0.0023250579833984375,
-0.0222015380859375,
-0.022186279296875,
0.060882568359375,
-0.0841064453125,
-0.0301055908203125,
-0.032623291015625,
-0.044769287109375,
0.03656005859375,
0.0215911865234375,
0.041534423828125,
0.0303192138671875,
-0.002162933349609375,
-0.00458526611328125,
0.048309326171875,
-0.041534423828125,
0.0562744140625,
0.01605224609375,
-0.00682830810546875,
-0.0433349609375,
0.0648193359375,
0.0029850006103515625,
0.00641632080078125,
0.00130462646484375,
0.024139404296875,
-0.03338623046875,
-0.02734375,
-0.0626220703125,
0.03021240234375,
-0.060638427734375,
-0.01195526123046875,
-0.057647705078125,
-0.04278564453125,
-0.040283203125,
0.007266998291015625,
-0.035491943359375,
-0.012969970703125,
-0.03814697265625,
0.005889892578125,
0.0294342041015625,
0.05438232421875,
-0.0207977294921875,
0.016754150390625,
-0.054595947265625,
0.0253448486328125,
0.033111572265625,
0.00957489013671875,
0.020355224609375,
-0.06829833984375,
-0.01776123046875,
0.00705718994140625,
-0.0110931396484375,
-0.0460205078125,
0.052764892578125,
-0.004138946533203125,
0.05047607421875,
0.02362060546875,
-0.0020198822021484375,
0.07122802734375,
-0.0233001708984375,
0.072265625,
0.004852294921875,
-0.062164306640625,
0.009185791015625,
-0.025970458984375,
0.014801025390625,
0,
0.010040283203125,
-0.038848876953125,
-0.003658294677734375,
-0.0421142578125,
-0.058868408203125,
0.0552978515625,
0.0076141357421875,
0.0085906982421875,
0.0200347900390625,
0.039703369140625,
-0.00542449951171875,
-0.0037937164306640625,
-0.08465576171875,
-0.04058837890625,
-0.050018310546875,
0.00029850006103515625,
0.007770538330078125,
0.0026264190673828125,
-0.004364013671875,
-0.049835205078125,
0.0692138671875,
0.0002484321594238281,
0.039703369140625,
0.0261993408203125,
-0.0023345947265625,
-0.01102447509765625,
-0.01447296142578125,
0.0306243896484375,
0.027008056640625,
0.00428009033203125,
-0.0066986083984375,
0.032928466796875,
-0.043670654296875,
-0.008697509765625,
0.029998779296875,
-0.0257110595703125,
0.0007767677307128906,
0.0196380615234375,
0.068603515625,
-0.007015228271484375,
-0.0129241943359375,
0.026153564453125,
-0.0157928466796875,
-0.0181884765625,
-0.03173828125,
0.00847625732421875,
0.017913818359375,
0.0254058837890625,
0.039794921875,
0.01045989990234375,
-0.01270294189453125,
-0.006313323974609375,
0.0096588134765625,
0.03289794921875,
-0.022369384765625,
-0.0198974609375,
0.0806884765625,
0.0160980224609375,
-0.0157012939453125,
0.074462890625,
-0.0150909423828125,
-0.0472412109375,
0.06201171875,
0.03253173828125,
0.07159423828125,
-0.009521484375,
0.01031494140625,
0.0650634765625,
0.02630615234375,
-0.019317626953125,
-0.00673675537109375,
-0.007007598876953125,
-0.049102783203125,
-0.03631591796875,
-0.061126708984375,
-0.00492095947265625,
0.017608642578125,
-0.045196533203125,
0.033416748046875,
-0.0163421630859375,
0.0010900497436523438,
-0.00284576416015625,
0.00010883808135986328,
-0.04150390625,
0.01471710205078125,
0.0101776123046875,
0.0665283203125,
-0.06256103515625,
0.078125,
0.036102294921875,
-0.0282135009765625,
-0.08489990234375,
-0.00308990478515625,
-0.0289154052734375,
-0.07257080078125,
0.043212890625,
0.0247955322265625,
-0.00024008750915527344,
0.02374267578125,
-0.046600341796875,
-0.06341552734375,
0.09613037109375,
0.0012292861938476562,
-0.01537322998046875,
-0.004436492919921875,
0.003635406494140625,
0.038360595703125,
-0.00649261474609375,
0.031036376953125,
0.02105712890625,
0.04107666015625,
0.01235198974609375,
-0.057708740234375,
0.01898193359375,
-0.0258026123046875,
-0.01108551025390625,
0.00901031494140625,
-0.0556640625,
0.10302734375,
-0.02813720703125,
-0.0265350341796875,
0.01617431640625,
0.07086181640625,
0.023284912109375,
-0.00794219970703125,
0.037567138671875,
0.052459716796875,
0.0472412109375,
-0.0172271728515625,
0.07733154296875,
-0.044403076171875,
0.060882568359375,
0.033782958984375,
0.0113677978515625,
0.044464111328125,
0.0160675048828125,
-0.01617431640625,
0.03253173828125,
0.0631103515625,
-0.039764404296875,
0.034515380859375,
0.01416015625,
-0.0159149169921875,
-0.019622802734375,
0.011505126953125,
-0.0478515625,
0.01239776611328125,
0.011322021484375,
-0.0219573974609375,
-0.0135345458984375,
-0.004917144775390625,
0.002651214599609375,
-0.03350830078125,
-0.0171661376953125,
0.0302886962890625,
-0.0023708343505859375,
-0.05718994140625,
0.06866455078125,
0.005077362060546875,
0.07232666015625,
-0.04901123046875,
-0.00502777099609375,
-0.0025539398193359375,
0.025360107421875,
-0.02276611328125,
-0.043853759765625,
0.0142822265625,
-0.01232147216796875,
-0.01058197021484375,
-0.017333984375,
0.04388427734375,
-0.037841796875,
-0.0390625,
0.0150299072265625,
0.005886077880859375,
0.030792236328125,
0.0036602020263671875,
-0.0762939453125,
-0.0017986297607421875,
0.0172119140625,
-0.03997802734375,
0.014495849609375,
0.0197906494140625,
0.025604248046875,
0.053436279296875,
0.06298828125,
0.005126953125,
0.025054931640625,
-0.013336181640625,
0.066650390625,
-0.062103271484375,
-0.0296478271484375,
-0.06500244140625,
0.03900146484375,
0.0022182464599609375,
-0.040130615234375,
0.06884765625,
0.045989990234375,
0.058349609375,
-0.005779266357421875,
0.061309814453125,
-0.02496337890625,
0.01355743408203125,
-0.020477294921875,
0.0794677734375,
-0.03790283203125,
0.003009796142578125,
0.00878143310546875,
-0.040252685546875,
0.00473785400390625,
0.066650390625,
0.0034275054931640625,
0.01526641845703125,
0.038848876953125,
0.069580078125,
0.0007910728454589844,
-0.004962921142578125,
0.01235198974609375,
0.036651611328125,
0.0291290283203125,
0.0257110595703125,
0.052642822265625,
-0.057891845703125,
0.05322265625,
-0.036834716796875,
-0.01003265380859375,
0.019439697265625,
-0.0655517578125,
-0.0697021484375,
-0.03643798828125,
-0.0278167724609375,
-0.042327880859375,
-0.011016845703125,
0.043182373046875,
0.059112548828125,
-0.0452880859375,
-0.020416259765625,
-0.014190673828125,
-0.002841949462890625,
-0.022064208984375,
-0.019500732421875,
0.041717529296875,
-0.019439697265625,
-0.05999755859375,
0.0208892822265625,
0.00273895263671875,
0.018585205078125,
-0.043548583984375,
-0.0193634033203125,
-0.00677490234375,
-0.016082763671875,
0.01024627685546875,
0.0305328369140625,
-0.0589599609375,
-0.0112457275390625,
0.001918792724609375,
-0.015106201171875,
0.0011606216430664062,
0.03826904296875,
-0.061737060546875,
0.0208892822265625,
0.044281005859375,
0.0290985107421875,
0.06805419921875,
-0.0013608932495117188,
0.039642333984375,
-0.0222930908203125,
0.0204925537109375,
-0.002231597900390625,
0.02142333984375,
0.032806396484375,
-0.033782958984375,
0.0198516845703125,
0.035064697265625,
-0.0633544921875,
-0.0692138671875,
-0.0144500732421875,
-0.057281494140625,
-0.029541015625,
0.086669921875,
-0.0265350341796875,
-0.032257080078125,
0.004852294921875,
-0.0017061233520507812,
0.0479736328125,
-0.0012121200561523438,
0.06597900390625,
0.039337158203125,
-0.00917816162109375,
-0.00547027587890625,
-0.047760009765625,
0.055816650390625,
0.025970458984375,
-0.0401611328125,
0.017120361328125,
0.00774383544921875,
0.049774169921875,
0.01386260986328125,
0.031402587890625,
-0.0013484954833984375,
0.011871337890625,
0.0184173583984375,
0.031280517578125,
-0.0264892578125,
0.00890350341796875,
-0.02850341796875,
-0.00457763671875,
-0.0296630859375,
-0.0254974365234375
]
] |
IDEA-CCNL/Ziya-LLaMA-13B-v1 | 2023-09-13T08:50:47.000Z | [
"transformers",
"pytorch",
"llama",
"text-generation",
"en",
"zh",
"arxiv:2210.08590",
"license:gpl-3.0",
"has_space",
"text-generation-inference",
"region:us"
] | text-generation | IDEA-CCNL | null | null | IDEA-CCNL/Ziya-LLaMA-13B-v1 | 252 | 5,926 | transformers | 2023-05-16T10:32:58 | ---
license: gpl-3.0
language:
- en
- zh
inference: false
---
# Ziya-LLaMA-13B-v1
- Main Page:[Fengshenbang](https://fengshenbang-lm.com/)
- Github: [Fengshenbang-LM](https://github.com/IDEA-CCNL/Fengshenbang-LM)
（LLaMA权重的许可证限制，我们无法直接发布完整的模型权重，用户需要参考[使用说明](#-使用-usage-)进行合并)
# 姜子牙系列模型
- [Ziya-LLaMA-13B-v1.1](https://huggingface.co/IDEA-CCNL/Ziya-LLaMA-13B-v1.1)
- [Ziya-LLaMA-13B-v1](https://huggingface.co/IDEA-CCNL/Ziya-LLaMA-13B-v1)
- [Ziya-LLaMA-7B-Reward](https://huggingface.co/IDEA-CCNL/Ziya-LLaMA-7B-Reward)
- [Ziya-LLaMA-13B-Pretrain-v1](https://huggingface.co/IDEA-CCNL/Ziya-LLaMA-13B-Pretrain-v1)
- [Ziya-BLIP2-14B-Visual-v1](https://huggingface.co/IDEA-CCNL/Ziya-BLIP2-14B-Visual-v1)
## 简介 Brief Introduction
姜子牙通用大模型V1是基于LLaMa的130亿参数的大规模预训练模型，具备翻译，编程，文本分类，信息抽取，摘要，文案生成，常识问答和数学计算等能力。目前姜子牙通用大模型已完成大规模预训练、多任务有监督微调和人类反馈学习三阶段的训练过程。
The Ziya-LLaMA-13B-v1 is a large-scale pre-trained model based on LLaMA with 13 billion parameters. It has the ability to perform tasks such as translation, programming, text classification, information extraction, summarization, copywriting, common sense Q&A, and mathematical calculation. The Ziya-LLaMA-13B-v1 has undergone three stages of training: large-scale continual pre-training (PT), multi-task supervised fine-tuning (SFT), and human feedback learning (RM, PPO).
## 软件依赖
```
pip install torch==1.12.1 tokenizers==0.13.3 git+https://github.com/huggingface/transformers
```
## 模型分类 Model Taxonomy
| 需求 Demand | 任务 Task | 系列 Series | 模型 Model | 参数 Parameter | 额外 Extra |
| :----: | :----: | :----: | :----: | :----: | :----: |
| 通用 General | AGI模型 | 姜子牙 Ziya | LLaMA | 13B | English&Chinese |
## 模型信息 Model Information
### 继续预训练 Continual pretraining
原始数据包含英文和中文，其中英文数据来自openwebtext、Books、Wikipedia和Code，中文数据来自清洗后的悟道数据集、自建的中文数据集。在对原始数据进行去重、模型打分、数据分桶、规则过滤、敏感主题过滤和数据评估后，最终得到125B tokens的有效数据。
为了解决LLaMA原生分词对中文编解码效率低下的问题，我们在LLaMA词表的基础上增加了7k+个常见中文字，通过和LLaMA原生的词表去重，最终得到一个39410大小的词表，并通过复用Transformers里LlamaTokenizer来实现了这一效果。
在增量训练过程中，我们使用了160张40GB的A100，采用2.6M tokens的训练集样本数量和FP 16的混合精度，吞吐量达到118 TFLOP per GPU per second。因此我们能够在8天的时间里在原生的LLaMA-13B模型基础上，增量训练110B tokens的数据。
训练期间，虽然遇到了机器宕机、底层框架bug、loss spike等各种问题，但我们通过快速调整，保证了增量训练的稳定性。我们也放出训练过程的loss曲线，让大家了解可能出现的问题。
The original data contains both English and Chinese, with English data from openwebtext, Books, Wikipedia, and Code, and Chinese data from the cleaned Wudao dataset and self-built Chinese dataset. After deduplication, model scoring, data bucketing, rule filtering, sensitive topic filtering, and data evaluation, we finally obtained 125 billion tokens of valid data.
To address the issue of low efficiency in Chinese encoding and decoding caused by the native word segmentation of LLaMa, we added 8,000 commonly used Chinese characters to the LLaMa vocabulary. By removing duplicates with the original LLaMa vocabulary, we finally obtained a vocabulary of size 39,410. We achieved this by reusing the LlamaTokenizer in Transformers.
During the incremental training process, we used 160 A100s with a total of 40GB memory, using a training dataset with 2.6 million tokens and mixed precision of FP16. The throughput reached 118 TFLOP per GPU per second. As a result, we were able to incrementally train 110 billion tokens of data on top of the native LLaMa-13B model in just 8 days.
Throughout the training process, we encountered various issues such as machine crashes, underlying framework bugs, and loss spikes. However, we ensured the stability of the incremental training by making rapid adjustments. We have also released the loss curve during the training process to help everyone understand the potential issues that may arise.
<img src="https://huggingface.co/datasets/suolyer/testb/blob/main/loss.png" width=1000 height=600>
### 多任务有监督微调 Supervised finetuning
在多任务有监督微调阶段，采用了课程学习（curiculum learning）和增量训练（continual learning）的策略，用大模型辅助划分已有的数据难度，然后通过“Easy To Hard”的方式，分多个阶段进行SFT训练。
SFT训练数据包含多个高质量的数据集，均经过人工筛选和校验：
- Self-Instruct构造的数据（约2M）：BELLE、Alpaca、Alpaca-GPT4等多个数据集
- 内部收集Code数据（300K）：包含leetcode、多种Code任务形式
- 内部收集推理/逻辑相关数据（500K）：推理、申论、数学应用题、数值计算等
- 中英平行语料（2M）：中英互译语料、COT类型翻译语料、古文翻译语料等
- 多轮对话语料（500K）：Self-Instruct生成、任务型多轮对话、Role-Playing型多轮对话等
During the supervised fine-tuning (SFT) phase of multitask learning, we used a strategy of curriculum learning and incremental training. We used the large model to assist in partitioning the existing data by difficulty and then conducted SFT training in multiple stages using the "easy to hard" approach.
The SFT training data consists of multiple high-quality datasets that have been manually selected and verified, including approximately 2 million samples from datasets such as BELLE, Alpaca, and Alpaca-GPT4, 300,000 samples of internally collected code data including LeetCode and various code tasks, 500,000 samples of internally collected inference/logic-related data such as reasoning, argumentative essays, mathematical application questions, and numerical calculations, 2 million samples of Chinese-English parallel corpora including translation, COT-type translation, and classical Chinese translation, and 500,000 samples of multi-turn dialogue corpora including self-instructed generation, task-oriented multi-turn dialogue, and role-playing multi-turn dialogue.
### 人类反馈学习 Human-Feedback training
为了进一步提升模型的综合表现，使其能够充分理解人类意图、减少“幻觉”和不安全的输出，基于指令微调后的模型，进行了人类反馈训练（Human-Feedback Training，HFT）。在训练中，我们采用了以人类反馈强化学习（RM、PPO）为主，结合多种其他手段联合训练的方法，手段包括人类反馈微调（Human-Feedback Fine-tuning，HFFT）、后见链微调（Chain-of-Hindsight Fine-tuning，COHFT）、AI反馈（AI Feedback）和基于规则的奖励系统（Rule-based Reward System，RBRS）等，用来弥补PPO方法的短板，加速训练。
我们在内部自研的框架上实现了HFT的训练流程，该框架可以利用最少8张40G的A100显卡完成Ziya-LLaMA-13B-v1的全参数训练。在PPO训练中，我们没有限制生成样本的长度，以确保长文本任务的奖励准确性。每次训练的总经验池尺寸超过100k样本，确保了训练的充分性。
To further improve the overall performance of the model, enabling it to fully understand human intentions, reduce "hallucinations" and unsafe outputs, we conducted Human-Feedback Training (HFT) based on the model fine-tuned with instructions. In the training process, we used a variety of methods, including human feedback reinforcement learning (RM, PPO), combined with other methods such as Human-Feedback Fine-tuning (HFFT), Chain-of-Hindsight Fine-tuning (COHFT), AI feedback, and Rule-based Reward System (RBRS), to supplement the shortcomings of the PPO method and accelerate training.
We implemented the HFT training process on an internally developed framework, which can use a minimum of 8 40GB A100 GPUs to complete the full parameter training of Ziya-LLaMA-13B-v1. In the PPO training, we did not limit the length of the generated samples to ensure the accuracy of rewards for long-text tasks. The total experience pool size for each training exceeded 100k samples, ensuring the sufficiency of the training.
### 效果评估 Performance
<img src="https://huggingface.co/IDEA-CCNL/Ziya-LLaMA-13B-v1/resolve/main/pk.png" width=1000 height=600>
## <span id="jump"> 使用 Usage </span>
由于LLaMA权重的许可限制，该模型不能用于商业用途，请严格遵守LLaMA的使用政策。考虑到LLaMA权重的许可证限制，我们无法直接发布完整的模型权重。因此，我们使用了[FastChat开源工具](https://github.com/lm-sys/FastChat/blob/main/fastchat/model/apply_delta.py)作为基础，并对其进行了进一步的优化。我们计算并发布了Ziya-LLaMA-13B-v1权重与原始LLaMA权重之间的差值。用户可以按照以下步骤操作以获得Ziya-LLaMA-13B-v1完整权重，具体步骤如下：
Step 1:获取[LLaMA](https://docs.google.com/forms/d/e/1FAIpQLSfqNECQnMkycAp2jP4Z9TFX0cGR4uf7b_fBxjY_OjhJILlKGA/viewform)权重并转成Hugging Face Transformers模型格式，可参考转换[脚本](https://github.com/huggingface/transformers/blob/main/src/transformers/models/llama/convert_llama_weights_to_hf.py)（若已经有huggingface权重则跳过）
```
python src/transformers/models/llama/convert_llama_weights_to_hf.py \
--input_dir /path/to/downloaded/llama/weights --model_size 13B --output_dir /output/path
```
Step 2:下载Ziya-LLaMA-13B-v1的delta权重以及step 1中转换好的原始LLaMA权重，使用如下脚本转换：https://github.com/IDEA-CCNL/Fengshenbang-LM/blob/main/fengshen/utils/apply_delta.py
```
python3 -m apply_delta --base ~/model_weights/llama-13b --target ~/model_weights/Ziya-LLaMA-13B --delta ~/model_weights/Ziya-LLaMA-13B-v1
```
Step 3: 加载step 2得到的模型推理
```python3
from transformers import AutoTokenizer
from transformers import LlamaForCausalLM
import torch
device = torch.device("cuda")
ckpt = '基于delta参数合并后的完整模型权重'
query="帮我写一份去西安的旅游计划"
model = LlamaForCausalLM.from_pretrained(ckpt, torch_dtype=torch.float16, device_map="auto")
tokenizer = AutoTokenizer.from_pretrained(ckpt, use_fast=False)
inputs = '<human>:' + query.strip() + '\n<bot>:'
input_ids = tokenizer(inputs, return_tensors="pt").input_ids.to(device)
generate_ids = model.generate(
input_ids,
max_new_tokens=1024,
do_sample = True,
top_p = 0.85,
temperature = 1.0,
repetition_penalty=1.,
eos_token_id=2,
bos_token_id=1,
pad_token_id=0)
output = tokenizer.batch_decode(generate_ids)[0]
print(output)
```
NOTE: Due to the licensing restrictions of LLaMA weights, the utilization of the model for commercial purposes is precluded. Please strictly respect LLaMA's usage policy. Considering the licensing limitations on LLaMA weights, we are unable to directly release the complete model weights. Therefore, we utilized [the open-source FastChat tool](https://github.com/lm-sys/FastChat/blob/main/fastchat/model/apply_delta.py) and further optimized it to calculate the differences between Ziya-LLaMA-13B-v1 weights and the original LLaMA weights. Users can follow the steps to obtain the complete weights of Ziya-LLaMA-13B-v1. The steps are as follows:
Step 1: Obtain the [LLaMA](https://huggingface.co/docs/transformers/main/en/model_doc/llama#overview) weights and convert them into the Hugging Face Transformers format. You can refer to the [script](https://github.com/huggingface/transformers/blob/main/src/transformers/models/llama/convert_llama_weights_to_hf.py) (skip this step if you already have the Hugging Face weights).
```
python src/transformers/models/llama/convert_llama_weights_to_hf.py \
--input_dir /path/to/downloaded/llama/weights --model_size 13B --output_dir /output/path
```
Step 2: Download the delta weights for Ziya-LLaMA-13B-v1 and the pre-converted original LLaMA weights from step 1. Use the following script for conversion: https://github.com/IDEA-CCNL/Fengshenbang-LM/blob/main/fengshen/utils/apply_delta.py
```
python3 -m apply_delta --base ~/model_weights/llama-13b --target ~/model_weights/Ziya-LLaMA-13B --delta ~/model_weights/Ziya-LLaMA-13B-v1(huggingface下载)
```
Step 3: Load the model obtained in Step 2 for inference.
```python3
from transformers import AutoTokenizer
from transformers import LlamaForCausalLM
import torch
device = torch.device("cuda")
ckpt = '基于delta合并后完整模型权重'
query="帮我写一份去西安的旅游计划"
model = LlamaForCausalLM.from_pretrained(ckpt, torch_dtype=torch.float16, device_map="auto")
tokenizer = AutoTokenizer.from_pretrained(ckpt, use_fast=False)
inputs = '<human>:' + query.strip() + '\n<bot>:'
input_ids = tokenizer(inputs, return_tensors="pt").input_ids.to(device)
generate_ids = model.generate(
input_ids,
max_new_tokens=1024,
do_sample = True,
top_p = 0.85,
temperature = 1.0,
repetition_penalty=1.,
eos_token_id=2,
bos_token_id=1,
pad_token_id=0)
output = tokenizer.batch_decode(generate_ids)[0]
print(output)
```
## 微调示例 Finetune Example
Refer to [ziya_finetune](https://github.com/IDEA-CCNL/Fengshenbang-LM/tree/main/fengshen/examples/ziya_llama)
## 推理量化示例 Inference & Quantization Example
Refer to [ziya_inference](https://github.com/IDEA-CCNL/Fengshenbang-LM/tree/main/fengshen/examples/ziya_inference)
## 引用 Citation
如果您在您的工作中使用了我们的模型，可以引用我们的[论文](https://arxiv.org/abs/2210.08590)：
If you are using the resource for your work, please cite the our [paper](https://arxiv.org/abs/2210.08590):
```text
@article{fengshenbang,
author = {Jiaxing Zhang and Ruyi Gan and Junjie Wang and Yuxiang Zhang and Lin Zhang and Ping Yang and Xinyu Gao and Ziwei Wu and Xiaoqun Dong and Junqing He and Jianheng Zhuo and Qi Yang and Yongfeng Huang and Xiayu Li and Yanghan Wu and Junyu Lu and Xinyu Zhu and Weifeng Chen and Ting Han and Kunhao Pan and Rui Wang and Hao Wang and Xiaojun Wu and Zhongshen Zeng and Chongpei Chen},
title = {Fengshenbang 1.0: Being the Foundation of Chinese Cognitive Intelligence},
journal = {CoRR},
volume = {abs/2209.02970},
year = {2022}
}
```
You can also cite our [website](https://github.com/IDEA-CCNL/Fengshenbang-LM/):
欢迎引用我们的[网站](https://github.com/IDEA-CCNL/Fengshenbang-LM/):
```text
@misc{Fengshenbang-LM,
title={Fengshenbang-LM},
author={IDEA-CCNL},
year={2021},
howpublished={\url{https://github.com/IDEA-CCNL/Fengshenbang-LM}},
}
```
| 12,780 | [
[
-0.040740966796875,
-0.042449951171875,
0.005035400390625,
0.035980224609375,
-0.0511474609375,
-0.00048089027404785156,
-0.0171966552734375,
-0.041717529296875,
0.0224609375,
0.0155487060546875,
-0.05743408203125,
-0.0369873046875,
-0.0419921875,
0.020782470703125,
-0.00484466552734375,
0.07855224609375,
-0.019927978515625,
-0.01198577880859375,
0.0238800048828125,
0.00035309791564941406,
-0.039886474609375,
-0.041900634765625,
-0.03997802734375,
-0.02947998046875,
0.0297088623046875,
0.0298919677734375,
0.03533935546875,
0.04632568359375,
0.044952392578125,
0.0238189697265625,
-0.0136260986328125,
0.0214996337890625,
-0.03802490234375,
-0.03045654296875,
0.0223388671875,
-0.040802001953125,
-0.043121337890625,
-0.0045623779296875,
0.043365478515625,
0.01169586181640625,
-0.0235137939453125,
0.0216522216796875,
0.0222625732421875,
0.047454833984375,
-0.0240478515625,
0.0196990966796875,
-0.023529052734375,
0.01434326171875,
-0.0201873779296875,
-0.015228271484375,
-0.0012617111206054688,
-0.029388427734375,
-0.0168304443359375,
-0.054962158203125,
-0.00760650634765625,
-0.003147125244140625,
0.100830078125,
0.0229034423828125,
-0.0207672119140625,
-0.007965087890625,
-0.025787353515625,
0.07415771484375,
-0.0714111328125,
-0.0019931793212890625,
0.025390625,
0.0220184326171875,
-0.0175018310546875,
-0.046234130859375,
-0.06317138671875,
-0.02197265625,
-0.0167083740234375,
0.016876220703125,
0.007701873779296875,
-0.01132965087890625,
0.01554107666015625,
0.028167724609375,
-0.01287841796875,
0.0227508544921875,
-0.026397705078125,
-0.012786865234375,
0.060760498046875,
0.015899658203125,
0.0245513916015625,
-0.0292510986328125,
-0.046295166015625,
-0.0303802490234375,
-0.049652099609375,
0.032806396484375,
0.019287109375,
0.01800537109375,
-0.0411376953125,
0.032806396484375,
-0.0286712646484375,
0.0305328369140625,
0.0196685791015625,
-0.0535888671875,
0.036895751953125,
-0.041473388671875,
-0.0168914794921875,
-0.0083160400390625,
0.07171630859375,
0.0291748046875,
0.00908660888671875,
0.007266998291015625,
-0.0190582275390625,
-0.004512786865234375,
-0.004199981689453125,
-0.0654296875,
0.01222991943359375,
0.0171051025390625,
-0.047454833984375,
-0.041900634765625,
-0.004268646240234375,
-0.048126220703125,
-0.0027866363525390625,
-0.0197601318359375,
0.028900146484375,
-0.0404052734375,
-0.03607177734375,
0.0163421630859375,
0.006683349609375,
0.03375244140625,
0.039215087890625,
-0.059814453125,
0.01485443115234375,
0.039337158203125,
0.061798095703125,
-0.01221466064453125,
-0.045562744140625,
-0.0105133056640625,
0.0075531005859375,
-0.0178680419921875,
0.04986572265625,
-0.00315093994140625,
-0.026885986328125,
-0.0231475830078125,
0.01136016845703125,
-0.01171112060546875,
-0.037261962890625,
0.029266357421875,
-0.0167388916015625,
-0.0012693405151367188,
-0.026214599609375,
-0.017547607421875,
-0.0275115966796875,
0.0214996337890625,
-0.034576416015625,
0.07965087890625,
0.0058441162109375,
-0.06072998046875,
0.01198577880859375,
-0.047637939453125,
-0.026336669921875,
-0.009613037109375,
-0.00125885009765625,
-0.0305633544921875,
-0.025848388671875,
0.0272064208984375,
0.04827880859375,
-0.0308074951171875,
0.01554107666015625,
0.00018537044525146484,
-0.045196533203125,
0.02337646484375,
-0.0160675048828125,
0.069580078125,
0.0124664306640625,
-0.039703369140625,
0.01514434814453125,
-0.0589599609375,
-0.00637054443359375,
0.041839599609375,
-0.049285888671875,
0.0018482208251953125,
-0.00675201416015625,
-0.004322052001953125,
0.0282135009765625,
0.037506103515625,
-0.02655029296875,
0.03167724609375,
-0.036163330078125,
0.042877197265625,
0.07843017578125,
-0.0028591156005859375,
0.01023101806640625,
-0.0426025390625,
0.033294677734375,
0.014739990234375,
0.0234375,
-0.0160675048828125,
-0.03521728515625,
-0.07379150390625,
-0.008575439453125,
0.01140594482421875,
0.05487060546875,
-0.047821044921875,
0.057281494140625,
-0.00981903076171875,
-0.053955078125,
-0.036651611328125,
0.0262451171875,
0.036407470703125,
0.034698486328125,
0.038116455078125,
-0.0235443115234375,
-0.04217529296875,
-0.0562744140625,
0.01739501953125,
-0.0205535888671875,
-0.006397247314453125,
0.01558685302734375,
0.05859375,
-0.0191497802734375,
0.036956787109375,
-0.031402587890625,
-0.035430908203125,
-0.0323486328125,
-0.003971099853515625,
0.0307769775390625,
0.041015625,
0.0443115234375,
-0.04376220703125,
-0.024627685546875,
0.0030651092529296875,
-0.06622314453125,
0.0242462158203125,
-0.017852783203125,
-0.0247344970703125,
0.0127105712890625,
0.00860595703125,
-0.044525146484375,
0.030426025390625,
0.031768798828125,
-0.0083465576171875,
0.052947998046875,
-0.003993988037109375,
-0.004032135009765625,
-0.08123779296875,
0.0029010772705078125,
-0.0019063949584960938,
0.0188751220703125,
-0.02911376953125,
0.01457977294921875,
-0.0008249282836914062,
0.02984619140625,
-0.040740966796875,
0.028778076171875,
-0.02899169921875,
0.0196533203125,
-0.0187530517578125,
-0.0010442733764648438,
-0.01476287841796875,
0.07745361328125,
-0.002716064453125,
0.05682373046875,
0.0229034423828125,
-0.056976318359375,
0.0276336669921875,
0.03802490234375,
-0.036346435546875,
0.01544952392578125,
-0.05792236328125,
0.0150146484375,
0.0027179718017578125,
0.0316162109375,
-0.075439453125,
-0.01177215576171875,
0.045745849609375,
-0.02984619140625,
0.016937255859375,
0.0130157470703125,
-0.037139892578125,
-0.041778564453125,
-0.04443359375,
0.004947662353515625,
0.041534423828125,
-0.045013427734375,
0.0201873779296875,
0.01226043701171875,
0.006183624267578125,
-0.062744140625,
-0.05377197265625,
-0.007171630859375,
-0.019195556640625,
-0.046661376953125,
0.03594970703125,
-0.0164031982421875,
0.00836944580078125,
-0.0218353271484375,
-0.006011962890625,
0.00823211669921875,
0.0106964111328125,
0.014404296875,
0.027923583984375,
-0.01180267333984375,
-0.0080108642578125,
0.003223419189453125,
-0.0041656494140625,
-0.01222991943359375,
0.0103302001953125,
0.05377197265625,
-0.0201263427734375,
-0.0178375244140625,
-0.06488037109375,
0.00264739990234375,
0.01549530029296875,
-0.00801849365234375,
0.05194091796875,
0.06378173828125,
-0.01293182373046875,
0.00989532470703125,
-0.043487548828125,
-0.00440216064453125,
-0.04046630859375,
0.0226898193359375,
-0.0279083251953125,
-0.061309814453125,
0.03411865234375,
-0.00882720947265625,
0.016082763671875,
0.0452880859375,
0.040374755859375,
-0.0012178421020507812,
0.065185546875,
0.054229736328125,
-0.03277587890625,
0.043670654296875,
-0.0499267578125,
-0.00817108154296875,
-0.0677490234375,
-0.034332275390625,
-0.025238037109375,
-0.036590576171875,
-0.033782958984375,
-0.03411865234375,
0.0160675048828125,
0.0197296142578125,
-0.041778564453125,
0.0323486328125,
-0.05291748046875,
0.005832672119140625,
0.042327880859375,
0.027557373046875,
0.0160369873046875,
0.005008697509765625,
-0.00498199462890625,
0.006641387939453125,
-0.0423583984375,
-0.03826904296875,
0.07220458984375,
0.0313720703125,
0.048187255859375,
-0.003662109375,
0.061370849609375,
0.00042724609375,
0.027252197265625,
-0.056793212890625,
0.044952392578125,
0.01247406005859375,
-0.0230560302734375,
-0.0274658203125,
-0.017547607421875,
-0.06048583984375,
0.03118896484375,
-0.00632476806640625,
-0.0631103515625,
0.01348114013671875,
-0.007236480712890625,
-0.035003662109375,
0.0313720703125,
-0.039642333984375,
0.049652099609375,
-0.042510986328125,
-0.0241546630859375,
-0.0147247314453125,
-0.05291748046875,
0.044158935546875,
-0.0076141357421875,
0.01216888427734375,
-0.0154876708984375,
-0.010894775390625,
0.0751953125,
-0.0552978515625,
0.0731201171875,
-0.017852783203125,
-0.019927978515625,
0.041290283203125,
-0.01558685302734375,
0.05572509765625,
0.0016546249389648438,
-0.00533294677734375,
0.039306640625,
-0.00511932373046875,
-0.0197906494140625,
-0.021636962890625,
0.033477783203125,
-0.083984375,
-0.05194091796875,
-0.035186767578125,
-0.019256591796875,
0.007366180419921875,
0.01390838623046875,
0.02642822265625,
-0.017730712890625,
0.01459503173828125,
0.007785797119140625,
0.021270751953125,
-0.043060302734375,
0.03546142578125,
0.0360107421875,
-0.006801605224609375,
-0.035797119140625,
0.06976318359375,
0.01280975341796875,
0.003139495849609375,
0.03717041015625,
0.01061248779296875,
-0.0066986083984375,
-0.03643798828125,
-0.04327392578125,
0.03759765625,
-0.03680419921875,
-0.033050537109375,
-0.03741455078125,
-0.0247344970703125,
-0.0242767333984375,
-0.0017805099487304688,
-0.0240631103515625,
-0.01529693603515625,
-0.04010009765625,
-0.0178985595703125,
0.0418701171875,
0.040008544921875,
-0.0015287399291992188,
0.04852294921875,
-0.049072265625,
0.02838134765625,
0.00948333740234375,
0.020477294921875,
0.034027099609375,
-0.044921875,
-0.033203125,
0.0135955810546875,
-0.038330078125,
-0.05853271484375,
0.044677734375,
0.0013647079467773438,
0.03741455078125,
0.04083251953125,
-0.01180267333984375,
0.0723876953125,
-0.01837158203125,
0.06414794921875,
0.03253173828125,
-0.066162109375,
0.046478271484375,
-0.021942138671875,
0.00220489501953125,
0.0208587646484375,
0.03375244140625,
-0.034149169921875,
0.004482269287109375,
-0.03155517578125,
-0.060821533203125,
0.0809326171875,
0.0146942138671875,
-0.0009918212890625,
0.0213623046875,
0.030181884765625,
0.005405426025390625,
0.022491455078125,
-0.06072998046875,
-0.0299072265625,
-0.0313720703125,
0.0003407001495361328,
-0.001148223876953125,
-0.0173492431640625,
-0.0027637481689453125,
-0.051910400390625,
0.049041748046875,
-0.007366180419921875,
0.0443115234375,
0.01027679443359375,
0.0009083747863769531,
-0.012481689453125,
0.0079345703125,
0.042724609375,
0.03887939453125,
-0.020233154296875,
-0.0313720703125,
0.035064697265625,
-0.05267333984375,
0.023101806640625,
0.005126953125,
-0.0134429931640625,
-0.0172576904296875,
0.0255126953125,
0.0718994140625,
0.014556884765625,
-0.05084228515625,
0.033050537109375,
-0.000048160552978515625,
-0.014739990234375,
-0.03045654296875,
0.01103973388671875,
0.0124053955078125,
0.0197296142578125,
0.039337158203125,
-0.01410675048828125,
-0.0093841552734375,
-0.0296630859375,
-0.002117156982421875,
0.01454925537109375,
0.0079498291015625,
-0.025238037109375,
0.0452880859375,
0.005035400390625,
-0.002307891845703125,
0.0367431640625,
-0.0163421630859375,
-0.038726806640625,
0.07159423828125,
0.04254150390625,
0.03857421875,
-0.0249481201171875,
0.025146484375,
0.05615234375,
0.033172607421875,
-0.0247344970703125,
0.0269317626953125,
-0.019012451171875,
-0.049224853515625,
-0.03173828125,
-0.0657958984375,
-0.0264739990234375,
0.0235443115234375,
-0.037139892578125,
0.0272064208984375,
-0.0535888671875,
-0.005859375,
-0.01207733154296875,
0.0281219482421875,
-0.05242919921875,
0.01526641845703125,
0.002490997314453125,
0.06982421875,
-0.035614013671875,
0.072265625,
0.0440673828125,
-0.05914306640625,
-0.079345703125,
-0.00897979736328125,
-0.005886077880859375,
-0.08795166015625,
0.0565185546875,
0.0214691162109375,
-0.0006012916564941406,
-0.01058197021484375,
-0.05596923828125,
-0.09014892578125,
0.11138916015625,
0.01537322998046875,
-0.050628662109375,
-0.00989532470703125,
0.01105499267578125,
0.041534423828125,
-0.016082763671875,
0.017730712890625,
0.04156494140625,
0.045013427734375,
0.0097808837890625,
-0.08154296875,
0.03173828125,
-0.033355712890625,
0.00665283203125,
-0.009185791015625,
-0.09796142578125,
0.087646484375,
-0.0271148681640625,
-0.0170135498046875,
0.01654052734375,
0.0626220703125,
0.03717041015625,
0.0145111083984375,
0.03472900390625,
0.035552978515625,
0.052154541015625,
-0.005405426025390625,
0.0712890625,
-0.01534271240234375,
0.0184326171875,
0.046142578125,
-0.0040740966796875,
0.061187744140625,
0.0244140625,
-0.04254150390625,
0.046905517578125,
0.069091796875,
-0.018951416015625,
0.044342041015625,
0.010284423828125,
-0.00922393798828125,
-0.0176239013671875,
-0.0019083023071289062,
-0.049224853515625,
0.026885986328125,
0.031463623046875,
-0.0094451904296875,
-0.0006380081176757812,
0.0090484619140625,
0.0139923095703125,
-0.033111572265625,
-0.01476287841796875,
0.0546875,
0.0125579833984375,
-0.042999267578125,
0.07977294921875,
0.011138916015625,
0.09832763671875,
-0.05841064453125,
-0.005725860595703125,
-0.026275634765625,
0.01073455810546875,
-0.0244598388671875,
-0.041595458984375,
-0.006694793701171875,
0.0008783340454101562,
0.01332855224609375,
-0.000789642333984375,
0.05047607421875,
0.0008268356323242188,
-0.037322998046875,
0.030242919921875,
0.0172576904296875,
0.0205230712890625,
0.0025577545166015625,
-0.06756591796875,
0.0151519775390625,
0.0167999267578125,
-0.04705810546875,
0.044342041015625,
0.02069091796875,
0.0013265609741210938,
0.05816650390625,
0.05755615234375,
0.0028934478759765625,
-0.0072784423828125,
-0.0086212158203125,
0.08489990234375,
-0.06201171875,
-0.031951904296875,
-0.05877685546875,
0.041168212890625,
0.0040740966796875,
-0.035003662109375,
0.053680419921875,
0.01374053955078125,
0.0714111328125,
0.001575469970703125,
0.064453125,
-0.01062774658203125,
0.031951904296875,
-0.0258026123046875,
0.043243408203125,
-0.0552978515625,
0.01128387451171875,
-0.032440185546875,
-0.038818359375,
-0.0031070709228515625,
0.06658935546875,
-0.01157379150390625,
0.0132598876953125,
0.03533935546875,
0.0521240234375,
0.0164794921875,
-0.0019254684448242188,
0.0006871223449707031,
0.0201416015625,
0.015869140625,
0.075439453125,
0.060546875,
-0.05279541015625,
0.05780029296875,
-0.0570068359375,
-0.01422119140625,
-0.032318115234375,
-0.047821044921875,
-0.06317138671875,
-0.033660888671875,
-0.01123809814453125,
-0.01395416259765625,
-0.006641387939453125,
0.0706787109375,
0.037567138671875,
-0.05194091796875,
-0.040191650390625,
0.006427764892578125,
0.014739990234375,
-0.049163818359375,
-0.01332855224609375,
0.0465087890625,
-0.01110076904296875,
-0.05224609375,
0.01617431640625,
0.0014209747314453125,
0.01654052734375,
-0.020050048828125,
-0.0218048095703125,
-0.0289459228515625,
-0.0145111083984375,
0.049072265625,
0.0233001708984375,
-0.057647705078125,
-0.006450653076171875,
0.01088714599609375,
-0.0064849853515625,
0.0199432373046875,
0.020233154296875,
-0.0535888671875,
-0.00811004638671875,
0.0228424072265625,
0.02337646484375,
0.047210693359375,
-0.002162933349609375,
0.008758544921875,
-0.03271484375,
0.023681640625,
-0.015228271484375,
0.0303802490234375,
0.0162200927734375,
-0.03680419921875,
0.054229736328125,
0.0244293212890625,
-0.026397705078125,
-0.057525634765625,
-0.01172637939453125,
-0.08612060546875,
-0.0220947265625,
0.08355712890625,
-0.01212310791015625,
-0.035003662109375,
0.0221710205078125,
-0.045074462890625,
0.044921875,
-0.034423828125,
0.07257080078125,
0.048828125,
-0.00843048095703125,
-0.00911712646484375,
-0.040771484375,
0.03033447265625,
0.0303802490234375,
-0.06072998046875,
-0.0240936279296875,
0.01532745361328125,
0.0237884521484375,
0.019287109375,
0.05841064453125,
0.007678985595703125,
0.0180511474609375,
-0.01134490966796875,
0.0198974609375,
-0.0198974609375,
0.0012073516845703125,
-0.016693115234375,
-0.0027637481689453125,
-0.00884246826171875,
-0.022247314453125
]
] |
MBZUAI/LaMini-Cerebras-1.3B | 2023-04-28T13:07:55.000Z | [
"transformers",
"pytorch",
"gpt2",
"text-generation",
"en",
"arxiv:2304.14402",
"license:cc-by-nc-4.0",
"endpoints_compatible",
"has_space",
"text-generation-inference",
"region:us"
] | text-generation | MBZUAI | null | null | MBZUAI/LaMini-Cerebras-1.3B | 1 | 5,925 | transformers | 2023-04-16T13:17:24 | ---
license: cc-by-nc-4.0
language:
- en
pipeline_tag: text-generation
widget:
- text: >-
Below is an instruction that describes a task.
Write a response that appropriately completes the request.
### Instruction:
how can I become more healthy?
### Response:
example_title: example
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
<p align="center" width="100%">
<a><img src="https://raw.githubusercontent.com/mbzuai-nlp/lamini-lm/main/images/lamini.png" alt="Title" style="width: 100%; min-width: 300px; display: block; margin: auto;"></a>
</p>
# LaMini-Cerebras-1.3B
[]()
This model is one of our LaMini-LM model series in paper "[LaMini-LM: A Diverse Herd of Distilled Models from Large-Scale Instructions](https://github.com/mbzuai-nlp/lamini-lm)".
This model is a fine-tuned version of [cerebras/Cerebras-GPT-1.3B](https://huggingface.co/cerebras/Cerebras-GPT-1.3B) on [LaMini-instruction dataset](https://huggingface.co/datasets/MBZUAI/LaMini-instruction) that contains 2.58M samples for instruction fine-tuning. For more information about our dataset, please refer to our [project repository](https://github.com/mbzuai-nlp/lamini-lm/).
You can view other models of LaMini-LM series as follows. Models with ✩ are those with the best overall performance given their size/architecture, hence we recommend using them. More details can be seen in our paper.
<table>
<thead>
<tr>
<th>Base model</th>
<th colspan="4">LaMini-LM series (#parameters)</th>
</tr>
</thead>
<tbody>
<tr>
<td>T5</td>
<td><a href="https://huggingface.co/MBZUAI/lamini-t5-61m" target="_blank" rel="noopener noreferrer">LaMini-T5-61M</a></td>
<td><a href="https://huggingface.co/MBZUAI/lamini-t5-223m" target="_blank" rel="noopener noreferrer">LaMini-T5-223M</a></td>
<td><a href="https://huggingface.co/MBZUAI/lamini-t5-738m" target="_blank" rel="noopener noreferrer">LaMini-T5-738M</a></td>
<td></td>
</tr>
<tr>
<td>Flan-T5</td>
<td><a href="https://huggingface.co/MBZUAI/lamini-flan-t5-77m" target="_blank" rel="noopener noreferrer">LaMini-Flan-T5-77M</a>✩</td>
<td><a href="https://huggingface.co/MBZUAI/lamini-flan-t5-248m" target="_blank" rel="noopener noreferrer">LaMini-Flan-T5-248M</a>✩</td>
<td><a href="https://huggingface.co/MBZUAI/lamini-flan-t5-783m" target="_blank" rel="noopener noreferrer">LaMini-Flan-T5-783M</a>✩</td>
<td></td>
</tr>
<tr>
<td>Cerebras-GPT</td>
<td><a href="https://huggingface.co/MBZUAI/lamini-cerebras-111m" target="_blank" rel="noopener noreferrer">LaMini-Cerebras-111M</a></td>
<td><a href="https://huggingface.co/MBZUAI/lamini-cerebras-256m" target="_blank" rel="noopener noreferrer">LaMini-Cerebras-256M</a></td>
<td><a href="https://huggingface.co/MBZUAI/lamini-cerebras-590m" target="_blank" rel="noopener noreferrer">LaMini-Cerebras-590M</a></td>
<td><a href="https://huggingface.co/MBZUAI/lamini-cerebras-1.3b" target="_blank" rel="noopener noreferrer">LaMini-Cerebras-1.3B</a></td>
</tr>
<tr>
<td>GPT-2</td>
<td><a href="https://huggingface.co/MBZUAI/lamini-gpt-124m" target="_blank" rel="noopener noreferrer">LaMini-GPT-124M</a>✩</td>
<td><a href="https://huggingface.co/MBZUAI/lamini-gpt-774m" target="_blank" rel="noopener noreferrer">LaMini-GPT-774M</a>✩</td>
<td><a href="https://huggingface.co/MBZUAI/lamini-gpt-1.5b" target="_blank" rel="noopener noreferrer">LaMini-GPT-1.5B</a>✩</td>
<td></td>
</tr>
<tr>
<td>GPT-Neo</td>
<td><a href="https://huggingface.co/MBZUAI/lamini-neo-125m" target="_blank" rel="noopener noreferrer">LaMini-Neo-125M</a></td>
<td><a href="https://huggingface.co/MBZUAI/lamini-neo-1.3b" target="_blank" rel="noopener noreferrer">LaMini-Neo-1.3B</a></td>
<td></td>
<td></td>
</tr>
<tr>
<td>GPT-J</td>
<td colspan="4">coming soon</td>
</tr>
<tr>
<td>LLaMA</td>
<td colspan="4">coming soon</td>
</tr>
</tbody>
</table>
## Use
### Intended use
We recommend using the model to respond to human instructions written in natural language.
Since this decoder-only model is fine-tuned with wrapper text, we suggest using the same wrapper text to achieve the best performance.
See the example on the right or the code below.
We now show you how to load and use our model using HuggingFace `pipeline()`.
```python
# pip install -q transformers
from transformers import pipeline
checkpoint = "{model_name}"
model = pipeline('text-generation', model = checkpoint)
instruction = 'Please let me know your thoughts on the given place and why you think it deserves to be visited: \n"Barcelona, Spain"'
input_prompt = f"Below is an instruction that describes a task. Write a response that appropriately completes the request.\n\n### Instruction:\n{instruction}\n\n### Response:"
generated_text = model(input_prompt, max_length=512, do_sample=True)[0]['generated_text']
print("Response", generated_text)
```
## Training Procedure
<p align="center" width="100%">
<a><img src="https://raw.githubusercontent.com/mbzuai-nlp/lamini-lm/main/images/lamini-pipeline.drawio.png" alt="Title" style="width: 100%; min-width: 250px; display: block; margin: auto;"></a>
</p>
We initialize with [cerebras/Cerebras-GPT-1.3B](https://huggingface.co/cerebras/Cerebras-GPT-1.3B) and fine-tune it on our [LaMini-instruction dataset](https://huggingface.co/datasets/MBZUAI/LaMini-instruction). Its total number of parameters is 1.3B.
### Training Hyperparameters
## Evaluation
We conducted two sets of evaluations: automatic evaluation on downstream NLP tasks and human evaluation on user-oriented instructions. For more detail, please refer to our [paper]().
## Limitations
More information needed
# Citation
```bibtex
@article{lamini-lm,
author = {Minghao Wu and
Abdul Waheed and
Chiyu Zhang and
Muhammad Abdul-Mageed and
Alham Fikri Aji
},
title = {LaMini-LM: A Diverse Herd of Distilled Models from Large-Scale Instructions},
journal = {CoRR},
volume = {abs/2304.14402},
year = {2023},
url = {https://arxiv.org/abs/2304.14402},
eprinttype = {arXiv},
eprint = {2304.14402}
}
``` | 6,579 | [
[
-0.0452880859375,
-0.0543212890625,
0.0134429931640625,
0.020416259765625,
-0.019805908203125,
-0.031951904296875,
-0.012176513671875,
-0.0455322265625,
0.02734375,
0.0207977294921875,
-0.05926513671875,
-0.033416748046875,
-0.03948974609375,
0.0017404556274414062,
-0.002391815185546875,
0.0631103515625,
-0.016265869140625,
-0.00736236572265625,
0.0106658935546875,
-0.00966644287109375,
-0.0164642333984375,
-0.031951904296875,
-0.0654296875,
-0.03253173828125,
0.01812744140625,
0.0013332366943359375,
0.054107666015625,
0.06207275390625,
0.0241546630859375,
0.0289764404296875,
-0.019195556640625,
0.0222320556640625,
-0.00556182861328125,
-0.0146331787109375,
0.00748443603515625,
-0.025909423828125,
-0.0723876953125,
0.0009760856628417969,
0.052978515625,
0.021636962890625,
0.01763916015625,
0.0304107666015625,
0.0190887451171875,
0.05810546875,
-0.031280517578125,
0.0124359130859375,
-0.0026187896728515625,
0.007175445556640625,
-0.0185089111328125,
-0.00125885009765625,
-0.0160675048828125,
-0.0361328125,
-0.000278472900390625,
-0.0458984375,
-0.00992584228515625,
0.00777435302734375,
0.11322021484375,
0.00891876220703125,
-0.006259918212890625,
-0.00754547119140625,
-0.0265045166015625,
0.0687255859375,
-0.06085205078125,
0.012237548828125,
0.04083251953125,
-0.0108489990234375,
0.005016326904296875,
-0.031829833984375,
-0.053955078125,
-0.0014677047729492188,
-0.038604736328125,
0.0263671875,
-0.0236663818359375,
-0.0268402099609375,
0.0458984375,
0.00980377197265625,
-0.03436279296875,
-0.0008897781372070312,
-0.025115966796875,
-0.005931854248046875,
0.04925537109375,
0.01739501953125,
0.0499267578125,
-0.022186279296875,
-0.0265350341796875,
-0.0178985595703125,
-0.026702880859375,
0.0248870849609375,
0.02923583984375,
0.021942138671875,
-0.0577392578125,
0.0243682861328125,
-0.00337982177734375,
0.0670166015625,
0.021331787109375,
-0.021484375,
0.047149658203125,
-0.015869140625,
-0.0292816162109375,
-0.0187225341796875,
0.08184814453125,
0.04638671875,
0.01763916015625,
0.003704071044921875,
-0.0015649795532226562,
-0.0192413330078125,
-0.00145721435546875,
-0.07159423828125,
-0.004940032958984375,
0.021240234375,
-0.04412841796875,
-0.031463623046875,
0.0038967132568359375,
-0.0667724609375,
0.004238128662109375,
-0.031097412109375,
0.0174407958984375,
-0.041351318359375,
-0.02252197265625,
0.0183258056640625,
-0.00039577484130859375,
0.0241851806640625,
0.0206451416015625,
-0.060760498046875,
0.00847625732421875,
0.0242462158203125,
0.0506591796875,
0.005191802978515625,
-0.02142333984375,
-0.019622802734375,
0.0158843994140625,
0.00888824462890625,
0.05194091796875,
-0.0179290771484375,
-0.02886962890625,
-0.017333984375,
0.0264434814453125,
-0.035858154296875,
-0.016021728515625,
0.06732177734375,
-0.00540924072265625,
0.025299072265625,
-0.03619384765625,
-0.0289459228515625,
-0.001049041748046875,
0.01346588134765625,
-0.050048828125,
0.07452392578125,
0.00791168212890625,
-0.08685302734375,
-0.0011777877807617188,
-0.059295654296875,
-0.01263427734375,
-0.02313232421875,
0.016021728515625,
-0.050323486328125,
-0.0211334228515625,
0.0257110595703125,
0.03167724609375,
-0.0234832763671875,
-0.0279083251953125,
-0.0230712890625,
-0.0188140869140625,
0.038970947265625,
-0.01538848876953125,
0.07366943359375,
0.01148223876953125,
-0.04913330078125,
-0.01369476318359375,
-0.06768798828125,
0.0208892822265625,
0.0258331298828125,
-0.025909423828125,
-0.00823211669921875,
-0.0223236083984375,
0.0183563232421875,
0.037506103515625,
0.032989501953125,
-0.0293426513671875,
0.008270263671875,
-0.034454345703125,
0.0273895263671875,
0.06097412109375,
0.001979827880859375,
0.031585693359375,
-0.056793212890625,
0.0244903564453125,
-0.006542205810546875,
0.02117919921875,
0.0082244873046875,
-0.0215301513671875,
-0.06951904296875,
-0.015472412109375,
0.0234375,
0.04547119140625,
-0.0285491943359375,
0.048797607421875,
-0.002567291259765625,
-0.03533935546875,
-0.048004150390625,
0.0068359375,
0.050872802734375,
0.035400390625,
0.042266845703125,
-0.011077880859375,
-0.053009033203125,
-0.05975341796875,
-0.003936767578125,
-0.01491546630859375,
0.0035762786865234375,
0.04473876953125,
0.04754638671875,
-0.02325439453125,
0.035552978515625,
-0.04205322265625,
-0.01322174072265625,
-0.02392578125,
0.005939483642578125,
0.0184326171875,
0.0599365234375,
0.05108642578125,
-0.062042236328125,
-0.047088623046875,
0.0018405914306640625,
-0.0701904296875,
-0.01013946533203125,
-0.0187225341796875,
-0.0338134765625,
0.0163116455078125,
0.00970458984375,
-0.033111572265625,
0.04217529296875,
0.023834228515625,
-0.037841796875,
0.039703369140625,
-0.019134521484375,
0.0115203857421875,
-0.08978271484375,
0.036651611328125,
0.0333251953125,
0.00740814208984375,
-0.067138671875,
0.01155853271484375,
-0.00782012939453125,
0.029937744140625,
-0.04107666015625,
0.06451416015625,
-0.031219482421875,
0.01763916015625,
-0.01337432861328125,
0.0246429443359375,
0.021728515625,
0.044219970703125,
0.0189361572265625,
0.0389404296875,
0.030853271484375,
-0.03424072265625,
0.0201568603515625,
0.035858154296875,
-0.011383056640625,
0.051666259765625,
-0.05926513671875,
0.00685882568359375,
-0.004974365234375,
0.0133819580078125,
-0.035186767578125,
-0.017608642578125,
0.040374755859375,
-0.0309906005859375,
0.049896240234375,
-0.0086822509765625,
-0.0303802490234375,
-0.05047607421875,
-0.02081298828125,
0.0111846923828125,
0.03790283203125,
-0.0287933349609375,
0.037200927734375,
0.0182647705078125,
0.0222320556640625,
-0.0556640625,
-0.052978515625,
-0.0215301513671875,
-0.0361328125,
-0.05828857421875,
0.038482666015625,
-0.01282501220703125,
-0.00730133056640625,
-0.0201416015625,
-0.007556915283203125,
-0.015899658203125,
0.0088043212890625,
0.0284881591796875,
0.03594970703125,
-0.017059326171875,
-0.012908935546875,
-0.0183563232421875,
-0.00928497314453125,
0.0089111328125,
-0.0014820098876953125,
0.053497314453125,
-0.0310211181640625,
-0.004215240478515625,
-0.09814453125,
0.00376129150390625,
0.038482666015625,
-0.021331787109375,
0.064697265625,
0.07904052734375,
-0.021881103515625,
0.01372528076171875,
-0.04229736328125,
-0.007099151611328125,
-0.03826904296875,
-0.0163726806640625,
-0.037689208984375,
-0.0340576171875,
0.050262451171875,
0.0012483596801757812,
-0.0159454345703125,
0.04180908203125,
0.0264434814453125,
-0.020721435546875,
0.053070068359375,
0.0281524658203125,
-0.02783203125,
0.03314208984375,
-0.057159423828125,
0.005748748779296875,
-0.10028076171875,
-0.03912353515625,
-0.03594970703125,
-0.035858154296875,
-0.03521728515625,
-0.02313232421875,
0.01247406005859375,
0.03887939453125,
-0.046661376953125,
0.04345703125,
-0.050048828125,
0.01337432861328125,
0.0352783203125,
0.044891357421875,
-0.004787445068359375,
-0.0126953125,
-0.0287017822265625,
-0.0014171600341796875,
-0.026214599609375,
-0.049530029296875,
0.069580078125,
0.0297088623046875,
0.03387451171875,
0.00791168212890625,
0.060272216796875,
0.0024013519287109375,
0.00492095947265625,
-0.0305023193359375,
0.03387451171875,
-0.005260467529296875,
-0.0287017822265625,
-0.023040771484375,
-0.0264434814453125,
-0.07269287109375,
0.0063934326171875,
-0.03369140625,
-0.08453369140625,
0.016845703125,
0.0153350830078125,
-0.032440185546875,
0.035736083984375,
-0.036285400390625,
0.068603515625,
-0.024658203125,
-0.06707763671875,
0.0248870849609375,
-0.04669189453125,
0.01114654541015625,
0.0281219482421875,
0.0183563232421875,
-0.0009937286376953125,
0.0113677978515625,
0.051239013671875,
-0.047576904296875,
0.06829833984375,
-0.0207061767578125,
-0.006744384765625,
0.039337158203125,
-0.0133819580078125,
0.0435791015625,
-0.0004279613494873047,
-0.0252532958984375,
-0.0091400146484375,
-0.005855560302734375,
-0.03204345703125,
-0.03558349609375,
0.05804443359375,
-0.07025146484375,
-0.038238525390625,
-0.039764404296875,
-0.0273284912109375,
0.0147705078125,
0.01160430908203125,
0.02691650390625,
0.036163330078125,
0.0023040771484375,
0.00582122802734375,
0.053466796875,
-0.0156402587890625,
0.045074462890625,
0.013031005859375,
0.00445556640625,
-0.01708984375,
0.061859130859375,
-0.0038299560546875,
0.01018524169921875,
0.042083740234375,
0.018096923828125,
-0.034149169921875,
-0.021026611328125,
-0.044464111328125,
0.042510986328125,
-0.0233306884765625,
-0.016448974609375,
-0.042694091796875,
-0.0243377685546875,
-0.02734375,
-0.029144287109375,
-0.01271820068359375,
-0.029693603515625,
-0.048614501953125,
-0.00501251220703125,
0.035552978515625,
0.037200927734375,
-0.0179290771484375,
0.0237884521484375,
-0.039215087890625,
0.0146026611328125,
0.013031005859375,
0.00823211669921875,
0.00942230224609375,
-0.03533935546875,
-0.00814056396484375,
0.021575927734375,
-0.03814697265625,
-0.05059814453125,
0.049774169921875,
-0.01007080078125,
0.042877197265625,
0.0312347412109375,
0.0013380050659179688,
0.058563232421875,
-0.0221405029296875,
0.04412841796875,
0.0262603759765625,
-0.07177734375,
0.0491943359375,
-0.0307464599609375,
0.0328369140625,
0.03558349609375,
0.038360595703125,
-0.02703857421875,
-0.016357421875,
-0.047393798828125,
-0.055694580078125,
0.06158447265625,
0.0229949951171875,
0.0007638931274414062,
0.0048065185546875,
0.038360595703125,
-0.03155517578125,
-0.004268646240234375,
-0.07489013671875,
-0.045562744140625,
-0.031036376953125,
-0.006359100341796875,
0.0264739990234375,
-0.0023784637451171875,
-0.0098114013671875,
-0.035736083984375,
0.06390380859375,
-0.005031585693359375,
0.0435791015625,
0.0177001953125,
-0.009033203125,
-0.004680633544921875,
0.0233154296875,
0.0592041015625,
0.0350341796875,
-0.0260772705078125,
-0.0207672119140625,
0.0236663818359375,
-0.03375244140625,
0.0017719268798828125,
-0.005901336669921875,
-0.028228759765625,
-0.006137847900390625,
0.01727294921875,
0.07574462890625,
0.0172576904296875,
-0.01033782958984375,
0.036956787109375,
0.00797271728515625,
-0.015655517578125,
-0.0255126953125,
0.014801025390625,
0.0175323486328125,
0.027435302734375,
0.001972198486328125,
0.00946807861328125,
0.00146484375,
-0.041717529296875,
0.020050048828125,
0.027587890625,
-0.027557373046875,
-0.017578125,
0.064453125,
-0.00441741943359375,
-0.01258087158203125,
0.022735595703125,
-0.0163726806640625,
-0.05938720703125,
0.046417236328125,
0.05377197265625,
0.044647216796875,
-0.0223388671875,
0.0255126953125,
0.0706787109375,
-0.0030689239501953125,
-0.0092620849609375,
0.01181793212890625,
0.0016908645629882812,
-0.0445556640625,
0.004608154296875,
-0.0758056640625,
-0.0014619827270507812,
0.0191497802734375,
-0.07452392578125,
0.0259552001953125,
-0.0357666015625,
-0.03045654296875,
-0.003910064697265625,
0.0281829833984375,
-0.053253173828125,
0.048736572265625,
0.01148223876953125,
0.0584716796875,
-0.049957275390625,
0.07672119140625,
0.03753662109375,
-0.056121826171875,
-0.06884765625,
0.00792694091796875,
0.0021820068359375,
-0.0704345703125,
0.06121826171875,
0.0008358955383300781,
-0.00124359130859375,
-0.00656890869140625,
-0.023040771484375,
-0.053680419921875,
0.1005859375,
-0.0094757080078125,
-0.0196990966796875,
-0.0202789306640625,
0.0203857421875,
0.04949951171875,
-0.0322265625,
0.054351806640625,
0.037139892578125,
0.051666259765625,
0.00466156005859375,
-0.06475830078125,
0.04345703125,
-0.044952392578125,
0.00402069091796875,
0.0007290840148925781,
-0.1031494140625,
0.0777587890625,
0.00394439697265625,
0.0003631114959716797,
0.018524169921875,
0.037109375,
0.025238037109375,
0.0164642333984375,
0.0109710693359375,
0.057952880859375,
0.0416259765625,
-0.0186614990234375,
0.08319091796875,
-0.03155517578125,
0.04095458984375,
0.07537841796875,
0.0031108856201171875,
0.06982421875,
0.014556884765625,
-0.0196685791015625,
0.05621337890625,
0.03009033203125,
-0.026031494140625,
0.01537322998046875,
0.0211334228515625,
-0.0125274658203125,
-0.00942230224609375,
-0.0057525634765625,
-0.0404052734375,
0.0169219970703125,
0.027740478515625,
-0.038726806640625,
0.00601959228515625,
-0.0230712890625,
0.031097412109375,
0.00920867919921875,
-0.01837158203125,
0.039947509765625,
0.0129852294921875,
-0.033966064453125,
0.0654296875,
-0.000728607177734375,
0.051788330078125,
-0.0361328125,
0.0153045654296875,
-0.0131378173828125,
0.01023101806640625,
-0.023895263671875,
-0.048492431640625,
0.00792694091796875,
0.0081787109375,
-0.0083465576171875,
-0.0250396728515625,
0.035247802734375,
-0.015350341796875,
-0.047088623046875,
0.030487060546875,
0.017730712890625,
0.0096435546875,
0.0237579345703125,
-0.09283447265625,
0.0223846435546875,
0.02435302734375,
-0.033447265625,
0.0245361328125,
0.0155181884765625,
0.0178070068359375,
0.048736572265625,
0.036712646484375,
-0.00276947021484375,
0.010223388671875,
-0.0006003379821777344,
0.06549072265625,
-0.033935546875,
-0.0071868896484375,
-0.06866455078125,
0.05963134765625,
-0.0300445556640625,
-0.02337646484375,
0.0714111328125,
0.044281005859375,
0.0548095703125,
-0.00984954833984375,
0.05108642578125,
-0.01678466796875,
0.026153564453125,
-0.04608154296875,
0.0712890625,
-0.047821044921875,
0.0111846923828125,
-0.03485107421875,
-0.048919677734375,
-0.0155792236328125,
0.075439453125,
-0.0189971923828125,
0.016937255859375,
0.0506591796875,
0.056365966796875,
0.0015153884887695312,
-0.004383087158203125,
-0.007785797119140625,
0.018890380859375,
-0.00205230712890625,
0.0687255859375,
0.038421630859375,
-0.064697265625,
0.0123443603515625,
-0.043701171875,
-0.00702667236328125,
-0.0250701904296875,
-0.053253173828125,
-0.08184814453125,
-0.047332763671875,
-0.03839111328125,
-0.041107177734375,
-0.00439453125,
0.0714111328125,
0.04388427734375,
-0.0618896484375,
-0.0258636474609375,
0.00556182861328125,
0.0011606216430664062,
-0.008270263671875,
-0.0196685791015625,
0.0582275390625,
0.0006146430969238281,
-0.0780029296875,
0.006137847900390625,
-0.00771331787109375,
0.0406494140625,
0.01491546630859375,
-0.021728515625,
-0.03472900390625,
0.00951385498046875,
0.0154876708984375,
0.041595458984375,
-0.044097900390625,
-0.023345947265625,
-0.0036449432373046875,
-0.0186004638671875,
0.01654052734375,
0.0203399658203125,
-0.033233642578125,
0.0088653564453125,
0.03753662109375,
0.01371002197265625,
0.05657958984375,
0.01654052734375,
0.0220794677734375,
-0.037139892578125,
0.0086822509765625,
-0.00766754150390625,
0.03363037109375,
0.00899505615234375,
-0.03155517578125,
0.043182373046875,
0.0159759521484375,
-0.03607177734375,
-0.05596923828125,
-0.007843017578125,
-0.09320068359375,
-0.0030460357666015625,
0.083740234375,
-0.0245361328125,
-0.0380859375,
0.023712158203125,
-0.02264404296875,
0.03759765625,
-0.034515380859375,
0.040283203125,
0.048431396484375,
-0.0271148681640625,
-0.01165008544921875,
-0.046417236328125,
0.049102783203125,
0.0172271728515625,
-0.06085205078125,
-0.0197601318359375,
0.0145721435546875,
0.021331787109375,
0.032928466796875,
0.034027099609375,
-0.00791168212890625,
0.0106201171875,
-0.01027679443359375,
0.0019178390502929688,
-0.006496429443359375,
0.0012845993041992188,
-0.00843048095703125,
0.00046181678771972656,
-0.021240234375,
-0.00782012939453125
]
] |
stabilityai/stablelm-base-alpha-7b-v2 | 2023-09-11T20:48:53.000Z | [
"transformers",
"safetensors",
"stablelm_alpha",
"text-generation",
"causal-lm",
"custom_code",
"en",
"dataset:tiiuae/falcon-refinedweb",
"dataset:togethercomputer/RedPajama-Data-1T",
"dataset:CarperAI/pilev2-dev",
"dataset:bigcode/starcoderdata",
"dataset:JeanKaddour/minipile",
"arxiv:2002.05202",
"arxiv:2104.09864",
"arxiv:2101.00027",
"arxiv:2305.06161",
"arxiv:1910.02054",
"license:cc-by-sa-4.0",
"has_space",
"region:us"
] | text-generation | stabilityai | null | null | stabilityai/stablelm-base-alpha-7b-v2 | 39 | 5,925 | transformers | 2023-08-04T04:38:56 | ---
datasets:
- tiiuae/falcon-refinedweb
- togethercomputer/RedPajama-Data-1T
- CarperAI/pilev2-dev
- bigcode/starcoderdata
- JeanKaddour/minipile
language:
- en
tags:
- causal-lm
license: cc-by-sa-4.0
---
# `StableLM-Base-Alpha-7B-v2`
## Model Description
`StableLM-Base-Alpha-7B-v2` is a 7 billion parameter decoder-only language model pre-trained on diverse English datasets. This model is the successor to the first [`StableLM-Base-Alpha-7B`](https://huggingface.co/stabilityai/stablelm-base-alpha-7b) model, addressing previous shortcomings through the use of improved data sources and mixture ratios.
## Usage
Get started generating text with `StableLM-Base-Alpha-7B-v2` by using the following code snippet:
```python
from transformers import AutoModelForCausalLM, AutoTokenizer
tokenizer = AutoTokenizer.from_pretrained("stabilityai/stablelm-base-alpha-7b-v2")
model = AutoModelForCausalLM.from_pretrained(
"stabilityai/stablelm-base-alpha-7b-v2",
trust_remote_code=True,
torch_dtype="auto",
)
model.cuda()
inputs = tokenizer("The weather is always wonderful", return_tensors="pt").to("cuda")
tokens = model.generate(
**inputs,
max_new_tokens=64,
temperature=0.75,
top_p=0.95,
do_sample=True,
)
print(tokenizer.decode(tokens[0], skip_special_tokens=True))
```
## Model Details
* **Developed by**: [Stability AI](https://stability.ai/)
* **Model type**: `StableLM-Base-Alpha-v2` models are auto-regressive language models based on the transformer decoder architecture.
* **Language(s)**: English
* **Library**: [GPT-NeoX](https://github.com/EleutherAI/gpt-neox)
* **License**: Model checkpoints are licensed under the Creative Commons license ([CC BY-SA-4.0](https://creativecommons.org/licenses/by-sa/4.0/)). Under this license, you must give [credit](https://creativecommons.org/licenses/by/4.0/#) to Stability AI, provide a link to the license, and [indicate if changes were made](https://creativecommons.org/licenses/by/4.0/#). You may do so in any reasonable manner, but not in any way that suggests the Stability AI endorses you or your use.
* **Contact**: For questions and comments about the model, please email `lm@stability.ai`
### Model Architecture
| Parameters | Hidden Size | Layers | Heads | Sequence Length |
|----------------|-------------|--------|-------|-----------------|
| 6,890,209,280 | 4096 | 32 | 32 | 4096 |
The model is a decoder-only transformer similar to the `StableLM-Base-Alpha` (v1) with the following configurations:
* **Activation**: SwiGLU ([Shazeer, 2020](https://arxiv.org/abs/2002.05202))
* **Decoder Layer**: Parallel Attention and MLP residuals with a single input LayerNorm ([Wang & Komatsuzaki, 2021](https://github.com/kingoflolz/mesh-transformer-jax/tree/master))
* **Position Embeddings**: Rotary Position Embeddings ([Su et al., 2021](https://arxiv.org/abs/2104.09864))
* **Bias**: LayerNorm bias terms only
## Training
`StableLM-Base-Alpha-7B-v2` is pre-trained using a multi-stage context length extension schedule following similar work ([Nijkamp et al. 2023](https://blog.salesforceairesearch.com/xgen/)); first pre-training at a context length of 2048 for 1 trillion tokens, then fine-tuning at a context length of 4096 for another 100B tokens.
### Training Dataset
The first pre-training stage relies on 1 trillion tokens sourced from a mix of the public Falcon RefinedWeb extract ([Penedo et al., 2023](https://huggingface.co/datasets/tiiuae/falcon-refinedweb)), RedPajama-Data ([Together Computer 2023](https://github.com/togethercomputer/RedPajama-Data), The Pile ([Gao et al., 2020](https://arxiv.org/abs/2101.00027)), and internal datasets with web text sampled at a rate of 71%.
In the second stage, we include the StarCoder ([Li et al., 2023](https://arxiv.org/abs/2305.06161)) dataset and down sample web text to 55% while increasing sampling proportions of naturally long text examples in the aforementioned sources.
### Training Procedure
The model is pre-trained on the dataset mixes mentioned above in mixed-precision (FP16), optimized with AdamW, and trained using the NeoX tokenizer with a vocabulary size of 50,257. We outline the complete hyperparameters choices in the project's [GitHub repository - config](https://github.com/Stability-AI/StableLM/blob/main/configs/stablelm-base-alpha-7b-v2.yaml).
### Training Infrastructure
* **Hardware**: `StableLM-Base-Alpha-7B-v2` was trained on the Stability AI cluster - occupying 384 NVIDIA A100 40GB GPUs across AWS P4d instances. Training took approximately 16.33 days to complete across both stages.
* **Software**: We use a fork of gpt-neox ([EleutherAI, 2021](https://github.com/EleutherAI/gpt-neox)) and train under 2D parallelism (Data and Tensor Parallel) with ZeRO-1 ([Rajbhandari et al., 2019](https://arxiv.org/abs/1910.02054v3)) and rely on flash-attention as well as rotary embedding kernels from FlashAttention-2 ([Dao et al., 2023](https://tridao.me/publications/flash2/flash2.pdf))
## Use and Limitations
### Intended Use
These models are intended to be used by all individuals as foundational models for application-specific fine-tuning without strict limitations on commercial use.
### Limitations and bias
The pre-training dataset may have contained offensive or inappropriate content even after applying data cleansing filters which can be reflected in the model-generated text. We recommend that users exercise caution when using these models in production systems. Do not use the models for any applications that may cause harm or distress to individuals or groups.
### How to cite
```bibtex
@misc{StableLMAlphaV2Models,
url={[https://huggingface.co/stabilityai/stablelm-base-alpha-7b-v2](https://huggingface.co/stabilityai/stablelm-base-alpha-7b-v2)},
title={StableLM Alpha v2 Models},
author={Tow, Jonathan}
}
```
| 5,852 | [
[
-0.029144287109375,
-0.053924560546875,
0.01201629638671875,
0.01540374755859375,
-0.0274658203125,
-0.018310546875,
-0.0157012939453125,
-0.04571533203125,
0.0013141632080078125,
0.0193634033203125,
-0.036865234375,
-0.036956787109375,
-0.04949951171875,
0.003810882568359375,
-0.027130126953125,
0.0799560546875,
0.007183074951171875,
-0.00771331787109375,
-0.0062103271484375,
-0.0150909423828125,
-0.01904296875,
-0.036773681640625,
-0.050811767578125,
-0.01343536376953125,
0.0164794921875,
0.0016565322875976562,
0.0662841796875,
0.0714111328125,
0.039337158203125,
0.0243072509765625,
-0.0180206298828125,
-0.002292633056640625,
-0.048736572265625,
-0.002593994140625,
0.0109405517578125,
-0.0164337158203125,
-0.0416259765625,
-0.00555419921875,
0.050537109375,
0.0223388671875,
-0.01190948486328125,
0.0220794677734375,
-0.0022907257080078125,
0.0308074951171875,
-0.040252685546875,
0.015960693359375,
-0.049835205078125,
-0.0235595703125,
-0.00652313232421875,
0.024871826171875,
-0.029876708984375,
0.00011980533599853516,
0.0009722709655761719,
-0.054443359375,
0.01490020751953125,
0.00272369384765625,
0.10107421875,
0.027374267578125,
-0.0216217041015625,
0.00475311279296875,
-0.044189453125,
0.0628662109375,
-0.07098388671875,
0.037872314453125,
0.03009033203125,
0.01476287841796875,
0.0037403106689453125,
-0.0714111328125,
-0.035980224609375,
-0.0117645263671875,
-0.002010345458984375,
0.01465606689453125,
-0.00991058349609375,
-0.006092071533203125,
0.0253143310546875,
0.00044846534729003906,
-0.048736572265625,
0.004238128662109375,
-0.033599853515625,
-0.02313232421875,
0.042633056640625,
0.0146026611328125,
0.0013427734375,
-0.00800323486328125,
-0.03546142578125,
-0.02154541015625,
-0.0406494140625,
0.010589599609375,
0.018218994140625,
0.0211944580078125,
-0.03326416015625,
0.0197906494140625,
0.00006628036499023438,
0.03515625,
0.01340484619140625,
-0.0245361328125,
0.039520263671875,
-0.0361328125,
-0.014556884765625,
0.00273895263671875,
0.0823974609375,
0.0258941650390625,
-0.0113677978515625,
-0.004730224609375,
-0.0240020751953125,
0.0221710205078125,
0.0228729248046875,
-0.0703125,
0.0009379386901855469,
0.0196685791015625,
-0.037841796875,
-0.02618408203125,
0.00949859619140625,
-0.049560546875,
-0.0106964111328125,
-0.006011962890625,
0.0362548828125,
-0.03662109375,
-0.032318115234375,
0.012420654296875,
-0.006378173828125,
0.02618408203125,
0.006420135498046875,
-0.05877685546875,
0.02655029296875,
0.03778076171875,
0.05511474609375,
-0.003391265869140625,
-0.0374755859375,
-0.01702880859375,
0.0038967132568359375,
-0.011138916015625,
0.028839111328125,
-0.023162841796875,
-0.02099609375,
-0.010498046875,
0.0164947509765625,
0.0008230209350585938,
-0.0184173583984375,
0.041229248046875,
-0.03753662109375,
0.021087646484375,
-0.0005693435668945312,
-0.0217437744140625,
-0.012908935546875,
0.0298004150390625,
-0.041778564453125,
0.086181640625,
0.0253448486328125,
-0.065185546875,
0.01334381103515625,
-0.033905029296875,
-0.0260162353515625,
-0.01085662841796875,
-0.0037097930908203125,
-0.06005859375,
-0.018646240234375,
0.01364898681640625,
0.010986328125,
-0.03228759765625,
0.0255279541015625,
-0.0188446044921875,
-0.02978515625,
-0.003711700439453125,
-0.0295257568359375,
0.06451416015625,
0.0121917724609375,
-0.060760498046875,
0.020355224609375,
-0.057037353515625,
-0.00647735595703125,
0.027801513671875,
-0.01316070556640625,
-0.006504058837890625,
-0.014312744140625,
0.01239013671875,
0.0280609130859375,
0.02911376953125,
-0.0229034423828125,
0.01284027099609375,
-0.034820556640625,
0.03216552734375,
0.042266845703125,
-0.0164642333984375,
0.0290069580078125,
-0.018218994140625,
0.0430908203125,
0.0104522705078125,
0.033599853515625,
-0.01131439208984375,
-0.048004150390625,
-0.06134033203125,
-0.0046539306640625,
0.0235595703125,
0.04156494140625,
-0.0439453125,
0.037261962890625,
-0.0275726318359375,
-0.047515869140625,
-0.044677734375,
0.004840850830078125,
0.05096435546875,
0.04974365234375,
0.046417236328125,
-0.002391815185546875,
-0.056243896484375,
-0.066162109375,
0.0199432373046875,
-0.022796630859375,
0.02227783203125,
0.004482269287109375,
0.03924560546875,
-0.04656982421875,
0.056243896484375,
-0.01245880126953125,
-0.00795745849609375,
-0.0139007568359375,
0.003509521484375,
0.0222930908203125,
0.043426513671875,
0.0572509765625,
-0.051116943359375,
-0.02880859375,
-0.00827789306640625,
-0.0601806640625,
0.009613037109375,
-0.0015382766723632812,
-0.0177001953125,
0.036346435546875,
0.0249481201171875,
-0.06640625,
0.02996826171875,
0.044097900390625,
-0.038909912109375,
0.0452880859375,
-0.009674072265625,
-0.00933837890625,
-0.09344482421875,
0.0244293212890625,
0.0029087066650390625,
-0.01617431640625,
-0.042816162109375,
0.002079010009765625,
0.01415252685546875,
-0.002765655517578125,
-0.054595947265625,
0.04779052734375,
-0.04193115234375,
0.004207611083984375,
-0.0236053466796875,
0.0042877197265625,
0.003055572509765625,
0.032684326171875,
0.0019016265869140625,
0.05413818359375,
0.053466796875,
-0.04083251953125,
-0.0014085769653320312,
0.021820068359375,
-0.004657745361328125,
-0.0105133056640625,
-0.06341552734375,
0.0096282958984375,
-0.0021610260009765625,
0.01256561279296875,
-0.053314208984375,
0.007137298583984375,
0.0263824462890625,
-0.041961669921875,
0.02459716796875,
-0.0219573974609375,
-0.022125244140625,
-0.038543701171875,
-0.021331787109375,
0.0310211181640625,
0.061065673828125,
-0.0222930908203125,
0.04443359375,
0.031585693359375,
0.006458282470703125,
-0.0770263671875,
-0.0482177734375,
-0.00925445556640625,
-0.0102691650390625,
-0.05511474609375,
0.0286102294921875,
-0.006649017333984375,
-0.01081085205078125,
0.0094451904296875,
-0.0093536376953125,
0.0103912353515625,
0.0175628662109375,
0.0335693359375,
0.0306243896484375,
-0.0211334228515625,
-0.0199432373046875,
0.00652313232421875,
-0.0242462158203125,
0.0068511962890625,
-0.0221405029296875,
0.061859130859375,
-0.042816162109375,
0.005268096923828125,
-0.04364013671875,
0.008697509765625,
0.0631103515625,
-0.01373291015625,
0.08001708984375,
0.07080078125,
-0.027130126953125,
0.00823211669921875,
-0.024505615234375,
-0.0274658203125,
-0.03851318359375,
0.035980224609375,
-0.0112762451171875,
-0.05572509765625,
0.0599365234375,
0.032012939453125,
0.0132598876953125,
0.06463623046875,
0.0457763671875,
0.01654052734375,
0.09075927734375,
0.041534423828125,
-0.009368896484375,
0.02740478515625,
-0.0543212890625,
-0.00247955322265625,
-0.055816650390625,
-0.0248260498046875,
-0.035919189453125,
-0.01715087890625,
-0.03167724609375,
-0.0222930908203125,
0.006412506103515625,
0.01360321044921875,
-0.04901123046875,
0.0273895263671875,
-0.0396728515625,
0.0109405517578125,
0.0355224609375,
-0.00815582275390625,
-0.01013946533203125,
-0.0060272216796875,
-0.01172637939453125,
0.00957489013671875,
-0.052703857421875,
-0.033905029296875,
0.07476806640625,
0.040191650390625,
0.054473876953125,
0.005092620849609375,
0.037384033203125,
-0.0008540153503417969,
0.0198516845703125,
-0.040679931640625,
0.040313720703125,
-0.00952911376953125,
-0.047698974609375,
-0.016876220703125,
-0.032989501953125,
-0.07977294921875,
0.0133209228515625,
-0.0305938720703125,
-0.036468505859375,
0.0198211669921875,
0.0168304443359375,
-0.022247314453125,
0.010284423828125,
-0.044342041015625,
0.0751953125,
-0.034698486328125,
-0.033966064453125,
0.0006322860717773438,
-0.07373046875,
0.0227203369140625,
0.00853729248046875,
0.01540374755859375,
-0.00295257568359375,
-0.005680084228515625,
0.06396484375,
-0.039337158203125,
0.056427001953125,
-0.0244140625,
0.013214111328125,
0.024383544921875,
-0.010955810546875,
0.044586181640625,
0.0219573974609375,
-0.020416259765625,
0.036285400390625,
-0.00018537044525146484,
-0.030670166015625,
-0.0272979736328125,
0.051849365234375,
-0.09686279296875,
-0.039093017578125,
-0.04327392578125,
-0.03228759765625,
0.004756927490234375,
0.046478271484375,
0.032684326171875,
0.032440185546875,
0.0092315673828125,
0.035308837890625,
0.0313720703125,
0.002391815185546875,
0.04254150390625,
0.035308837890625,
-0.01995849609375,
-0.05084228515625,
0.054595947265625,
0.0030975341796875,
0.01003265380859375,
0.0015163421630859375,
0.01262664794921875,
-0.043304443359375,
-0.059722900390625,
-0.041473388671875,
0.0299530029296875,
-0.049407958984375,
-0.0323486328125,
-0.052001953125,
-0.0255889892578125,
-0.049407958984375,
-0.0018339157104492188,
-0.049346923828125,
-0.0182342529296875,
-0.039642333984375,
-0.009552001953125,
0.02789306640625,
0.03265380859375,
0.01500701904296875,
0.0295562744140625,
-0.06201171875,
0.0222320556640625,
-0.0025501251220703125,
0.0160064697265625,
-0.00726318359375,
-0.054412841796875,
-0.03228759765625,
0.0240325927734375,
-0.01381683349609375,
-0.047210693359375,
0.046112060546875,
0.0113677978515625,
0.055084228515625,
0.03448486328125,
0.01538848876953125,
0.050384521484375,
-0.02239990234375,
0.07373046875,
0.0172576904296875,
-0.060516357421875,
0.038238525390625,
-0.028350830078125,
0.033355712890625,
0.050384521484375,
0.0311737060546875,
-0.0102996826171875,
-0.0276336669921875,
-0.05828857421875,
-0.08648681640625,
0.06695556640625,
0.0237579345703125,
0.00499725341796875,
-0.007537841796875,
0.0447998046875,
-0.0003485679626464844,
0.00852203369140625,
-0.061676025390625,
-0.0401611328125,
-0.048004150390625,
-0.027740478515625,
-0.00734710693359375,
-0.01922607421875,
-0.01056671142578125,
-0.026824951171875,
0.06793212890625,
-0.00693511962890625,
0.03363037109375,
0.01328277587890625,
-0.0250091552734375,
-0.00577545166015625,
-0.01038360595703125,
0.057708740234375,
0.053131103515625,
-0.0306243896484375,
0.005741119384765625,
0.0031261444091796875,
-0.06500244140625,
0.01079559326171875,
0.026153564453125,
-0.037872314453125,
-0.01111602783203125,
0.0167083740234375,
0.09930419921875,
0.0073394775390625,
-0.040557861328125,
0.029632568359375,
-0.02679443359375,
-0.030029296875,
-0.022857666015625,
0.006015777587890625,
0.00415802001953125,
0.0081024169921875,
0.03167724609375,
0.00966644287109375,
-0.009979248046875,
-0.0272064208984375,
0.0168609619140625,
0.023895263671875,
-0.022857666015625,
-0.032867431640625,
0.06976318359375,
0.006168365478515625,
-0.0182952880859375,
0.05938720703125,
-0.01554107666015625,
-0.035980224609375,
0.053466796875,
0.07061767578125,
0.061065673828125,
-0.0079193115234375,
0.0076446533203125,
0.03704833984375,
0.035064697265625,
-0.025726318359375,
0.0174560546875,
0.0293426513671875,
-0.05511474609375,
-0.01464080810546875,
-0.055328369140625,
-0.0099334716796875,
0.0188446044921875,
-0.049896240234375,
0.035064697265625,
-0.054931640625,
-0.036376953125,
-0.023468017578125,
0.013519287109375,
-0.039306640625,
0.02099609375,
0.01641845703125,
0.06884765625,
-0.05877685546875,
0.0723876953125,
0.06878662109375,
-0.03759765625,
-0.07159423828125,
-0.0103759765625,
-0.01165008544921875,
-0.04705810546875,
0.0240020751953125,
0.0286407470703125,
-0.0090484619140625,
0.013824462890625,
-0.03594970703125,
-0.07867431640625,
0.10357666015625,
0.04296875,
-0.043304443359375,
-0.0009694099426269531,
-0.003704071044921875,
0.047119140625,
-0.0286865234375,
0.032135009765625,
0.0275421142578125,
0.03424072265625,
0.001087188720703125,
-0.055816650390625,
0.00954437255859375,
-0.048492431640625,
0.0010128021240234375,
0.0135650634765625,
-0.064208984375,
0.0711669921875,
-0.00847625732421875,
-0.0036792755126953125,
-0.0002351999282836914,
0.058563232421875,
0.03668212890625,
0.0187225341796875,
0.0406494140625,
0.063720703125,
0.042816162109375,
-0.00897216796875,
0.078857421875,
-0.056915283203125,
0.0457763671875,
0.067626953125,
0.0016660690307617188,
0.066162109375,
0.022247314453125,
-0.0015106201171875,
0.040618896484375,
0.04779052734375,
-0.0134429931640625,
0.037689208984375,
-0.0295257568359375,
0.0110015869140625,
-0.0159912109375,
0.0122222900390625,
-0.055145263671875,
0.01641845703125,
0.0256500244140625,
-0.0311431884765625,
0.0005602836608886719,
-0.0015192031860351562,
0.0155181884765625,
-0.024444580078125,
-0.016876220703125,
0.040679931640625,
0.00604248046875,
-0.042449951171875,
0.08135986328125,
0.00872039794921875,
0.053985595703125,
-0.05401611328125,
0.0236053466796875,
-0.01593017578125,
0.02099609375,
-0.0018701553344726562,
-0.045166015625,
0.017578125,
-0.00858306884765625,
-0.022216796875,
-0.006229400634765625,
0.05206298828125,
-0.0225830078125,
-0.04693603515625,
0.037109375,
0.00977325439453125,
0.0003898143768310547,
0.00899505615234375,
-0.0799560546875,
0.034820556640625,
-0.01059722900390625,
-0.0380859375,
0.0296630859375,
0.007274627685546875,
-0.007305145263671875,
0.045806884765625,
0.050628662109375,
-0.0130462646484375,
0.0160675048828125,
0.0053558349609375,
0.0704345703125,
-0.051483154296875,
-0.041839599609375,
-0.054473876953125,
0.057769775390625,
0.000598907470703125,
-0.037078857421875,
0.060577392578125,
0.04730224609375,
0.05035400390625,
0.0019083023071289062,
0.050811767578125,
-0.01200103759765625,
0.00588226318359375,
-0.0306854248046875,
0.06243896484375,
-0.04461669921875,
0.0180206298828125,
-0.0167236328125,
-0.071044921875,
-0.019622802734375,
0.049072265625,
-0.006015777587890625,
0.01526641845703125,
0.04400634765625,
0.0638427734375,
-0.0013980865478515625,
-0.02001953125,
0.0116119384765625,
0.041534423828125,
0.02783203125,
0.032257080078125,
0.04034423828125,
-0.06927490234375,
0.0567626953125,
-0.039337158203125,
-0.0029392242431640625,
-0.0259857177734375,
-0.052703857421875,
-0.06396484375,
-0.03564453125,
-0.02618408203125,
-0.056427001953125,
0.01419830322265625,
0.060577392578125,
0.05804443359375,
-0.05902099609375,
-0.00939178466796875,
-0.0140228271484375,
-0.00611114501953125,
-0.0209197998046875,
-0.01250457763671875,
0.03179931640625,
-0.0182952880859375,
-0.0418701171875,
0.027557373046875,
0.005420684814453125,
0.0182647705078125,
-0.0211334228515625,
-0.040740966796875,
-0.023895263671875,
-0.010406494140625,
0.0325927734375,
0.0275726318359375,
-0.0491943359375,
-0.01410675048828125,
0.0182952880859375,
-0.0167236328125,
0.01049041748046875,
0.02093505859375,
-0.052825927734375,
0.01435089111328125,
0.03729248046875,
0.04443359375,
0.048065185546875,
0.01336669921875,
0.02191162109375,
-0.033447265625,
0.041168212890625,
0.0183258056640625,
0.03253173828125,
0.0166778564453125,
-0.0277099609375,
0.0250244140625,
0.03594970703125,
-0.041717529296875,
-0.07122802734375,
-0.007427215576171875,
-0.0968017578125,
-0.0116424560546875,
0.10400390625,
-0.0144500732421875,
-0.041839599609375,
-0.0041046142578125,
-0.0149383544921875,
0.0251312255859375,
-0.03900146484375,
0.04901123046875,
0.0274810791015625,
0.005817413330078125,
-0.038818359375,
-0.01493072509765625,
0.0338134765625,
0.019927978515625,
-0.04620361328125,
-0.00977325439453125,
0.0299530029296875,
0.02520751953125,
0.0277862548828125,
0.0372314453125,
-0.0180511474609375,
0.028350830078125,
0.01123046875,
0.01381683349609375,
-0.0304412841796875,
-0.02972412109375,
-0.027496337890625,
0.0149993896484375,
-0.00423431396484375,
0.00328826904296875
]
] |
cmarkea/bloomz-7b1-mt-sft-chat | 2023-09-23T12:27:58.000Z | [
"transformers",
"pytorch",
"safetensors",
"bloom",
"text-generation",
"fr",
"en",
"dataset:ehartford/wizard_vicuna_70k_unfiltered",
"dataset:shahules786/orca-chat",
"dataset:timdettmers/openassistant-guanaco",
"dataset:laion/OIG",
"arxiv:2012.15613",
"arxiv:2001.09977",
"license:bigscience-bloom-rail-1.0",
"endpoints_compatible",
"text-generation-inference",
"region:us"
] | text-generation | cmarkea | null | null | cmarkea/bloomz-7b1-mt-sft-chat | 10 | 5,925 | transformers | 2023-09-11T16:59:44 | ---
license: bigscience-bloom-rail-1.0
datasets:
- ehartford/wizard_vicuna_70k_unfiltered
- shahules786/orca-chat
- timdettmers/openassistant-guanaco
- laion/OIG
language:
- fr
- en
library_name: transformers
pipeline_tag: text-generation
widget:
- text: </s>Bonjour, qui es-tu ?<s>
- text: </s>Hello, who are you?<s>
---
bloomz-7b1-mt-sft-chat
--------------------
We introduce the bloomz-7b1-mt-sft-chat model, which is a fine-tuning of a Large Language Model (LLM) [bigscience/bloomz-7b1-mt](https://huggingface.co/bigscience/bloomz-7b1-mt). This model is notable for being pre-trained for a chatbot context and undergoing a transposition from float16 to bfloat16. Therefore, this model serves as a solid starting point for fine-tuning towards other more specific tasks.
The model was trained equally on both French and English data, ensuring maximum efficiency for these two languages (and their interactions). Due to the transition from float16 to bfloat16, we do not guarantee the preservation of the original model's multilingual capabilities. However, fine-tuning can restore reasonable performance on other languages.
The objective is to pre-train all three models (Bloomz-{560m, 3b, 7b1-mt}-sft-chat) to ensure high-performing, energy-efficient, and fast "foundation" models for inference on "realistic" infrastructures suitable for a business with standard industrial capabilities.
Bloomz, through its license, enables free and flexible industrial use. Its tokenizer has been designed with true multi-lingual context in mind, with a significantly lower token generation per word compared to other LLM models. This capability not only leads to improved performance but also enhanced efficiency during inference by making fewer model calls when generating text with shorter contexts. Here is a table illustrating our points using French as an example, where we tokenized Marcel Proust's longest sentence (823 words):
```
Sans honneur que précaire, sans liberté que provisoire, [...], et de façon qu’à eux-mêmes il ne leur paraisse pas un vice.
```
| model | GPT 3.5 | Boris | Flan-T5 | LLaMA | Dolly | MPT | Falcon | Bloomz |
|:--------------:|:-------:|:-----:|:-------:|:-----:|:-----:|:---:|:------:|:------:|
| tokens per word | 2.3 | 2.3 | 2 | 1.9 | 1.9 | 1.9 | 1.8 | 1.4 |
For comparison, with a specialized French tokenizer like [CamemBERT](https://huggingface.co/camembert/camembert-base) or [DistilCamemBERT](cmarkea/distilcamembert-base), we have 1.5 tokens per word. In addition to its positive impact on inference time and resource consumption, there has already been [shown that there is a direct relationship](https://arxiv.org/abs/2012.15613) between the number of tokens per word required for modeling and the predictive performance of the model.
Dataset
-------
After analyzing a substantial set of modelings, we have observed that the most effective pre-training for zero-shot use cases is pre-training for chatbot contexts. This study was conducted internally, focusing specifically on the French context. As a result, we trained the model on a dataset comprising 0.9 billion tokens. This dataset consists of interactions between an individual and a third party. To balance the French and English data, we utilized the Google Translate API.
Training
--------
Here is the table summarizing the architecture used for training, along with the training time and average inference speed per token on the target architecture in tokens per second:
| model | Architecture | Training time (h) | Inference speed (tokens per second) |
|:----------------------:|:-------------:|:-----------------:|:-----------------------------------:|
| [bloomz-560m-sft-chat](https://huggingface.co/cmarkea/bloomz-560m-sft-chat) | 1 x A100 40GB | 41 | 29 |
| [bloomz-3b-sft-chat](https://huggingface.co/cmarkea/bloomz-3b-sft-chat) | 1 x A100 40GB | 140 | 13 |
| [bloomz-7b1-mt-sft-chat](https://huggingface.co/cmarkea/bloomz-7b1-mt-sft-chat) | 4 x A100 40GB | 268 | 8 |
| Hyperparameter | Value |
|:---------------------:|:----------:|
| label smoothing | 0.05 |
| optimizer | AdamW |
| betas | 0.9, 0.999 |
| learning rate | 5e-6 |
| anneal strategy | cos |
| div factor | 100 |
| final div factor | 0.1 |
| batch size | 2 |
| gradient accumulation | 200 |
| max length | 2048 |
Experimentations
----------------
Since the model is trained only on English and French corpora, the performance of the model cannot be guaranteed in other languages. This degradation in performance in other languages is also due to the change in the model's data type from float16 to bfloat16. The conversation example below illustrates this point:
| French | English | Spanish |
|:-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------:|:--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------:|:------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------:|
| **human:** C'est quoi la première guerre mondiale ?<br>**bot:** La Première Guerre mondiale a eu lieu entre 1914 et 1918. C'était une guerre entre les puissances européennes, principalement la France, l'Allemagne et la Grande-Bretagne.<br>**human:** Y a-t-il eu également d’autres pays impliqués ?<br>**bot:** Oui, il y avait également des pays comme l'Italie, les États-Unis, la Russie et la Turquie.<br>**human:** Comment cela s'est-il fini et quels ont été ses impacts ?<br>**bot:** La Première Guerre mondiale s'est terminée par un accord de paix appelé le Traité de Versailles. Il a mis fin à la guerre et a établi un nouvel ordre mondial.<br>**human:** Combien y a-t-il eu de mort ?<br>**bot:** Il y a eu environ 10 millions de morts, principalement des soldats.<br>**human:** Et combien de mort civile ?<br>**bot:** Il y a eu environ 5 millions de morts civils, principalement des civils.<br>**human:** Tu es certain de ce chiffre, ça me parait peu...<br>**bot:** Je suis sûr de ce chiffre. | **human:** What is the first world war?<br>**bot:** The first world war was a war that took place between 1914 and 1918.<br>**human:** Which countries participated in this war?<br>**bot:** he countries that participated in the first world war were the United Kingdom, France, Belgium, the Netherlands, Italy, Austria-Hungary, Serbia, Montenegro, Bulgaria, Greece, Romania, Albania, and the Ottoman Empire.<br>**human:** How did it end and what were its impacts?<br>**bot:** The first world war ended with the Treaty of Versailles, which was signed on 28 June 1919. The Treaty of Versailles was a peace treaty that ended the war and established the international system that we know today.<br>**human:** How many deaths?<br>**bot:** The first world war resulted in approximately 50 million deaths, with approximately 30 million of those deaths occurring in the European Theater.<br>**human:** That seems a bit much to me, are you sure of these numbers?<br>**bot:** Yes, I am. | **human:** ¿Qué es la Primera Guerra Mundial?<br>**bot:** La Primera Guerra Mundial fue un conflicto bélico que ocurrió entre 1914 y 1918.<br>**human:** ¿En qué países participamos?<br>**bot:** La Primera Guerra Mundial involucró a muchos países, incluidos Alemania, Austria-Hungría, Bélgica, Bulgaria, Checoslovaquia, Francia, Gran Bretaña, Italia, Japón, Luxemburgo, Noruega, los Países Bajos, Portugal, Serbia, España, Turquía y Estados Unidos.<br>**human:** ¿Cuántas muertes?<br>**bot:** La Primera Guerra Mundial dejó más de 10 millones de muertes.<br>**human:** ¿Y cuántas muertes de civiles?<br>**bot:** La Primera Guerra Mundial dejó más de 5 millones de muertes de civiles.<br>**human:** ¿Estás seguro? Eso no me parece mucho.<br>**bot:** No, no estoy seguro. |
Evaluating the performance of generative models can be challenging, and quantitative measures may not always provide the most relevant insights for this type of approach. In this regard, we conducted a qualitative performance study. It involves assessing the relevance of model responses to a pool of French questions using blind evaluation. The metric used for evaluation is the average of two criteria: response accuracy and completeness, similar to what the [SSA metric](https://arxiv.org/abs/2001.09977) aims to capture. Please note that this measure lacks rigorous scientific validity due to the limited number of questions in the test dataset and the fact that it involves only a few individuals with similar socio-demographic characteristics. The prompts take the same format for all models:
```
[Instruction]
Question : [Question]
Réponse :
```
As a result, the prompts do not exploit the structures of chatbot models to ensure fairness, and the evaluation quantifies performance in a purely instruction-based approach. The figure below illustrates the results. The higher a model is positioned in the top-left corner with a small circle radius, the better the model; conversely, if a model is towards the bottom-right with a large circle, it performs less favorably.
We observe that across all models, the performance gain is logarithmic in relation to the increase in model parameters. However, for models that undergo multiple pre-trainings (vanilla, instruction, and chat), models pre-trained on instruction and chat perform significantly better in zero-shot contexts, with a notable improvement for chat-based approaches. The models we have trained demonstrate promising efficiency in this test compared to the number of parameters, indicating cost-effectiveness in a production context.
How to use bloomz-7b1-mt-sft-chat
-------------------------------
There are no specific instructions for using these models in a normal causal inference context. However, to leverage the chatbot capability of the model, an individual's prompt should be preceded by the EOS token (</s>), and the generated part should be preceded by the BOS token (<s>). The structure takes the following form:
```
</s>[human prompt 1]<s>[bot answer 1]</s>[human prompt 2]<s>
```
For example, to load the model using the HuggingFace pipeline interface:
```python
from transformers import pipeline
model = pipeline("text-generation", "cmarkea/bloomz-7b1-mt-sft-chat")
result = model("</s>C'est quoi le deep learning ?<s>", max_new_tokens=512)
result
[{'generated_text': "</s>C'est quoi le deep learning ?<s>Le deep learning
est un type de machine learning qui utilise des réseaux de neurones
artificiels pour apprendre à partir de données. Il est utilisé dans de
nombreux domaines, notamment la reconnaissance d'images, la
reconnaissance vocale, la traduction automatique et la reconnaissance
de l'écriture."}]
```
Citation
--------
```bibtex
@online{DeBloomzChat,
AUTHOR = {Cyrile Delestre},
URL = {https://huggingface.co/cmarkea/bloomz-7b1-mt-sft-chat},
YEAR = {2023},
KEYWORDS = {NLP ; Transformers ; LLM ; Bloomz},
}
``` | 16,466 | [
[
-0.037994384765625,
-0.06719970703125,
0.01715087890625,
0.03594970703125,
-0.01172637939453125,
0.00934600830078125,
-0.0190887451171875,
-0.033721923828125,
0.012298583984375,
0.02947998046875,
-0.047698974609375,
-0.032867431640625,
-0.042755126953125,
-0.005523681640625,
-0.0202484130859375,
0.06982421875,
0.00972747802734375,
0.006816864013671875,
0.00919342041015625,
-0.00629425048828125,
-0.04144287109375,
-0.060455322265625,
-0.06097412109375,
-0.0118560791015625,
0.0260772705078125,
0.0256195068359375,
0.062103271484375,
0.036712646484375,
0.0357666015625,
0.021759033203125,
-0.0142974853515625,
0.0123443603515625,
-0.05126953125,
-0.020477294921875,
0.008087158203125,
-0.0272979736328125,
-0.042083740234375,
-0.0125732421875,
0.0295867919921875,
0.0347900390625,
-0.007228851318359375,
0.004734039306640625,
0.005107879638671875,
0.049468994140625,
-0.034942626953125,
0.01507568359375,
-0.051544189453125,
-0.005069732666015625,
-0.0250701904296875,
-0.0290069580078125,
-0.021728515625,
-0.00814056396484375,
-0.004909515380859375,
-0.0311126708984375,
0.01502227783203125,
0.01107025146484375,
0.098876953125,
0.00890350341796875,
-0.022064208984375,
0.0005927085876464844,
-0.06353759765625,
0.05438232421875,
-0.06597900390625,
0.037933349609375,
0.021820068359375,
0.017425537109375,
-0.01335906982421875,
-0.041748046875,
-0.051910400390625,
-0.0030231475830078125,
-0.024658203125,
0.0263671875,
-0.01776123046875,
0.006710052490234375,
0.014556884765625,
0.0345458984375,
-0.0560302734375,
0.0259246826171875,
-0.04248046875,
-0.0280303955078125,
0.04718017578125,
0.007434844970703125,
0.0202789306640625,
-0.0242156982421875,
-0.033721923828125,
-0.0035800933837890625,
-0.05279541015625,
0.026336669921875,
0.0277557373046875,
0.03717041015625,
-0.02294921875,
0.03179931640625,
-0.015350341796875,
0.053680419921875,
-0.0141143798828125,
-0.0164642333984375,
0.038665771484375,
-0.0438232421875,
-0.013946533203125,
-0.01549530029296875,
0.07879638671875,
0.030426025390625,
0.01175689697265625,
-0.006435394287109375,
-0.007137298583984375,
0.0045013427734375,
0.0158233642578125,
-0.0660400390625,
-0.0023097991943359375,
0.037841796875,
-0.03204345703125,
-0.01873779296875,
-0.010589599609375,
-0.048187255859375,
0.021392822265625,
-0.0150909423828125,
0.031707763671875,
-0.050018310546875,
-0.0264434814453125,
0.0139617919921875,
-0.002803802490234375,
0.0249481201171875,
0.0292205810546875,
-0.06524658203125,
0.0300750732421875,
0.041656494140625,
0.079345703125,
-0.0224456787109375,
-0.02008056640625,
-0.0127105712890625,
-0.00899505615234375,
-0.0380859375,
0.05645751953125,
-0.01206207275390625,
-0.03240966796875,
-0.00632476806640625,
0.010040283203125,
-0.01922607421875,
-0.023468017578125,
0.038909912109375,
-0.0399169921875,
0.0248870849609375,
-0.00646209716796875,
-0.0496826171875,
-0.0169677734375,
0.015838623046875,
-0.047760009765625,
0.0887451171875,
0.01438140869140625,
-0.056610107421875,
0.033935546875,
-0.054534912109375,
-0.031829833984375,
0.006816864013671875,
-0.0088958740234375,
-0.041290283203125,
-0.0117340087890625,
0.0190887451171875,
0.041351318359375,
-0.0181427001953125,
0.0231781005859375,
-0.01061248779296875,
-0.0249481201171875,
-0.0002453327178955078,
-0.040496826171875,
0.06317138671875,
0.0250091552734375,
-0.03204345703125,
0.00847625732421875,
-0.06256103515625,
0.004192352294921875,
0.0181121826171875,
-0.0303802490234375,
0.00792694091796875,
0.0006608963012695312,
0.01522064208984375,
0.02996826171875,
0.020416259765625,
-0.029083251953125,
0.00414276123046875,
-0.04840087890625,
0.040313720703125,
0.048248291015625,
0.0061492919921875,
0.0252838134765625,
-0.038970947265625,
0.0296478271484375,
0.01554107666015625,
0.01439666748046875,
-0.0278778076171875,
-0.03216552734375,
-0.07305908203125,
-0.041412353515625,
0.01812744140625,
0.06097412109375,
-0.039581298828125,
0.0494384765625,
-0.0293426513671875,
-0.043609619140625,
-0.031463623046875,
0.01006317138671875,
0.0258026123046875,
0.0308837890625,
0.0311737060546875,
-0.00849151611328125,
-0.035003662109375,
-0.07049560546875,
-0.0023288726806640625,
-0.004077911376953125,
0.01125335693359375,
0.0369873046875,
0.037750244140625,
-0.0004477500915527344,
0.0736083984375,
-0.0294189453125,
-0.01406097412109375,
-0.025421142578125,
-0.004817962646484375,
0.040252685546875,
0.041351318359375,
0.07080078125,
-0.065185546875,
-0.051025390625,
-0.0007519721984863281,
-0.04693603515625,
0.0189208984375,
-0.00766754150390625,
-0.0194549560546875,
0.0206451416015625,
0.032562255859375,
-0.04412841796875,
0.042938232421875,
0.04559326171875,
-0.0189971923828125,
0.046905517578125,
-0.01727294921875,
0.0059967041015625,
-0.10888671875,
0.008514404296875,
-0.01013946533203125,
-0.0172119140625,
-0.0557861328125,
-0.017730712890625,
-0.002948760986328125,
-0.006771087646484375,
-0.046661376953125,
0.05419921875,
-0.0254974365234375,
0.016357421875,
-0.0011606216430664062,
0.00812530517578125,
-0.004947662353515625,
0.059906005859375,
0.0167236328125,
0.0703125,
0.046905517578125,
-0.058502197265625,
0.01528167724609375,
0.0208740234375,
-0.03790283203125,
0.02752685546875,
-0.054718017578125,
-0.0021209716796875,
0.0019292831420898438,
0.0171661376953125,
-0.0858154296875,
-0.00466156005859375,
0.01477813720703125,
-0.044830322265625,
0.015350341796875,
-0.007129669189453125,
-0.05023193359375,
-0.044281005859375,
-0.01306915283203125,
-0.0017642974853515625,
0.03436279296875,
-0.0232086181640625,
0.024017333984375,
-0.0016031265258789062,
-0.000591278076171875,
-0.0455322265625,
-0.055389404296875,
0.0017070770263671875,
-0.01513671875,
-0.059783935546875,
0.0288848876953125,
-0.0069732666015625,
0.0165863037109375,
-0.006916046142578125,
0.00807952880859375,
-0.006473541259765625,
0.00969696044921875,
0.0171051025390625,
0.038238525390625,
-0.01013946533203125,
0.01141357421875,
-0.01348876953125,
0.003643035888671875,
-0.0303192138671875,
-0.01012420654296875,
0.0721435546875,
-0.007411956787109375,
-0.0261383056640625,
-0.047607421875,
0.0196685791015625,
0.05084228515625,
-0.025482177734375,
0.09649658203125,
0.05548095703125,
-0.0168609619140625,
0.0026874542236328125,
-0.0489501953125,
-0.0178985595703125,
-0.03448486328125,
0.0399169921875,
-0.037445068359375,
-0.0758056640625,
0.053955078125,
0.00731658935546875,
0.0240631103515625,
0.0433349609375,
0.04931640625,
-0.00815582275390625,
0.06292724609375,
0.041168212890625,
-0.0157928466796875,
0.046417236328125,
-0.035064697265625,
0.0238189697265625,
-0.04498291015625,
-0.0088653564453125,
-0.04412841796875,
-0.014556884765625,
-0.054290771484375,
-0.0288848876953125,
0.0179595947265625,
0.011566162109375,
-0.0246734619140625,
0.03717041015625,
-0.030609130859375,
0.01210784912109375,
0.041168212890625,
0.00580596923828125,
0.00974273681640625,
0.005031585693359375,
-0.01091766357421875,
-0.0012845993041992188,
-0.06512451171875,
-0.048583984375,
0.08111572265625,
0.019195556640625,
0.0301666259765625,
0.01171112060546875,
0.044281005859375,
0.006824493408203125,
0.012664794921875,
-0.051116943359375,
0.0286102294921875,
-0.00428009033203125,
-0.07861328125,
-0.0202484130859375,
-0.0404052734375,
-0.07867431640625,
0.03411865234375,
-0.027069091796875,
-0.07427978515625,
0.0058441162109375,
0.02203369140625,
-0.033203125,
0.03240966796875,
-0.07061767578125,
0.087646484375,
-0.0208740234375,
-0.015777587890625,
-0.0157318115234375,
-0.045135498046875,
0.0203704833984375,
-0.0029621124267578125,
0.035614013671875,
-0.016143798828125,
0.010772705078125,
0.049041748046875,
-0.04620361328125,
0.055389404296875,
-0.006130218505859375,
0.00907135009765625,
0.02459716796875,
-0.0006165504455566406,
0.045806884765625,
-0.006732940673828125,
-0.0021343231201171875,
0.01824951171875,
0.00963592529296875,
-0.041900634765625,
-0.03277587890625,
0.03997802734375,
-0.081298828125,
-0.0447998046875,
-0.027801513671875,
-0.0213470458984375,
-0.0159149169921875,
0.01554107666015625,
0.040802001953125,
0.025848388671875,
-0.0189361572265625,
0.033477783203125,
0.03668212890625,
-0.0284881591796875,
0.038543701171875,
0.04986572265625,
-0.0175628662109375,
-0.020782470703125,
0.0628662109375,
0.01441192626953125,
0.032257080078125,
0.03594970703125,
0.00334930419921875,
-0.0101776123046875,
-0.048858642578125,
-0.026763916015625,
0.009674072265625,
-0.025604248046875,
-0.0225677490234375,
-0.0712890625,
-0.038360595703125,
-0.046905517578125,
-0.01024627685546875,
-0.042877197265625,
-0.027862548828125,
-0.029510498046875,
-0.01995849609375,
0.0364990234375,
0.015380859375,
0.00881195068359375,
0.0400390625,
-0.07086181640625,
0.013763427734375,
0.00940704345703125,
0.0312042236328125,
0.001247406005859375,
-0.052154541015625,
-0.02001953125,
0.015228271484375,
-0.0279998779296875,
-0.054595947265625,
0.042449951171875,
0.0233306884765625,
0.05029296875,
0.0364990234375,
-0.01143646240234375,
0.0545654296875,
-0.04461669921875,
0.06787109375,
0.016387939453125,
-0.0594482421875,
0.0494384765625,
-0.03411865234375,
0.0197601318359375,
0.036834716796875,
0.052947998046875,
-0.054290771484375,
-0.0153961181640625,
-0.06109619140625,
-0.0751953125,
0.07342529296875,
0.0287933349609375,
0.0257720947265625,
-0.0039043426513671875,
0.01727294921875,
-0.007106781005859375,
0.0208282470703125,
-0.06756591796875,
-0.0321044921875,
-0.0177459716796875,
-0.00862884521484375,
-0.0163726806640625,
-0.0133819580078125,
0.005367279052734375,
-0.017913818359375,
0.0501708984375,
0.01160430908203125,
0.0290985107421875,
-0.0005602836608886719,
-0.009429931640625,
0.01436614990234375,
0.0269317626953125,
0.05377197265625,
0.051239013671875,
-0.02191162109375,
-0.006786346435546875,
0.0203094482421875,
-0.029937744140625,
0.0026111602783203125,
-0.005153656005859375,
-0.0189361572265625,
-0.0028629302978515625,
0.038970947265625,
0.08447265625,
0.011260986328125,
-0.041961669921875,
0.03839111328125,
0.00616455078125,
-0.0228271484375,
-0.0214691162109375,
0.00531768798828125,
0.0161895751953125,
0.017730712890625,
0.02008056640625,
0.004985809326171875,
-0.002910614013671875,
-0.050628662109375,
0.0002269744873046875,
0.03448486328125,
-0.0271453857421875,
-0.025146484375,
0.04241943359375,
0.01546478271484375,
-0.025115966796875,
0.03887939453125,
-0.009246826171875,
-0.053955078125,
0.036529541015625,
0.03497314453125,
0.046142578125,
-0.026763916015625,
0.0157318115234375,
0.05108642578125,
0.0279998779296875,
-0.00228118896484375,
0.0237579345703125,
-0.0013713836669921875,
-0.06787109375,
-0.0276947021484375,
-0.060394287109375,
-0.012237548828125,
0.01377105712890625,
-0.033966064453125,
0.032623291015625,
-0.030487060546875,
-0.0272979736328125,
0.001861572265625,
0.006439208984375,
-0.060089111328125,
0.0170440673828125,
-0.0023403167724609375,
0.0718994140625,
-0.057647705078125,
0.05438232421875,
0.03729248046875,
-0.034912109375,
-0.06622314453125,
-0.028778076171875,
-0.0123443603515625,
-0.06439208984375,
0.047027587890625,
0.00455474853515625,
0.00412750244140625,
-0.004497528076171875,
-0.03790283203125,
-0.07257080078125,
0.07666015625,
0.0190582275390625,
-0.05560302734375,
-0.0089874267578125,
0.01306915283203125,
0.05438232421875,
-0.01049041748046875,
0.0210418701171875,
0.032928466796875,
0.0256195068359375,
0.02337646484375,
-0.0850830078125,
-0.00453948974609375,
-0.0225372314453125,
0.01258087158203125,
0.005962371826171875,
-0.08349609375,
0.06787109375,
-0.002994537353515625,
-0.0200653076171875,
-0.0196990966796875,
0.048248291015625,
0.00958251953125,
-0.004215240478515625,
0.029296875,
0.049835205078125,
0.039825439453125,
-0.0241241455078125,
0.07171630859375,
-0.044921875,
0.0290069580078125,
0.08099365234375,
0.01111602783203125,
0.0704345703125,
0.022735595703125,
-0.042755126953125,
0.0310821533203125,
0.0694580078125,
0.0010824203491210938,
0.031524658203125,
-0.004207611083984375,
-0.0114288330078125,
-0.0188140869140625,
0.01403045654296875,
-0.032684326171875,
0.0296630859375,
0.03277587890625,
-0.01611328125,
0.0009555816650390625,
-0.001064300537109375,
0.036102294921875,
-0.0185089111328125,
-0.00969696044921875,
0.066162109375,
-0.006023406982421875,
-0.0467529296875,
0.0587158203125,
0.0009713172912597656,
0.059967041015625,
-0.05792236328125,
0.022613525390625,
-0.011260986328125,
0.009765625,
-0.019989013671875,
-0.039642333984375,
0.0225372314453125,
-0.004924774169921875,
-0.002033233642578125,
-0.009857177734375,
0.0330810546875,
-0.0204315185546875,
-0.0465087890625,
0.0244293212890625,
0.040985107421875,
0.007305145263671875,
-0.0054931640625,
-0.05322265625,
-0.00167083740234375,
0.00003552436828613281,
-0.041229248046875,
0.0167694091796875,
0.038421630859375,
0.00862884521484375,
0.04986572265625,
0.05841064453125,
0.0167388916015625,
0.01543426513671875,
0.002758026123046875,
0.0777587890625,
-0.04522705078125,
-0.045745849609375,
-0.06756591796875,
0.0309295654296875,
-0.00925445556640625,
-0.021209716796875,
0.06500244140625,
0.05242919921875,
0.057647705078125,
0.009002685546875,
0.053985595703125,
-0.011199951171875,
0.048187255859375,
-0.0293731689453125,
0.04364013671875,
-0.046844482421875,
-0.0010404586791992188,
-0.0257720947265625,
-0.07110595703125,
-0.01357269287109375,
0.045989990234375,
-0.026336669921875,
0.01690673828125,
0.044708251953125,
0.06390380859375,
0.00859832763671875,
-0.009857177734375,
0.03192138671875,
0.01509857177734375,
0.03314208984375,
0.042083740234375,
0.03619384765625,
-0.042999267578125,
0.059326171875,
-0.0206756591796875,
-0.0151519775390625,
-0.0301971435546875,
-0.04693603515625,
-0.0743408203125,
-0.051483154296875,
-0.024383544921875,
-0.0201263427734375,
-0.00032830238342285156,
0.06671142578125,
0.0716552734375,
-0.0657958984375,
-0.0249786376953125,
-0.00629425048828125,
0.0016469955444335938,
-0.0206298828125,
-0.0164794921875,
0.033050537109375,
-0.03863525390625,
-0.05816650390625,
0.006591796875,
0.0238037109375,
0.0226593017578125,
-0.0155792236328125,
-0.00313568115234375,
-0.033935546875,
0.0296478271484375,
0.0465087890625,
0.01401519775390625,
-0.041290283203125,
-0.01727294921875,
0.00812530517578125,
-0.016082763671875,
0.027557373046875,
0.032928466796875,
-0.0296630859375,
0.0300750732421875,
0.035003662109375,
0.02569580078125,
0.0643310546875,
-0.0066375732421875,
0.0236663818359375,
-0.050048828125,
0.030975341796875,
0.00868988037109375,
0.022308349609375,
0.0180206298828125,
-0.023773193359375,
0.0350341796875,
0.0199127197265625,
-0.03302001953125,
-0.05096435546875,
-0.0055389404296875,
-0.0721435546875,
-0.0162811279296875,
0.07904052734375,
-0.0168609619140625,
-0.023040771484375,
-0.00045108795166015625,
-0.021270751953125,
0.0172576904296875,
-0.03936767578125,
0.043609619140625,
0.0823974609375,
-0.006519317626953125,
-0.023162841796875,
-0.055145263671875,
0.046661376953125,
0.0213470458984375,
-0.06976318359375,
-0.004512786865234375,
0.0222015380859375,
0.0214691162109375,
0.0161895751953125,
0.0655517578125,
-0.01262664794921875,
0.0093231201171875,
-0.0020885467529296875,
0.00714874267578125,
0.004901885986328125,
-0.004848480224609375,
0.003894805908203125,
0.001873016357421875,
-0.00736236572265625,
-0.01186370849609375
]
] |
Locutusque/gpt2-large-conversational | 2023-10-09T00:05:12.000Z | [
"transformers",
"pytorch",
"safetensors",
"gpt2",
"text-generation",
"en",
"dataset:Locutusque/ColumnedChatCombined",
"dataset:crumb/Clean-Instruct-440k",
"doi:10.57967/hf/1215",
"license:openrail",
"endpoints_compatible",
"has_space",
"text-generation-inference",
"region:us"
] | text-generation | Locutusque | null | null | Locutusque/gpt2-large-conversational | 3 | 5,921 | transformers | 2023-06-16T04:43:45 | ---
license: openrail
datasets:
- Locutusque/ColumnedChatCombined
- crumb/Clean-Instruct-440k
language:
- en
metrics:
- bleu
- perplexity
- loss
- reward
- penalty
pipeline_tag: text-generation
---
# Model Card
* this model is deprecated, please see https://huggingface.co/Locutusque/gpt2-large-conversational-retrain for a better performing model.
## Model Details
- Model Name: gpt2-large-conversational
- Model Type: Language Modeling
- Task: Generating Conversational Responses
- Hardware: 1x Nvidia Titan V
- Description: This model is trained on a dataset of conversations between a user and an AI assistant, with the goal of generating a coherent and relevant response to the user's input. It uses the GPT-2 architecture, a state-of-the-art transformer-based language model that is capable of generating high-quality text with a wide range of styles and tones. The model is fine-tuned on the conversational data using maximum likelihood estimation, and is evaluated based on its ability to generate responses that are both grammatically correct and semantically relevant to the user's input.
## Intended Use
This model is intended to be used for generating conversational responses in a variety of contexts, such as chatbots, virtual assistants, and customer service applications. It is designed to provide natural and engaging responses to user input, with a focus on maintaining a consistent tone and style throughout the conversation. The model is suitable for use in both text-based and voice-based interfaces, and can be easily integrated into existing applications using the PyTorch and Transformers frameworks.
## Training Data
The model is trained on a large dataset of conversational data, consisting of interactions between users and an AI assistant. The data is preprocessed to remove any sensitive information and is formatted in a way that is suitable for training a language model. The training data is split into a training set and a validation set, with the training set used to update the model parameters and the validation set used to evaluate the model performance. The model was trained on 550,000 examples over 687,500 steps, it achieved decent metrics.
## Model Architecture
The model architecture used in this model is GPT-2, a transformer-based language model that is capable of generating high-quality text with a wide range of styles and tones. The GPT-2 architecture consists of a multi-layered decoder-only transformer, with self-attention mechanisms that allow the model to capture long-term dependencies and generate coherent text.
## Evaluation Metrics
The model is evaluated based on several metrics, including loss, reward, penalty, BLEU score, and perplexity. The loss metric is calculated during training and reflects the difference between the predicted output and the actual output. The reward metric is based on the number of correct words generated by the model, while the penalty metric penalizes the model for repeating words consecutively. The BLEU score measures the similarity between the generated text and the ground truth text, while the perplexity metric measures how well the model is able to predict the next word in a sequence. During validation, the model achieved the following metrics:
- BLEU score: 12
- perplexity: 38
- loss: 3.1
## Limitations and Bias
This model is not suitable for all use cases due to its limited training time on a weak computer. As a result, it may produce irrelevant or nonsensical responses. Additionally, it has not been fine-tuned to remember the chat history, is unable to provide follow-up responses, and it does not know the answer to many questions (it was only fine-tuned to respond in a conversational way). For optimal performance, I recommend using a GPU with at least 12 GB of VRAM and downloading the model manually instead of using the Transformers library. Here's how you should deploy the model:
```python
import torch
from transformers import GPT2Tokenizer, GPT2LMHeadModel
start_token = "<|ASSISTANT|>"
end_token = "<|"
tokenizer = GPT2Tokenizer.from_pretrained('gpt2-large')
model = GPT2LMHeadModel.from_pretrained('gpt2-large')
tokenizer.add_special_tokens({'pad_token': '[PAD]'})
special_tokens = {
"additional_special_tokens": ["<|USER|>", "<|ASSISTANT|>"]
}
tokenizer.add_special_tokens(special_tokens)
model.resize_token_embeddings(len(tokenizer))
model.load_state_dict(torch.load("path/to/model"))
device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
model.to(device)
def generate_text(model, tokenizer, prompt, max_length=256):
prompt = f'<|USER|> {prompt} <|ASSISTANT|> '
input_ids = tokenizer.encode(prompt, add_special_tokens=True, return_tensors="pt").to(device)
attention_mask = torch.ones_like(input_ids).to(device)
output = model.generate(input_ids,
max_length=max_length,
do_sample=True,
top_k=35,
top_p=0.80,
pad_token_id=tokenizer.pad_token_id,
eos_token_id=tokenizer.eos_token_id,
attention_mask=attention_mask)
output_ids = tokenizer.decode(output[0], skip_special_tokens=False)
return output_ids
# Loop to interact with the model
while True:
prompt = input("Enter a prompt (or 'q' to quit): ")
if prompt == "q":
break
output_text = generate_text(model, tokenizer, prompt)
text_between_tokens = output_text[output_text.find(start_token) + len(start_token):]
out = text_between_tokens[:text_between_tokens.find(end_token)]
print(out)
```
## Deploying and training the model
The model has been fine-tuned on a specific input format that goes like this ```"<|USER|> {user prompt} <|ASSISTANT|> {model prediction} ".``` For the best performance from the model the input text should be as follows ```<|USER|> {dataset prompt} <|ASSISTANT|> ``` and the target/label should be as follows ```<|USER|> {dataset prompt} <|ASSISTANT|> {dataset output} ```. This model is also very fun to play with in text generation webui | 6,119 | [
[
-0.0147705078125,
-0.07537841796875,
0.0214691162109375,
0.0189056396484375,
-0.021820068359375,
-0.01406097412109375,
-0.021942138671875,
-0.0303955078125,
-0.020294189453125,
0.0191802978515625,
-0.036529541015625,
-0.0285797119140625,
-0.049957275390625,
-0.01175689697265625,
-0.01373291015625,
0.09423828125,
0.00547027587890625,
-0.010467529296875,
-0.004367828369140625,
0.0134124755859375,
-0.053619384765625,
-0.04302978515625,
-0.07110595703125,
-0.036773681640625,
0.0185089111328125,
0.0167694091796875,
0.04302978515625,
0.0286407470703125,
0.0267486572265625,
0.0212860107421875,
-0.0169830322265625,
0.007678985595703125,
-0.043365478515625,
-0.0073394775390625,
-0.005641937255859375,
-0.0276641845703125,
-0.039947509765625,
0.005184173583984375,
0.036651611328125,
0.01462554931640625,
-0.006439208984375,
0.01488494873046875,
0.0186004638671875,
0.007843017578125,
-0.01013946533203125,
0.02813720703125,
-0.04473876953125,
-0.01336669921875,
0.006496429443359375,
0.004283905029296875,
-0.037841796875,
-0.01218414306640625,
0.0187530517578125,
-0.03887939453125,
0.0250701904296875,
-0.0009608268737792969,
0.08819580078125,
0.013641357421875,
-0.0258636474609375,
-0.0175018310546875,
-0.0587158203125,
0.0506591796875,
-0.061920166015625,
0.02423095703125,
0.0302886962890625,
0.0171356201171875,
-0.0003440380096435547,
-0.07305908203125,
-0.05169677734375,
-0.028961181640625,
-0.01398468017578125,
0.0181427001953125,
-0.023681640625,
0.0007572174072265625,
0.0303497314453125,
0.01132965087890625,
-0.067626953125,
0.006885528564453125,
-0.03118896484375,
-0.022064208984375,
0.035491943359375,
0.021240234375,
0.0194091796875,
-0.0152435302734375,
-0.02349853515625,
-0.01690673828125,
-0.035369873046875,
0.0032444000244140625,
0.0248870849609375,
0.0174713134765625,
-0.031585693359375,
0.041015625,
-0.007045745849609375,
0.03631591796875,
0.0115814208984375,
-0.008575439453125,
0.023956298828125,
-0.02484130859375,
-0.030487060546875,
-0.0198822021484375,
0.0850830078125,
0.029052734375,
0.01318359375,
-0.01024627685546875,
-0.01535797119140625,
0.009429931640625,
0.00965118408203125,
-0.07733154296875,
-0.0270843505859375,
0.0287017822265625,
-0.021026611328125,
-0.0276641845703125,
-0.0142364501953125,
-0.05029296875,
-0.0103302001953125,
-0.00791168212890625,
0.0361328125,
-0.03717041015625,
-0.032257080078125,
-0.0027141571044921875,
-0.015625,
0.023773193359375,
0.0168304443359375,
-0.09234619140625,
0.00876617431640625,
0.0322265625,
0.06695556640625,
-0.00125885009765625,
-0.0254364013671875,
-0.019195556640625,
0.0005249977111816406,
-0.01059722900390625,
0.039642333984375,
-0.02081298828125,
-0.0357666015625,
-0.01032257080078125,
0.00884246826171875,
-0.01470947265625,
-0.0263824462890625,
0.037841796875,
-0.0294189453125,
0.04254150390625,
-0.011016845703125,
-0.035614013671875,
-0.01207733154296875,
0.0243377685546875,
-0.0292510986328125,
0.09521484375,
0.023681640625,
-0.064697265625,
0.00988006591796875,
-0.05609130859375,
-0.0270843505859375,
0.007076263427734375,
-0.00469970703125,
-0.0457763671875,
-0.0040740966796875,
0.0116729736328125,
0.031646728515625,
-0.033355712890625,
0.0271453857421875,
0.006359100341796875,
-0.025787353515625,
0.01152801513671875,
-0.0389404296875,
0.060455322265625,
0.01296234130859375,
-0.04498291015625,
0.0200347900390625,
-0.047393798828125,
0.005451202392578125,
0.03045654296875,
-0.032623291015625,
0.01462554931640625,
-0.013153076171875,
0.03778076171875,
0.016082763671875,
0.02386474609375,
-0.037109375,
0.0150146484375,
-0.0379638671875,
0.057708740234375,
0.046234130859375,
-0.00905609130859375,
0.018585205078125,
-0.0127105712890625,
0.023712158203125,
0.0029754638671875,
0.00921630859375,
-0.006572723388671875,
-0.053253173828125,
-0.0628662109375,
-0.0009527206420898438,
0.0213775634765625,
0.06317138671875,
-0.05010986328125,
0.042938232421875,
-0.0203857421875,
-0.047760009765625,
-0.0197906494140625,
-0.002197265625,
0.036865234375,
0.0504150390625,
0.0264892578125,
-0.01236724853515625,
-0.03662109375,
-0.05743408203125,
0.002613067626953125,
-0.0247955322265625,
-0.01306915283203125,
0.0280609130859375,
0.052459716796875,
-0.0299530029296875,
0.065673828125,
-0.04132080078125,
-0.013641357421875,
-0.0428466796875,
0.0224609375,
0.01216888427734375,
0.047393798828125,
0.029449462890625,
-0.05133056640625,
-0.028900146484375,
0.0033855438232421875,
-0.056549072265625,
0.008941650390625,
-0.007289886474609375,
-0.0029754638671875,
0.0208282470703125,
0.0185089111328125,
-0.0689697265625,
0.0288238525390625,
0.03570556640625,
-0.03472900390625,
0.0467529296875,
-0.0220489501953125,
0.0009832382202148438,
-0.10601806640625,
0.023712158203125,
0.00319671630859375,
-0.014190673828125,
-0.0533447265625,
-0.0007886886596679688,
-0.005901336669921875,
-0.01461029052734375,
-0.033782958984375,
0.057708740234375,
-0.020477294921875,
0.01515960693359375,
-0.0271453857421875,
0.0071258544921875,
0.002346038818359375,
0.054046630859375,
0.0097198486328125,
0.075927734375,
0.032958984375,
-0.04302978515625,
0.045257568359375,
0.033203125,
-0.01641845703125,
0.0240325927734375,
-0.0853271484375,
0.036224365234375,
0.002277374267578125,
0.01428985595703125,
-0.10125732421875,
-0.0246734619140625,
0.021759033203125,
-0.062164306640625,
0.0281829833984375,
-0.01084136962890625,
-0.044281005859375,
-0.04229736328125,
-0.00025844573974609375,
0.02618408203125,
0.056884765625,
-0.0260467529296875,
0.03955078125,
0.0166168212890625,
-0.01422119140625,
-0.0252685546875,
-0.0657958984375,
0.00354766845703125,
-0.0159912109375,
-0.052886962890625,
0.0213775634765625,
0.003604888916015625,
0.01319122314453125,
-0.0179901123046875,
0.028900146484375,
0.00832366943359375,
0.00042629241943359375,
0.0215911865234375,
0.02752685546875,
-0.00836944580078125,
0.0147247314453125,
-0.00511932373046875,
-0.01361846923828125,
0.00620269775390625,
-0.0247344970703125,
0.07269287109375,
-0.003124237060546875,
0.0013256072998046875,
-0.04986572265625,
0.01180267333984375,
0.032928466796875,
-0.0213623046875,
0.058502197265625,
0.07647705078125,
-0.038665771484375,
0.01291656494140625,
-0.04345703125,
-0.0228424072265625,
-0.03570556640625,
0.06817626953125,
-0.0313720703125,
-0.054046630859375,
0.038238525390625,
0.0139617919921875,
0.0123291015625,
0.0413818359375,
0.070068359375,
0.006114959716796875,
0.08770751953125,
0.0413818359375,
-0.006084442138671875,
0.038604736328125,
-0.0399169921875,
0.01416015625,
-0.0709228515625,
-0.026275634765625,
-0.0237274169921875,
-0.0030345916748046875,
-0.039031982421875,
-0.0238189697265625,
0.01410675048828125,
0.00699615478515625,
-0.038299560546875,
0.0272369384765625,
-0.052154541015625,
0.01336669921875,
0.051666259765625,
0.01264190673828125,
-0.002666473388671875,
0.0089874267578125,
-0.004573822021484375,
0.0003457069396972656,
-0.037994384765625,
-0.047882080078125,
0.08807373046875,
0.0458984375,
0.04339599609375,
0.001102447509765625,
0.0226593017578125,
-0.0004749298095703125,
0.011138916015625,
-0.049957275390625,
0.031982421875,
0.003948211669921875,
-0.04412841796875,
-0.02923583984375,
-0.0309600830078125,
-0.06390380859375,
0.00794219970703125,
-0.0015716552734375,
-0.08099365234375,
-0.0152740478515625,
0.01462554931640625,
-0.033660888671875,
0.017242431640625,
-0.07061767578125,
0.082275390625,
-0.012054443359375,
-0.035125732421875,
0.0052032470703125,
-0.045501708984375,
0.022064208984375,
0.01959228515625,
-0.0163726806640625,
0.01137542724609375,
0.0164794921875,
0.07366943359375,
-0.022918701171875,
0.064697265625,
-0.01371002197265625,
-0.0024738311767578125,
0.03289794921875,
-0.01116180419921875,
0.05230712890625,
0.007160186767578125,
0.01227569580078125,
0.005939483642578125,
-0.01383209228515625,
-0.01947021484375,
-0.037567138671875,
0.044677734375,
-0.08428955078125,
-0.0338134765625,
-0.0247955322265625,
-0.04046630859375,
-0.01001739501953125,
0.0110626220703125,
0.056427001953125,
0.03643798828125,
-0.0143585205078125,
-0.00689697265625,
0.048828125,
-0.02587890625,
0.0355224609375,
0.00559234619140625,
-0.0098876953125,
-0.043243408203125,
0.06866455078125,
-0.0016727447509765625,
0.019561767578125,
0.0187835693359375,
0.006351470947265625,
-0.05010986328125,
-0.03271484375,
-0.053619384765625,
0.0177764892578125,
-0.038818359375,
-0.01132965087890625,
-0.062347412109375,
-0.0274658203125,
-0.04949951171875,
0.0168609619140625,
-0.02423095703125,
-0.0123138427734375,
-0.03887939453125,
-0.01517486572265625,
0.03656005859375,
0.049346923828125,
0.01885986328125,
0.049835205078125,
-0.041473388671875,
0.01904296875,
0.03582763671875,
0.019866943359375,
0.0031566619873046875,
-0.06878662109375,
-0.01279449462890625,
0.0193023681640625,
-0.04779052734375,
-0.0626220703125,
0.0308685302734375,
-0.00362396240234375,
0.02130126953125,
0.0178070068359375,
-0.0284423828125,
0.037841796875,
-0.0323486328125,
0.07635498046875,
0.0200347900390625,
-0.06591796875,
0.050384521484375,
-0.03546142578125,
0.02984619140625,
0.0207366943359375,
0.007152557373046875,
-0.051422119140625,
-0.01436614990234375,
-0.04217529296875,
-0.06329345703125,
0.064208984375,
0.037841796875,
0.0294189453125,
0.001773834228515625,
0.0277862548828125,
-0.01412200927734375,
0.0105438232421875,
-0.06640625,
-0.0218505859375,
-0.028533935546875,
-0.0196380615234375,
-0.00548553466796875,
-0.0280609130859375,
0.002445220947265625,
-0.024749755859375,
0.057708740234375,
0.0003387928009033203,
0.05914306640625,
0.005535125732421875,
-0.00455474853515625,
0.010498046875,
0.00795745849609375,
0.05572509765625,
0.039306640625,
-0.01263427734375,
-0.01120758056640625,
0.019500732421875,
-0.052459716796875,
0.005279541015625,
0.021697998046875,
-0.0130615234375,
-0.0007276535034179688,
0.0125732421875,
0.09051513671875,
-0.0106353759765625,
-0.0230865478515625,
0.031829833984375,
-0.0102386474609375,
-0.023345947265625,
-0.019378662109375,
0.0007414817810058594,
0.02197265625,
0.0149383544921875,
0.022064208984375,
-0.01163482666015625,
-0.0009555816650390625,
-0.0377197265625,
0.00490570068359375,
0.028564453125,
-0.0230560302734375,
-0.02667236328125,
0.08087158203125,
0.023773193359375,
-0.035552978515625,
0.06988525390625,
-0.026763916015625,
-0.05120849609375,
0.04827880859375,
0.0560302734375,
0.06787109375,
-0.0167388916015625,
0.01556396484375,
0.04351806640625,
0.020599365234375,
-0.00890350341796875,
0.01367950439453125,
0.01454925537109375,
-0.045135498046875,
-0.0157623291015625,
-0.038818359375,
-0.01212310791015625,
0.03594970703125,
-0.027618408203125,
0.029937744140625,
-0.03411865234375,
-0.03277587890625,
-0.0117340087890625,
0.0135955810546875,
-0.06109619140625,
0.033538818359375,
0.01299285888671875,
0.05120849609375,
-0.0643310546875,
0.060791015625,
0.04180908203125,
-0.047607421875,
-0.0733642578125,
-0.012298583984375,
-0.01467132568359375,
-0.060943603515625,
0.036163330078125,
0.026031494140625,
0.01311492919921875,
0.016845703125,
-0.04229736328125,
-0.05230712890625,
0.09197998046875,
0.021270751953125,
-0.0411376953125,
-0.0311737060546875,
0.0232086181640625,
0.045654296875,
-0.0156402587890625,
0.0611572265625,
0.0472412109375,
0.0220489501953125,
0.0144805908203125,
-0.087158203125,
0.0151824951171875,
-0.02337646484375,
0.0144805908203125,
0.000017344951629638672,
-0.06396484375,
0.07659912109375,
-0.018463134765625,
-0.01297760009765625,
0.0214385986328125,
0.052032470703125,
0.018585205078125,
0.006229400634765625,
0.0153045654296875,
0.048095703125,
0.053924560546875,
-0.0264129638671875,
0.075927734375,
-0.0220184326171875,
0.05584716796875,
0.07989501953125,
-0.0013332366943359375,
0.05706787109375,
0.0185089111328125,
-0.01219940185546875,
0.0287322998046875,
0.05279541015625,
-0.0194549560546875,
0.0343017578125,
0.00928497314453125,
-0.0036773681640625,
-0.001308441162109375,
0.004222869873046875,
-0.03704833984375,
0.0313720703125,
0.0185546875,
-0.036865234375,
-0.00457763671875,
0.00531005859375,
0.0148162841796875,
-0.0276031494140625,
0.00109100341796875,
0.0601806640625,
-0.00890350341796875,
-0.05108642578125,
0.05792236328125,
0.0215911865234375,
0.05889892578125,
-0.046173095703125,
-0.002666473388671875,
-0.0182647705078125,
0.0192718505859375,
-0.00743865966796875,
-0.047088623046875,
0.004878997802734375,
0.004180908203125,
-0.0139312744140625,
0.0020542144775390625,
0.035552978515625,
-0.037384033203125,
-0.047393798828125,
0.0093841552734375,
0.0265045166015625,
0.0195159912109375,
-0.02020263671875,
-0.06109619140625,
0.00836944580078125,
0.004039764404296875,
-0.036407470703125,
0.015533447265625,
0.0248870849609375,
0.007678985595703125,
0.044281005859375,
0.040618896484375,
-0.00949859619140625,
0.0003936290740966797,
0.0082855224609375,
0.06744384765625,
-0.0416259765625,
-0.0364990234375,
-0.071044921875,
0.049530029296875,
-0.01036834716796875,
-0.05194091796875,
0.046112060546875,
0.056243896484375,
0.0513916015625,
-0.006526947021484375,
0.072509765625,
-0.011444091796875,
0.0270233154296875,
-0.029693603515625,
0.05914306640625,
-0.0268402099609375,
0.006816864013671875,
-0.01953125,
-0.065673828125,
0.0000017881393432617188,
0.058197021484375,
-0.027496337890625,
0.0252838134765625,
0.0430908203125,
0.0697021484375,
-0.005672454833984375,
0.00015556812286376953,
0.0210418701171875,
0.0191802978515625,
0.040283203125,
0.066162109375,
0.042694091796875,
-0.064697265625,
0.05377197265625,
-0.007740020751953125,
-0.017333984375,
-0.00794219970703125,
-0.03814697265625,
-0.08380126953125,
-0.044464111328125,
-0.02288818359375,
-0.046112060546875,
0.005344390869140625,
0.0645751953125,
0.054290771484375,
-0.053863525390625,
-0.01207733154296875,
-0.016937255859375,
-0.0125732421875,
-0.0014085769653320312,
-0.023681640625,
0.034423828125,
-0.020416259765625,
-0.07232666015625,
-0.005649566650390625,
-0.00806427001953125,
0.025970458984375,
-0.0163116455078125,
-0.008209228515625,
-0.00864410400390625,
-0.0037860870361328125,
0.042144775390625,
0.009429931640625,
-0.05364990234375,
-0.0245819091796875,
0.001712799072265625,
-0.0260467529296875,
0.004970550537109375,
0.03399658203125,
-0.05047607421875,
0.0208892822265625,
0.028533935546875,
0.034210205078125,
0.050750732421875,
-0.004192352294921875,
0.03564453125,
-0.053192138671875,
0.020843505859375,
0.01227569580078125,
0.0164642333984375,
0.03472900390625,
-0.040740966796875,
0.026214599609375,
0.02130126953125,
-0.05010986328125,
-0.054931640625,
0.0063629150390625,
-0.06915283203125,
-0.004116058349609375,
0.1112060546875,
-0.0125732421875,
-0.01226806640625,
-0.0015659332275390625,
-0.041259765625,
0.036590576171875,
-0.035003662109375,
0.0506591796875,
0.054473876953125,
0.000012934207916259766,
-0.01654052734375,
-0.03692626953125,
0.037322998046875,
0.01800537109375,
-0.060028076171875,
0.01320648193359375,
0.0167388916015625,
0.041595458984375,
0.01177215576171875,
0.051788330078125,
0.005069732666015625,
0.0230865478515625,
0.0121002197265625,
0.00879669189453125,
-0.01259613037109375,
-0.01041412353515625,
-0.0145416259765625,
-0.0014057159423828125,
-0.0022068023681640625,
-0.031646728515625
]
] |
darkstorm2150/Protogen_Infinity_Official_Release | 2023-01-27T17:43:23.000Z | [
"diffusers",
"stable-diffusion",
"stable-diffusion-diffusers",
"text-to-image",
"art",
"artistic",
"en",
"license:creativeml-openrail-m",
"endpoints_compatible",
"has_space",
"diffusers:StableDiffusionPipeline",
"region:us"
] | text-to-image | darkstorm2150 | null | null | darkstorm2150/Protogen_Infinity_Official_Release | 61 | 5,920 | diffusers | 2023-01-13T07:57:14 | ---
language:
- en
tags:
- stable-diffusion
- stable-diffusion-diffusers
- text-to-image
- art
- artistic
- diffusers
inference: true
license: creativeml-openrail-m
---
## Pending info card
I will be updating soon
## Model Weights
 | 351 | [
[
-0.0246429443359375,
-0.0014390945434570312,
0.040374755859375,
0.028594970703125,
-0.01824951171875,
0.0112152099609375,
0.0135040283203125,
-0.043701171875,
0.0335693359375,
0.039825439453125,
-0.02947998046875,
-0.039459228515625,
-0.042724609375,
-0.021453857421875,
-0.00861358642578125,
0.053558349609375,
0.00731658935546875,
0.0148162841796875,
-0.0191497802734375,
-0.0179595947265625,
-0.0236663818359375,
-0.024749755859375,
-0.050079345703125,
-0.0465087890625,
0.05584716796875,
0.04443359375,
0.055694580078125,
0.04437255859375,
0.05023193359375,
0.022735595703125,
-0.0157012939453125,
-0.02484130859375,
-0.01194000244140625,
0.0084075927734375,
0.005901336669921875,
-0.038421630859375,
-0.0989990234375,
0.007144927978515625,
0.06915283203125,
0.046234130859375,
-0.0005464553833007812,
0.054046630859375,
-0.0020732879638671875,
0.033355712890625,
-0.03662109375,
0.0186920166015625,
-0.0162506103515625,
-0.0057220458984375,
-0.0357666015625,
0.008026123046875,
-0.010284423828125,
-0.01507568359375,
-0.0114593505859375,
-0.0570068359375,
0.01419830322265625,
0.0190277099609375,
0.083251953125,
0.014984130859375,
-0.02264404296875,
0.0054473876953125,
-0.0406494140625,
0.0300445556640625,
-0.040771484375,
0.03729248046875,
0.0147247314453125,
0.033477783203125,
-0.0010747909545898438,
-0.07904052734375,
-0.0129547119140625,
-0.01708984375,
0.0008616447448730469,
0.017333984375,
-0.03277587890625,
0.0154571533203125,
0.06353759765625,
0.045135498046875,
-0.0072021484375,
0.013671875,
-0.055450439453125,
-0.0235137939453125,
0.053497314453125,
0.00156402587890625,
0.032440185546875,
0.0245513916015625,
-0.0252532958984375,
-0.0028934478759765625,
-0.034454345703125,
-0.0178985595703125,
0.0220489501953125,
0.014984130859375,
-0.06396484375,
0.06475830078125,
-0.01154327392578125,
0.05841064453125,
0.0277557373046875,
0.036163330078125,
0.040069580078125,
0.0042724609375,
-0.0240325927734375,
-0.01149749755859375,
0.039886474609375,
0.041748046875,
-0.01004791259765625,
0.0200958251953125,
-0.0085601806640625,
-0.021453857421875,
0.0118408203125,
-0.061767578125,
0.0010805130004882812,
0.0262908935546875,
-0.046356201171875,
-0.032684326171875,
0.016387939453125,
-0.07171630859375,
-0.03436279296875,
0.00855255126953125,
0.0242767333984375,
0.005977630615234375,
-0.032867431640625,
0.01395416259765625,
-0.0272064208984375,
0.001750946044921875,
0.046234130859375,
-0.05712890625,
0.00225830078125,
0.010162353515625,
0.036285400390625,
0.055877685546875,
0.0171966552734375,
-0.0233612060546875,
0.0072784423828125,
-0.029205322265625,
0.049163818359375,
-0.0264892578125,
-0.0268707275390625,
-0.01183319091796875,
-0.0035991668701171875,
0.0281982421875,
-0.03497314453125,
0.0693359375,
-0.045806884765625,
-0.0016222000122070312,
-0.0310516357421875,
-0.0261688232421875,
0.0203094482421875,
0.0157318115234375,
-0.09197998046875,
0.06256103515625,
0.0265045166015625,
-0.05230712890625,
0.0212249755859375,
-0.053924560546875,
0.002960205078125,
0.0439453125,
-0.0019588470458984375,
-0.03985595703125,
0.0303955078125,
-0.02532958984375,
0.0264434814453125,
-0.012725830078125,
0.0040130615234375,
-0.057647705078125,
-0.0134429931640625,
0.01275634765625,
0.0075836181640625,
0.034027099609375,
0.04998779296875,
0.004825592041015625,
0.005947113037109375,
-0.0635986328125,
-0.0082550048828125,
0.01275634765625,
-0.020660400390625,
0.0316162109375,
-0.0253753662109375,
-0.009765625,
0.013427734375,
0.064208984375,
-0.053009033203125,
0.032073974609375,
0.01418304443359375,
-0.00873565673828125,
0.057861328125,
-0.0035762786865234375,
0.0121307373046875,
-0.0269317626953125,
0.05596923828125,
-0.0297088623046875,
0.02508544921875,
0.027069091796875,
-0.047332763671875,
-0.06048583984375,
0.01126861572265625,
0.055572509765625,
0.0249481201171875,
-0.00510406494140625,
0.02423095703125,
0.006252288818359375,
-0.045745849609375,
-0.0228729248046875,
-0.0229034423828125,
0.0170135498046875,
0.01212310791015625,
0.00870513916015625,
-0.0236968994140625,
-0.04931640625,
-0.054656982421875,
-0.0078277587890625,
-0.02520751953125,
-0.0160369873046875,
0.007106781005859375,
0.0209503173828125,
-0.020843505859375,
0.02923583984375,
-0.06219482421875,
-0.01042938232421875,
-0.0226287841796875,
0.039093017578125,
0.02764892578125,
0.05340576171875,
0.0618896484375,
-0.07427978515625,
-0.0252532958984375,
-0.01078033447265625,
-0.030731201171875,
-0.03070068359375,
0.007171630859375,
-0.009735107421875,
-0.041107177734375,
0.0482177734375,
-0.047393798828125,
0.052490234375,
0.036895751953125,
-0.0609130859375,
0.027374267578125,
-0.04248046875,
-0.00029730796813964844,
-0.06524658203125,
0.020172119140625,
0.003391265869140625,
-0.01311492919921875,
-0.0189666748046875,
0.0260162353515625,
0.01352691650390625,
-0.0223388671875,
-0.038848876953125,
0.068359375,
-0.051910400390625,
0.005077362060546875,
-0.01165008544921875,
-0.00666046142578125,
0.0009598731994628906,
-0.014434814453125,
-0.00974273681640625,
0.036865234375,
0.047821044921875,
-0.039764404296875,
0.049285888671875,
0.0062255859375,
0.0176849365234375,
0.03729248046875,
-0.08111572265625,
-0.020599365234375,
-0.02734375,
0.017578125,
-0.0670166015625,
-0.050537109375,
0.036376953125,
-0.04962158203125,
0.040985107421875,
-0.00768280029296875,
-0.04254150390625,
-0.06634521484375,
-0.01067352294921875,
0.03302001953125,
0.05157470703125,
-0.01169586181640625,
0.06317138671875,
0.0162506103515625,
0.00838470458984375,
-0.0229034423828125,
-0.059417724609375,
0.0022735595703125,
0.011016845703125,
-0.017364501953125,
0.048675537109375,
-0.0277862548828125,
-0.035125732421875,
0.0012102127075195312,
-0.01323699951171875,
-0.00997161865234375,
-0.01617431640625,
0.01446533203125,
0.01329803466796875,
0.00977325439453125,
0.0275115966796875,
0.006420135498046875,
-0.046234130859375,
-0.0079193115234375,
-0.01177215576171875,
0.03271484375,
-0.024688720703125,
-0.0009350776672363281,
-0.032867431640625,
0.007110595703125,
0.05816650390625,
-0.010284423828125,
0.033294677734375,
0.057647705078125,
-0.060394287109375,
0.03021240234375,
-0.05413818359375,
-0.00946044921875,
-0.035919189453125,
0.01450347900390625,
-0.0428466796875,
-0.047882080078125,
0.0556640625,
-0.0160369873046875,
0.0002751350402832031,
0.07720947265625,
0.033447265625,
0.0002560615539550781,
0.0943603515625,
0.078125,
-0.00440216064453125,
0.0311737060546875,
-0.0268707275390625,
-0.01169586181640625,
-0.071533203125,
-0.02911376953125,
-0.0380859375,
-0.0266571044921875,
-0.042724609375,
-0.04791259765625,
0.0017633438110351562,
0.037841796875,
-0.003551483154296875,
0.047088623046875,
-0.03466796875,
0.0131988525390625,
0.032196044921875,
0.01910400390625,
-0.0021457672119140625,
-0.03778076171875,
-0.0147705078125,
-0.004131317138671875,
-0.0285491943359375,
-0.012451171875,
0.0877685546875,
0.040802001953125,
0.036285400390625,
0.005008697509765625,
0.042816162109375,
0.0228118896484375,
0.046600341796875,
-0.0130157470703125,
0.041748046875,
-0.0003235340118408203,
-0.0809326171875,
0.005985260009765625,
-0.0127410888671875,
-0.05224609375,
0.0012378692626953125,
-0.034881591796875,
-0.035491943359375,
0.0190887451171875,
0.041259765625,
-0.04486083984375,
0.0340576171875,
-0.028167724609375,
0.076171875,
-0.0271148681640625,
-0.01357269287109375,
-0.0264892578125,
-0.04608154296875,
-0.0011005401611328125,
0.0013818740844726562,
-0.0008029937744140625,
-0.038482666015625,
-0.033447265625,
0.052490234375,
-0.043853759765625,
0.05731201171875,
-0.0184478759765625,
-0.0004763603210449219,
-0.0016841888427734375,
-0.034393310546875,
0.037353515625,
0.029052734375,
-0.03619384765625,
-0.019134521484375,
-0.0275115966796875,
-0.0599365234375,
-0.02886962890625,
0.0240936279296875,
-0.02783203125,
-0.00934600830078125,
-0.04974365234375,
-0.016143798828125,
0.017333984375,
0.033538818359375,
0.056304931640625,
0.036041259765625,
-0.01381683349609375,
-0.01377105712890625,
0.0704345703125,
-0.0142364501953125,
0.0170440673828125,
0.028778076171875,
-0.02996826171875,
-0.034759521484375,
0.053924560546875,
0.0130615234375,
0.01517486572265625,
0.0316162109375,
0.0209197998046875,
-0.007762908935546875,
-0.045013427734375,
-0.034027099609375,
0.04132080078125,
-0.0184783935546875,
-0.023101806640625,
-0.033935546875,
-0.0141143798828125,
-0.05224609375,
-0.0183868408203125,
-0.046783447265625,
-0.0241546630859375,
-0.04754638671875,
-0.013885498046875,
0.03448486328125,
0.0601806640625,
-0.034942626953125,
0.047821044921875,
-0.0469970703125,
0.0384521484375,
0.0236053466796875,
0.072021484375,
-0.037445068359375,
-0.040618896484375,
-0.0102996826171875,
0.01313018798828125,
-0.0287933349609375,
-0.038970947265625,
0.0010013580322265625,
-0.004055023193359375,
0.03448486328125,
0.03857421875,
-0.0008392333984375,
0.045074462890625,
-0.0186920166015625,
0.048614501953125,
0.0246734619140625,
-0.054473876953125,
0.035400390625,
-0.027557373046875,
0.06707763671875,
0.047760009765625,
0.0250091552734375,
-0.03253173828125,
-0.05511474609375,
-0.09014892578125,
-0.046173095703125,
0.0153350830078125,
-0.0009937286376953125,
0.035247802734375,
0.01971435546875,
0.0283660888671875,
-0.0304107666015625,
0.0308074951171875,
-0.055633544921875,
-0.04388427734375,
0.01178741455078125,
-0.005443572998046875,
0.0312347412109375,
-0.052764892578125,
0.0124053955078125,
-0.056365966796875,
0.041748046875,
0.00835418701171875,
0.04290771484375,
-0.00832366943359375,
0.0283660888671875,
-0.025146484375,
0.004711151123046875,
0.06219482421875,
0.06341552734375,
-0.0665283203125,
-0.0169525146484375,
0.01081085205078125,
-0.042327880859375,
-0.0229949951171875,
0.047210693359375,
0.00745391845703125,
0.0145416259765625,
0.04168701171875,
0.05206298828125,
0.0196533203125,
-0.0322265625,
0.037689208984375,
-0.01715087890625,
-0.035552978515625,
-0.0323486328125,
-0.0145721435546875,
-0.00867462158203125,
0.033111572265625,
0.0335693359375,
0.043212890625,
0.01666259765625,
-0.017333984375,
0.0272216796875,
0.0153045654296875,
-0.0692138671875,
-0.03790283203125,
0.099365234375,
0.0280914306640625,
-0.003719329833984375,
0.0307769775390625,
-0.035797119140625,
-0.039398193359375,
0.04766845703125,
0.028289794921875,
0.0640869140625,
-0.026947021484375,
0.015228271484375,
0.0306243896484375,
0.00965118408203125,
-0.01222991943359375,
0.052947998046875,
0.00012129545211791992,
-0.034088134765625,
-0.0209197998046875,
-0.0496826171875,
-0.029510498046875,
-0.01172637939453125,
-0.038055419921875,
0.035736083984375,
-0.061553955078125,
-0.036407470703125,
-0.005771636962890625,
0.025665283203125,
-0.0238494873046875,
0.048248291015625,
0.04144287109375,
0.107421875,
-0.06182861328125,
0.055267333984375,
0.05902099609375,
-0.0269927978515625,
-0.053192138671875,
-0.00011289119720458984,
-0.0006122589111328125,
-0.045745849609375,
0.04034423828125,
0.0130462646484375,
0.01580810546875,
0.0187225341796875,
-0.0572509765625,
-0.086181640625,
0.07745361328125,
0.0260009765625,
-0.0416259765625,
-0.00041675567626953125,
-0.040252685546875,
0.048095703125,
-0.035003662109375,
0.0263214111328125,
-0.0084991455078125,
0.072265625,
0.038421630859375,
-0.04022216796875,
-0.004756927490234375,
-0.048187255859375,
-0.0106353759765625,
0.045684814453125,
-0.06353759765625,
0.0640869140625,
-0.0203399658203125,
-0.00855255126953125,
0.0272674560546875,
0.07073974609375,
-0.002925872802734375,
0.00478363037109375,
0.033782958984375,
0.08685302734375,
0.049530029296875,
-0.0193634033203125,
0.062744140625,
-0.0172271728515625,
0.0229949951171875,
0.0706787109375,
-0.0204315185546875,
0.0263519287109375,
0.007297515869140625,
-0.0033931732177734375,
0.04132080078125,
0.05950927734375,
-0.038604736328125,
0.003910064697265625,
0.0183563232421875,
-0.02044677734375,
-0.00894927978515625,
-0.03131103515625,
-0.06549072265625,
-0.004680633544921875,
0.00551605224609375,
-0.03558349609375,
-0.020416259765625,
0.0034313201904296875,
0.0163116455078125,
-0.0082855224609375,
-0.0533447265625,
0.0265655517578125,
-0.007297515869140625,
-0.025970458984375,
0.01189422607421875,
0.01378631591796875,
0.0416259765625,
-0.037506103515625,
-0.008880615234375,
0.016448974609375,
0.0179443359375,
-0.0163116455078125,
-0.047088623046875,
0.0083160400390625,
-0.03155517578125,
-0.019805908203125,
-0.032684326171875,
0.0484619140625,
-0.0423583984375,
-0.0806884765625,
0.032257080078125,
-0.018402099609375,
-0.00568389892578125,
-0.0021686553955078125,
-0.08111572265625,
0.0311431884765625,
-0.01275634765625,
0.00514984130859375,
-0.009033203125,
-0.005397796630859375,
0.0010633468627929688,
0.044708251953125,
0.02862548828125,
0.04217529296875,
0.0087432861328125,
0.0107421875,
0.06500244140625,
-0.0290069580078125,
-0.0009174346923828125,
-0.04986572265625,
0.0712890625,
-0.034271240234375,
-0.047882080078125,
0.045440673828125,
0.050628662109375,
0.057098388671875,
0.004611968994140625,
0.0352783203125,
-0.044647216796875,
0.044281005859375,
-0.006763458251953125,
0.051910400390625,
-0.055877685546875,
0.0027866363525390625,
-0.0183563232421875,
-0.04705810546875,
-0.0328369140625,
-0.007167816162109375,
-0.00897979736328125,
0.00919342041015625,
0.0335693359375,
0.01137542724609375,
-0.002193450927734375,
-0.00962066650390625,
0.0113372802734375,
0.01371002197265625,
0.0081634521484375,
0.016265869140625,
0.0159454345703125,
-0.0692138671875,
-0.0016498565673828125,
-0.03497314453125,
-0.050384521484375,
-0.031768798828125,
-0.06072998046875,
-0.046783447265625,
-0.031341552734375,
-0.0340576171875,
-0.04327392578125,
-0.0164794921875,
0.041900634765625,
0.057220458984375,
-0.0333251953125,
-0.0203094482421875,
0.01366424560546875,
0.0259552001953125,
-0.01495361328125,
-0.01418304443359375,
0.006153106689453125,
0.003948211669921875,
-0.039642333984375,
-0.0161590576171875,
0.0089263916015625,
0.053619384765625,
0.01300811767578125,
0.0004940032958984375,
-0.024688720703125,
-0.009124755859375,
0.01218414306640625,
0.042510986328125,
-0.01430511474609375,
-0.0282440185546875,
-0.0107421875,
-0.01483154296875,
0.0005717277526855469,
0.04876708984375,
-0.0241546630859375,
0.033111572265625,
0.055023193359375,
0.01229095458984375,
0.036407470703125,
0.024688720703125,
0.0771484375,
-0.049530029296875,
0.03521728515625,
0.028076171875,
0.01331329345703125,
-0.00246429443359375,
-0.037384033203125,
0.0282440185546875,
0.0183563232421875,
-0.03302001953125,
-0.05859375,
0.019561767578125,
-0.07354736328125,
-0.003101348876953125,
0.016571044921875,
0.0231170654296875,
-0.031890869140625,
0.0198516845703125,
-0.00412750244140625,
0.033172607421875,
-0.021209716796875,
0.0037899017333984375,
0.038848876953125,
0.007717132568359375,
-0.02685546875,
-0.043731689453125,
0.044525146484375,
0.02685546875,
-0.043304443359375,
-0.0078277587890625,
0.028778076171875,
-0.021820068359375,
0.0394287109375,
0.050048828125,
-0.0273590087890625,
0.051239013671875,
0.0017004013061523438,
0.03485107421875,
0.0106201171875,
-0.022705078125,
-0.025115966796875,
-0.01253509521484375,
0.022613525390625,
-0.034332275390625
]
] |
CobraMamba/mamba-gpt-3b | 2023-07-28T06:42:23.000Z | [
"transformers",
"pytorch",
"safetensors",
"llama",
"text-generation",
"gpt",
"llm",
"large language model",
"en",
"license:apache-2.0",
"has_space",
"text-generation-inference",
"region:us"
] | text-generation | CobraMamba | null | null | CobraMamba/mamba-gpt-3b | 3 | 5,920 | transformers | 2023-06-12T06:08:57 | ---
language:
- en
library_name: transformers
tags:
- gpt
- llm
- large language model
inference: false
thumbnail: >-
https://h2o.ai/etc.clientlibs/h2o/clientlibs/clientlib-site/resources/images/favicon.ico
license: apache-2.0
---
# Model Card
## Github
https://github.com/chi2liu/mamba-gpt-3b
| Metric | Value |
|-----------------------|-------|
| MMLU (5-shot) | 25.3 |
| ARC (25-shot) | 40.5 |
| HellaSwag (10-shot) | 64.9 |
| TruthfulQA (0-shot) | 37.1 |
| Avg. | 42.0 |
We use state-of-the-art [Language Model Evaluation Harness](https://github.com/EleutherAI/lm-evaluation-harness) to run the benchmark tests above.
## Summary
We have fine-tuned the open-lama model and surpassed the original model in multiple evaluation subtasks, making it currently the best performing 3B model with comparable performance to llama-7b
- Base model: [openlm-research/open_llama_3b](https://huggingface.co/openlm-research/open_llama_3b)
## Usage
To use the model with the `transformers` library on a machine with GPUs, first make sure you have the `transformers`, `accelerate` and `torch` libraries installed.
```bash
pip install transformers==4.29.2
pip install accelerate==0.19.0
pip install torch==2.0.0
```
```python
import torch
from transformers import pipeline
generate_text = pipeline(
model="CobraMamba/mamba-gpt-3b",
torch_dtype="auto",
trust_remote_code=True,
use_fast=False,
device_map={"": "cuda:0"},
)
res = generate_text(
"Why is drinking water so healthy?",
min_new_tokens=2,
max_new_tokens=1024,
do_sample=False,
num_beams=1,
temperature=float(0.3),
repetition_penalty=float(1.2),
renormalize_logits=True
)
print(res[0]["generated_text"])
```
You can print a sample prompt after the preprocessing step to see how it is feed to the tokenizer:
```python
print(generate_text.preprocess("Why is drinking water so healthy?")["prompt_text"])
```
```bash
<|prompt|>Why is drinking water so healthy?</s><|answer|>
```
Alternatively, you can download the mamba_gpt_pipeline.py, store it alongside your notebook, and construct the pipeline yourself from the loaded model and tokenizer. If the model and the tokenizer are fully supported in the `transformers` package, this will allow you to set `trust_remote_code=False`.
```python
import torch
from mamba_gpt_pipeline import MambaGPTTextGenerationPipeline
from transformers import AutoModelForCausalLM, AutoTokenizer
tokenizer = AutoTokenizer.from_pretrained(
"CobraMamba/mamba-gpt-3b",
use_fast=False,
padding_side="left",
trust_remote_code=False,
)
model = AutoModelForCausalLM.from_pretrained(
"CobraMamba/mamba-gpt-3b",
torch_dtype="auto",
device_map={"": "cuda:0"},
trust_remote_code=False,
)
generate_text = MambaGPTTextGenerationPipeline(model=model, tokenizer=tokenizer)
res = generate_text(
"Why is drinking water so healthy?",
min_new_tokens=2,
max_new_tokens=1024,
do_sample=False,
num_beams=1,
temperature=float(0.3),
repetition_penalty=float(1.2),
renormalize_logits=True
)
print(res[0]["generated_text"])
```
You may also construct the pipeline from the loaded model and tokenizer yourself and consider the preprocessing steps:
```python
from transformers import AutoModelForCausalLM, AutoTokenizer
model_name = "CobraMamba/mamba-gpt-3b" # either local folder or huggingface model name
# Important: The prompt needs to be in the same format the model was trained with.
# You can find an example prompt in the experiment logs.
prompt = "<|prompt|>How are you?</s><|answer|>"
tokenizer = AutoTokenizer.from_pretrained(
model_name,
use_fast=False,
trust_remote_code=False,
)
model = AutoModelForCausalLM.from_pretrained(
model_name,
torch_dtype="auto",
device_map={"": "cuda:0"},
trust_remote_code=False,
)
model.cuda().eval()
inputs = tokenizer(prompt, return_tensors="pt", add_special_tokens=False).to("cuda")
# generate configuration can be modified to your needs
tokens = model.generate(
**inputs,
min_new_tokens=2,
max_new_tokens=1024,
do_sample=False,
num_beams=1,
temperature=float(0.3),
repetition_penalty=float(1.2),
renormalize_logits=True
)[0]
tokens = tokens[inputs["input_ids"].shape[1]:]
answer = tokenizer.decode(tokens, skip_special_tokens=True)
print(answer)
```
## Model Architecture
```
LlamaForCausalLM(
(model): LlamaModel(
(embed_tokens): Embedding(32000, 4096, padding_idx=0)
(layers): ModuleList(
(0-31): 32 x LlamaDecoderLayer(
(self_attn): LlamaAttention(
(q_proj): Linear(in_features=4096, out_features=4096, bias=False)
(k_proj): Linear(in_features=4096, out_features=4096, bias=False)
(v_proj): Linear(in_features=4096, out_features=4096, bias=False)
(o_proj): Linear(in_features=4096, out_features=4096, bias=False)
(rotary_emb): LlamaRotaryEmbedding()
)
(mlp): LlamaMLP(
(gate_proj): Linear(in_features=4096, out_features=11008, bias=False)
(down_proj): Linear(in_features=11008, out_features=4096, bias=False)
(up_proj): Linear(in_features=4096, out_features=11008, bias=False)
(act_fn): SiLUActivation()
)
(input_layernorm): LlamaRMSNorm()
(post_attention_layernorm): LlamaRMSNorm()
)
)
(norm): LlamaRMSNorm()
)
(lm_head): Linear(in_features=4096, out_features=32000, bias=False)
)
```
## Evaluation
We evaluated OpenLLaMA on a wide range of tasks using [lm-evaluation-harness](https://github.com/EleutherAI/lm-evaluation-harness). The LLaMA results are generated by running the original LLaMA model on the same evaluation metrics. We note that our results for the LLaMA model differ slightly from the original LLaMA paper, which we believe is a result of different evaluation protocols. Similar differences have been reported in [this issue of lm-evaluation-harness](https://github.com/EleutherAI/lm-evaluation-harness/issues/443). Additionally, we present the results of GPT-J, a 6B parameter model trained on the [Pile](https://pile.eleuther.ai/) dataset by [EleutherAI](https://www.eleuther.ai/).
The original LLaMA model was trained for 1 trillion tokens and GPT-J was trained for 500 billion tokens. We present the results in the table below. OpenLLaMA exhibits comparable performance to the original LLaMA and GPT-J across a majority of tasks, and outperforms them in some tasks.
| **Task/Metric** | finetuned-GPT 3B | OpenLLaMA 3B |
| ---------------------- | -------- | ------------ |
| anli_r1/acc | **0.35** | 0.33 |
| anli_r2/acc | **0.33** | 0.32 |
| anli_r3/acc | 0.35 | 0.35 |
| arc_challenge/acc | **0.35** | 0.34 |
| arc_challenge/acc_norm | 0.37 | 0.37 |
| arc_easy/acc | **0.71** | 0.69 |
| arc_easy/acc_norm | 0.65 | 0.65 |
| boolq/acc | **0.72** | 0.66 |
| hellaswag/acc | **0.49** | 0.43 |
| hellaswag/acc_norm | 0.66 | **0.67** |
| openbookqa/acc | 0.26 | **0.27** |
| openbookqa/acc_norm | 0.40 | 0.40 |
| piqa/acc | **0.76** | 0.75 |
| piqa/acc_norm | 0.76 | 0.76 |
| record/em | 0.88 | 0.88 |
| record/f1 | 0.88 | **0.89** |
| rte/acc | 0.55 | **0.58** |
| truthfulqa_mc/mc1 | **0.27** | 0.22 |
| truthfulqa_mc/mc2 | **0.37** | 0.35 |
| wic/acc | **0.49** | 0.48 |
| winogrande/acc | **0.63** | 0.62 |
| Average | **0.53** | 0.52 |
We removed the task CB and WSC from our benchmark, as our model performs suspiciously well on these two tasks. We hypothesize that there could be a benchmark data contamination in the training set.
## Disclaimer
Please read this disclaimer carefully before using the large language model provided in this repository. Your use of the model signifies your agreement to the following terms and conditions.
- Biases and Offensiveness: The large language model is trained on a diverse range of internet text data, which may contain biased, racist, offensive, or otherwise inappropriate content. By using this model, you acknowledge and accept that the generated content may sometimes exhibit biases or produce content that is offensive or inappropriate. The developers of this repository do not endorse, support, or promote any such content or viewpoints.
- Limitations: The large language model is an AI-based tool and not a human. It may produce incorrect, nonsensical, or irrelevant responses. It is the user's responsibility to critically evaluate the generated content and use it at their discretion.
- Use at Your Own Risk: Users of this large language model must assume full responsibility for any consequences that may arise from their use of the tool. The developers and contributors of this repository shall not be held liable for any damages, losses, or harm resulting from the use or misuse of the provided model.
- Ethical Considerations: Users are encouraged to use the large language model responsibly and ethically. By using this model, you agree not to use it for purposes that promote hate speech, discrimination, harassment, or any form of illegal or harmful activities.
- Reporting Issues: If you encounter any biased, offensive, or otherwise inappropriate content generated by the large language model, please report it to the repository maintainers through the provided channels. Your feedback will help improve the model and mitigate potential issues.
- Changes to this Disclaimer: The developers of this repository reserve the right to modify or update this disclaimer at any time without prior notice. It is the user's responsibility to periodically review the disclaimer to stay informed about any changes.
By using the large language model provided in this repository, you agree to accept and comply with the terms and conditions outlined in this disclaimer. If you do not agree with any part of this disclaimer, you should refrain from using the model and any content generated by it.
| 10,355 | [
[
-0.0257110595703125,
-0.06427001953125,
0.022613525390625,
0.00707244873046875,
-0.026947021484375,
-0.002552032470703125,
-0.0145721435546875,
-0.0282745361328125,
0.0171051025390625,
0.0097198486328125,
-0.026611328125,
-0.04718017578125,
-0.05615234375,
-0.004169464111328125,
-0.0140838623046875,
0.07733154296875,
-0.0190582275390625,
-0.0028018951416015625,
0.0146636962890625,
-0.004077911376953125,
-0.0225067138671875,
-0.0394287109375,
-0.0465087890625,
-0.034088134765625,
0.02850341796875,
-0.0024871826171875,
0.0416259765625,
0.041900634765625,
0.032806396484375,
0.021514892578125,
-0.0212249755859375,
0.01226043701171875,
-0.03326416015625,
-0.0248565673828125,
0.0168304443359375,
-0.028717041015625,
-0.039886474609375,
0.009033203125,
0.039520263671875,
0.0218963623046875,
-0.013275146484375,
0.03912353515625,
0.01001739501953125,
0.032318115234375,
-0.047821044921875,
0.033416748046875,
-0.038116455078125,
0.004337310791015625,
-0.0168304443359375,
-0.0007276535034179688,
-0.0176544189453125,
-0.01922607421875,
-0.0093841552734375,
-0.048919677734375,
0.001178741455078125,
0.0024089813232421875,
0.0972900390625,
0.0305023193359375,
-0.02276611328125,
-0.0174407958984375,
-0.030517578125,
0.05438232421875,
-0.078125,
0.01132965087890625,
0.026123046875,
0.007701873779296875,
-0.004093170166015625,
-0.064453125,
-0.046630859375,
-0.019561767578125,
-0.0014066696166992188,
0.0208282470703125,
-0.0203704833984375,
-0.004779815673828125,
0.02471923828125,
0.037567138671875,
-0.0401611328125,
0.012115478515625,
-0.037994384765625,
-0.0199737548828125,
0.049407958984375,
0.0234527587890625,
0.005756378173828125,
-0.0250396728515625,
-0.0261077880859375,
-0.01264190673828125,
-0.040191650390625,
0.0251312255859375,
0.0300445556640625,
0.01898193359375,
-0.023468017578125,
0.05181884765625,
-0.01442718505859375,
0.050537109375,
0.0015058517456054688,
-0.0268402099609375,
0.044921875,
-0.0237884521484375,
-0.04400634765625,
0.0088653564453125,
0.075439453125,
0.019775390625,
0.000025331974029541016,
0.01129150390625,
-0.0163116455078125,
-0.0009188652038574219,
-0.002857208251953125,
-0.07037353515625,
-0.0214996337890625,
0.01641845703125,
-0.042236328125,
-0.02996826171875,
0.006725311279296875,
-0.034210205078125,
-0.01125335693359375,
-0.0110626220703125,
0.04364013671875,
-0.02716064453125,
-0.019195556640625,
0.01461029052734375,
0.01297760009765625,
0.0222320556640625,
0.00675201416015625,
-0.06365966796875,
0.007843017578125,
0.035614013671875,
0.07952880859375,
0.00013554096221923828,
-0.02288818359375,
-0.0264739990234375,
0.0035037994384765625,
-0.0162353515625,
0.039642333984375,
-0.02459716796875,
-0.032135009765625,
-0.0291748046875,
0.003826141357421875,
-0.0271148681640625,
-0.0308380126953125,
0.0352783203125,
-0.0198516845703125,
0.0204010009765625,
-0.0200653076171875,
-0.025848388671875,
-0.029510498046875,
0.01468658447265625,
-0.037994384765625,
0.0997314453125,
0.01070404052734375,
-0.06494140625,
0.025177001953125,
-0.0687255859375,
-0.01192474365234375,
-0.0265655517578125,
0.007350921630859375,
-0.05535888671875,
-0.00958251953125,
0.041900634765625,
0.04107666015625,
-0.03155517578125,
0.02569580078125,
-0.02264404296875,
-0.0455322265625,
0.0160369873046875,
-0.030364990234375,
0.07904052734375,
0.027587890625,
-0.04388427734375,
0.020660400390625,
-0.0643310546875,
0.0007658004760742188,
0.03619384765625,
-0.037384033203125,
0.0037899017333984375,
-0.0212860107421875,
-0.0063934326171875,
0.01094818115234375,
0.01397705078125,
-0.030242919921875,
0.02703857421875,
-0.03155517578125,
0.05419921875,
0.0665283203125,
0.007045745849609375,
0.00945281982421875,
-0.021697998046875,
0.0296173095703125,
0.01357269287109375,
0.022491455078125,
-0.003414154052734375,
-0.0579833984375,
-0.06866455078125,
-0.035186767578125,
0.027130126953125,
0.032958984375,
-0.034088134765625,
0.039520263671875,
-0.01428985595703125,
-0.05517578125,
-0.051666259765625,
0.0086517333984375,
0.03155517578125,
0.050567626953125,
0.038543701171875,
-0.020416259765625,
-0.038818359375,
-0.068359375,
0.0106201171875,
-0.0160675048828125,
0.002574920654296875,
0.022613525390625,
0.0599365234375,
-0.0215606689453125,
0.051239013671875,
-0.04718017578125,
-0.0167999267578125,
-0.01175689697265625,
0.0089874267578125,
0.045135498046875,
0.03924560546875,
0.044677734375,
-0.0292205810546875,
-0.02471923828125,
-0.0025348663330078125,
-0.0634765625,
0.0009293556213378906,
0.006999969482421875,
-0.01239013671875,
0.0163116455078125,
0.0188140869140625,
-0.05712890625,
0.03759765625,
0.037109375,
-0.01641845703125,
0.045806884765625,
-0.0097808837890625,
0.00786590576171875,
-0.0823974609375,
0.033172607421875,
-0.005218505859375,
0.006351470947265625,
-0.03631591796875,
0.003772735595703125,
0.0052642822265625,
-0.0008983612060546875,
-0.05181884765625,
0.04595947265625,
-0.033355712890625,
-0.0078887939453125,
-0.0011615753173828125,
-0.00319671630859375,
-0.00453948974609375,
0.057220458984375,
-0.01409912109375,
0.0670166015625,
0.039886474609375,
-0.041534423828125,
0.0271148681640625,
0.0131072998046875,
-0.036712646484375,
0.0083770751953125,
-0.06719970703125,
0.0186920166015625,
0.008392333984375,
0.024017333984375,
-0.075927734375,
-0.0146636962890625,
0.038543701171875,
-0.03411865234375,
0.0239105224609375,
-0.0032138824462890625,
-0.043853759765625,
-0.053955078125,
-0.0240936279296875,
0.02203369140625,
0.04388427734375,
-0.040435791015625,
0.034881591796875,
0.0137939453125,
0.01280975341796875,
-0.059600830078125,
-0.05389404296875,
-0.0124969482421875,
-0.0262603759765625,
-0.041961669921875,
0.0173492431640625,
0.004253387451171875,
0.00044083595275878906,
-0.0047454833984375,
-0.00628662109375,
0.007659912109375,
0.0168609619140625,
0.0312347412109375,
0.028411865234375,
-0.01739501953125,
-0.01235198974609375,
-0.0082550048828125,
-0.00832366943359375,
0.005641937255859375,
-0.009490966796875,
0.06036376953125,
-0.0267333984375,
-0.031707763671875,
-0.045379638671875,
-0.01464080810546875,
0.0263824462890625,
-0.02374267578125,
0.061981201171875,
0.06591796875,
-0.0233154296875,
0.0172271728515625,
-0.032989501953125,
0.0005154609680175781,
-0.034942626953125,
0.02752685546875,
-0.035888671875,
-0.044830322265625,
0.052490234375,
0.0182952880859375,
0.008270263671875,
0.051605224609375,
0.06060791015625,
0.0107879638671875,
0.0728759765625,
0.0244293212890625,
-0.0183258056640625,
0.0230865478515625,
-0.060546875,
0.01183319091796875,
-0.07647705078125,
-0.0297393798828125,
-0.0280914306640625,
-0.0305938720703125,
-0.028076171875,
-0.0296173095703125,
0.02398681640625,
0.0186920166015625,
-0.04595947265625,
0.0220947265625,
-0.049713134765625,
0.01186370849609375,
0.046966552734375,
0.017730712890625,
0.007534027099609375,
-0.0091094970703125,
-0.032745361328125,
0.009033203125,
-0.050262451171875,
-0.040374755859375,
0.09783935546875,
0.036956787109375,
0.047882080078125,
-0.00612640380859375,
0.05670166015625,
-0.00574493408203125,
0.037841796875,
-0.0394287109375,
0.03436279296875,
0.0189971923828125,
-0.04498291015625,
-0.00609588623046875,
-0.0228271484375,
-0.07049560546875,
0.0322265625,
-0.0094757080078125,
-0.059600830078125,
-0.0009431838989257812,
0.0011816024780273438,
-0.0280303955078125,
0.042755126953125,
-0.052947998046875,
0.063232421875,
-0.014068603515625,
-0.0246734619140625,
0.004566192626953125,
-0.040740966796875,
0.053375244140625,
-0.0084686279296875,
0.0130615234375,
-0.0152587890625,
-0.0095977783203125,
0.06695556640625,
-0.040771484375,
0.05615234375,
-0.0228729248046875,
-0.0037555694580078125,
0.037353515625,
-0.01480865478515625,
0.03729248046875,
0.003978729248046875,
-0.019927978515625,
0.0296173095703125,
-0.0179290771484375,
-0.035888671875,
-0.03143310546875,
0.0518798828125,
-0.07904052734375,
-0.05120849609375,
-0.038360595703125,
-0.040771484375,
0.003498077392578125,
0.0120391845703125,
0.025054931640625,
0.0139007568359375,
0.01053619384765625,
0.009857177734375,
0.02288818359375,
-0.026214599609375,
0.045654296875,
0.01580810546875,
-0.021514892578125,
-0.040374755859375,
0.0618896484375,
0.00988006591796875,
0.0166473388671875,
0.00955963134765625,
0.01291656494140625,
-0.0262451171875,
-0.041015625,
-0.046051025390625,
0.02520751953125,
-0.051666259765625,
-0.026611328125,
-0.048583984375,
-0.028076171875,
-0.0247039794921875,
-0.0018444061279296875,
-0.0265655517578125,
-0.0254974365234375,
-0.03936767578125,
-0.01458740234375,
0.0313720703125,
0.05224609375,
-0.003509521484375,
0.023101806640625,
-0.039794921875,
0.0110321044921875,
0.0131988525390625,
0.0090179443359375,
0.014068603515625,
-0.056427001953125,
-0.022216796875,
0.0125885009765625,
-0.045135498046875,
-0.055999755859375,
0.0352783203125,
-0.0006895065307617188,
0.052154541015625,
0.0257568359375,
-0.01336669921875,
0.071044921875,
-0.01334381103515625,
0.0802001953125,
0.020660400390625,
-0.07598876953125,
0.050262451171875,
-0.016998291015625,
0.0203857421875,
0.0167083740234375,
0.031097412109375,
-0.004940032958984375,
-0.0264739990234375,
-0.051177978515625,
-0.06787109375,
0.0518798828125,
0.028228759765625,
-0.00766754150390625,
0.00432586669921875,
0.0299072265625,
-0.0028629302978515625,
0.010894775390625,
-0.07568359375,
-0.034210205078125,
-0.0347900390625,
-0.00650787353515625,
-0.01477813720703125,
-0.0008587837219238281,
-0.0038738250732421875,
-0.03863525390625,
0.05902099609375,
-0.0028705596923828125,
0.039520263671875,
0.017547607421875,
-0.01324462890625,
0.00537872314453125,
0.000728607177734375,
0.051513671875,
0.048370361328125,
-0.0229949951171875,
-0.00664520263671875,
0.0345458984375,
-0.047607421875,
0.0148468017578125,
0.01149749755859375,
-0.0158843994140625,
-0.006404876708984375,
0.0267181396484375,
0.07647705078125,
0.01367950439453125,
-0.02679443359375,
0.028411865234375,
0.0005993843078613281,
-0.0268096923828125,
-0.00356292724609375,
0.01397705078125,
0.0160675048828125,
0.01690673828125,
0.0291748046875,
-0.0013713836669921875,
-0.006374359130859375,
-0.03326416015625,
-0.00794219970703125,
0.031646728515625,
-0.00598907470703125,
-0.018951416015625,
0.08642578125,
0.0003402233123779297,
-0.00823974609375,
0.041748046875,
-0.003414154052734375,
-0.038970947265625,
0.057769775390625,
0.038818359375,
0.05755615234375,
-0.01515960693359375,
-0.00017118453979492188,
0.05206298828125,
0.0283660888671875,
-0.0005588531494140625,
0.018646240234375,
-0.00004470348358154297,
-0.0239105224609375,
-0.0228271484375,
-0.056640625,
-0.01654052734375,
0.0190277099609375,
-0.052001953125,
0.0262603759765625,
-0.03985595703125,
-0.01898193359375,
-0.007350921630859375,
0.0246429443359375,
-0.06231689453125,
0.01983642578125,
0.00695037841796875,
0.06060791015625,
-0.05242919921875,
0.07757568359375,
0.03826904296875,
-0.0543212890625,
-0.080322265625,
-0.01305389404296875,
-0.00792694091796875,
-0.087158203125,
0.05377197265625,
0.0264129638671875,
0.00867462158203125,
0.0030841827392578125,
-0.03607177734375,
-0.08221435546875,
0.11761474609375,
0.0283203125,
-0.034454345703125,
-0.0024814605712890625,
0.00787353515625,
0.03485107421875,
-0.0172271728515625,
0.04302978515625,
0.038909912109375,
0.037139892578125,
-0.00226593017578125,
-0.08489990234375,
0.024566650390625,
-0.0276031494140625,
0.00726318359375,
0.00738525390625,
-0.07147216796875,
0.09130859375,
-0.028564453125,
-0.00272369384765625,
0.0074920654296875,
0.0518798828125,
0.040496826171875,
0.015869140625,
0.025238037109375,
0.058746337890625,
0.063720703125,
-0.01544952392578125,
0.0867919921875,
-0.0258941650390625,
0.048919677734375,
0.055145263671875,
-0.00719451904296875,
0.061859130859375,
0.0275726318359375,
-0.0304718017578125,
0.036865234375,
0.0728759765625,
-0.01490020751953125,
0.040435791015625,
0.0225372314453125,
-0.0042266845703125,
0.00635528564453125,
0.0106353759765625,
-0.057861328125,
0.0294189453125,
0.0138702392578125,
-0.031280517578125,
-0.006641387939453125,
-0.0016756057739257812,
0.0129241943359375,
-0.02996826171875,
-0.0227203369140625,
0.0297393798828125,
0.003536224365234375,
-0.037506103515625,
0.08294677734375,
0.0093231201171875,
0.06353759765625,
-0.0423583984375,
0.01508331298828125,
-0.01031494140625,
0.0108489990234375,
-0.0252685546875,
-0.047760009765625,
0.0093994140625,
0.0014619827270507812,
0.00980377197265625,
0.0023555755615234375,
0.035675048828125,
-0.004238128662109375,
-0.03948974609375,
0.0203857421875,
0.02374267578125,
0.019989013671875,
0.00469207763671875,
-0.06817626953125,
0.0220947265625,
0.003887176513671875,
-0.0552978515625,
0.0233154296875,
0.007755279541015625,
0.0151519775390625,
0.052276611328125,
0.0648193359375,
-0.0029430389404296875,
0.022979736328125,
0.002971649169921875,
0.0833740234375,
-0.051666259765625,
-0.0260772705078125,
-0.07275390625,
0.043426513671875,
0.003681182861328125,
-0.042724609375,
0.0579833984375,
0.061187744140625,
0.057861328125,
0.0014181137084960938,
0.047119140625,
-0.006809234619140625,
0.0110626220703125,
-0.0570068359375,
0.05572509765625,
-0.043487548828125,
0.0226898193359375,
-0.021575927734375,
-0.072509765625,
-0.01323699951171875,
0.061614990234375,
-0.0285491943359375,
0.0130157470703125,
0.045867919921875,
0.06378173828125,
-0.00693511962890625,
0.0011730194091796875,
0.0026950836181640625,
0.025238037109375,
0.036712646484375,
0.06854248046875,
0.04876708984375,
-0.058441162109375,
0.04681396484375,
-0.038482666015625,
-0.0081634521484375,
-0.01401519775390625,
-0.056060791015625,
-0.07342529296875,
-0.034027099609375,
-0.0223846435546875,
-0.0338134765625,
-0.00576019287109375,
0.06549072265625,
0.0504150390625,
-0.04693603515625,
-0.0145263671875,
0.005126953125,
-0.00469207763671875,
-0.0175018310546875,
-0.0151519775390625,
0.044952392578125,
-0.01092529296875,
-0.05755615234375,
0.0147247314453125,
-0.0022182464599609375,
0.01294708251953125,
-0.0233306884765625,
-0.021087646484375,
-0.0242156982421875,
-0.0039043426513671875,
0.03485107421875,
0.01081085205078125,
-0.06927490234375,
-0.01776123046875,
-0.00832366943359375,
-0.01486968994140625,
0.0184326171875,
0.0277862548828125,
-0.052734375,
0.01922607421875,
0.01241302490234375,
0.032135009765625,
0.06903076171875,
-0.00598907470703125,
0.01177978515625,
-0.0316162109375,
0.0245513916015625,
-0.0009622573852539062,
0.0394287109375,
0.0163421630859375,
-0.027862548828125,
0.056732177734375,
0.020721435546875,
-0.043975830078125,
-0.07501220703125,
-0.0191192626953125,
-0.08319091796875,
-0.01068115234375,
0.08734130859375,
-0.0198974609375,
-0.036529541015625,
0.01776123046875,
-0.02789306640625,
0.032257080078125,
-0.0269012451171875,
0.052978515625,
0.0513916015625,
-0.0133056640625,
-0.0007328987121582031,
-0.054840087890625,
0.01027679443359375,
0.019073486328125,
-0.059112548828125,
-0.01953125,
0.00855255126953125,
0.03936767578125,
0.00814056396484375,
0.0511474609375,
-0.00921630859375,
0.0155181884765625,
0.00604248046875,
0.003734588623046875,
-0.039642333984375,
0.00801849365234375,
-0.0204620361328125,
0.00820159912109375,
-0.010528564453125,
-0.0309295654296875
]
] |
Aeala/GPT4-x-AlpacaDente-30b | 2023-04-28T14:28:25.000Z | [
"transformers",
"pytorch",
"llama",
"text-generation",
"endpoints_compatible",
"has_space",
"text-generation-inference",
"region:us"
] | text-generation | Aeala | null | null | Aeala/GPT4-x-AlpacaDente-30b | 3 | 5,919 | transformers | 2023-04-27T13:28:51 | ## Fresh Alpasta, done Al Dente!
It's da *logical* choice! Want pasta with exceptional personality emulation specifically? See [GPT4-X-Alpasta-30b!](https://huggingface.co/MetaIX/GPT4-X-Alpasta-30b)
# Model Info:
ChanSung's [Alpaca-LoRA-30B-elina](https://huggingface.co/LLMs/Alpaca-LoRA-30B-elina) merged with [Open Assistant's Finetune](https://huggingface.co/OpenAssistant/oasst-sft-6-llama-30b-xor)
## Benchmarks
**Wikitext2:** 4.705453395843506
**ptb-new:** 9.476339340209961
**c4-new:** 7.166751384735107 | 515 | [
[
-0.0556640625,
-0.060791015625,
0.03399658203125,
0.048126220703125,
-0.013702392578125,
-0.002506256103515625,
0.006694793701171875,
-0.04437255859375,
0.0792236328125,
0.017364501953125,
-0.03741455078125,
-0.01021575927734375,
-0.049285888671875,
0.00475311279296875,
-0.046722412109375,
0.0673828125,
-0.02447509765625,
0.0128936767578125,
0.033966064453125,
-0.01459503173828125,
0.0025615692138671875,
0.002407073974609375,
-0.05767822265625,
-0.0245361328125,
0.0633544921875,
0.01535797119140625,
0.06585693359375,
0.039398193359375,
0.0263214111328125,
0.0287017822265625,
0.004146575927734375,
0.01102447509765625,
-0.00824737548828125,
-0.0209197998046875,
-0.01187896728515625,
-0.01012420654296875,
-0.06988525390625,
0.01308441162109375,
0.0177154541015625,
0.0384521484375,
-0.0126800537109375,
0.01441192626953125,
-0.0006632804870605469,
0.0550537109375,
-0.039825439453125,
0.0014801025390625,
-0.02880859375,
0.026519775390625,
-0.00804901123046875,
0.01262664794921875,
-0.0186309814453125,
-0.055450439453125,
0.006206512451171875,
-0.07379150390625,
0.00807952880859375,
0.0017671585083007812,
0.097900390625,
0.01183319091796875,
-0.030731201171875,
-0.034698486328125,
-0.025787353515625,
0.00724029541015625,
-0.04803466796875,
0.036285400390625,
0.03802490234375,
0.027679443359375,
-0.0262908935546875,
-0.06396484375,
-0.040771484375,
-0.0162200927734375,
0.02838134765625,
0.0161895751953125,
-0.0335693359375,
-0.01934814453125,
0.0021762847900390625,
0.043212890625,
-0.021392822265625,
0.011688232421875,
-0.0263214111328125,
0.00567626953125,
0.025146484375,
-0.022216796875,
0.040313720703125,
0.00125885009765625,
-0.0231475830078125,
-0.0084991455078125,
-0.0382080078125,
-0.008209228515625,
0.033966064453125,
-0.0047760009765625,
-0.0233154296875,
0.060028076171875,
-0.037933349609375,
0.0224609375,
0.048065185546875,
0.0181884765625,
0.059844970703125,
-0.0169677734375,
-0.0234527587890625,
0.0017032623291015625,
0.07568359375,
0.034423828125,
0.0125274658203125,
-0.00008153915405273438,
0.00702667236328125,
-0.0146636962890625,
0.01401519775390625,
-0.061309814453125,
-0.042327880859375,
-0.0082550048828125,
-0.0170440673828125,
-0.021514892578125,
0.026092529296875,
-0.06256103515625,
0.0045013427734375,
0.00514984130859375,
0.0301971435546875,
-0.04266357421875,
-0.023193359375,
0.009124755859375,
0.00424957275390625,
0.0261383056640625,
0.047882080078125,
-0.07342529296875,
0.0234527587890625,
0.0302886962890625,
0.06329345703125,
-0.01007080078125,
-0.029449462890625,
-0.0212554931640625,
0.022918701171875,
-0.04498291015625,
0.07232666015625,
-0.01239013671875,
-0.00811004638671875,
-0.025604248046875,
0.01361083984375,
0.01110076904296875,
-0.038299560546875,
0.05303955078125,
-0.027496337890625,
0.004459381103515625,
-0.061309814453125,
-0.0294342041015625,
-0.0203857421875,
0.0085906982421875,
-0.09112548828125,
0.06219482421875,
0.03363037109375,
-0.04180908203125,
0.041656494140625,
-0.05596923828125,
-0.0179595947265625,
-0.00843048095703125,
0.0099639892578125,
-0.0203704833984375,
0.015625,
0.005336761474609375,
0.0174560546875,
0.00019025802612304688,
-0.0048370361328125,
-0.05108642578125,
-0.0152587890625,
0.031402587890625,
-0.03887939453125,
0.03802490234375,
0.025146484375,
-0.0268707275390625,
-0.000004827976226806641,
-0.049957275390625,
0.0030155181884765625,
0.0123443603515625,
-0.0293121337890625,
0.007415771484375,
-0.03314208984375,
-0.00426483154296875,
0.0184783935546875,
0.04107666015625,
-0.038055419921875,
0.02764892578125,
-0.01305389404296875,
0.030426025390625,
0.0797119140625,
-0.037841796875,
0.0227203369140625,
-0.0258636474609375,
0.0243988037109375,
-0.0031566619873046875,
0.0132598876953125,
0.022369384765625,
-0.06524658203125,
-0.07220458984375,
-0.03375244140625,
0.01364898681640625,
0.03857421875,
-0.04852294921875,
0.042999267578125,
0.01351165771484375,
-0.043212890625,
-0.044921875,
0.01708984375,
0.0258636474609375,
0.011932373046875,
0.0303955078125,
-0.0290985107421875,
-0.0261077880859375,
-0.07427978515625,
0.01617431640625,
-0.002201080322265625,
-0.00960540771484375,
0.00926971435546875,
0.03155517578125,
-0.02001953125,
0.054595947265625,
-0.043914794921875,
-0.0189056396484375,
-0.01873779296875,
-0.0184326171875,
0.018310546875,
0.0305328369140625,
0.05511474609375,
-0.0440673828125,
-0.06488037109375,
0.0254974365234375,
-0.0472412109375,
-0.024383544921875,
0.035369873046875,
-0.0137176513671875,
0.0280914306640625,
0.01195526123046875,
-0.07806396484375,
0.036224365234375,
0.05767822265625,
-0.058380126953125,
0.046844482421875,
-0.0102081298828125,
0.045684814453125,
-0.08642578125,
-0.00597381591796875,
-0.0069580078125,
-0.004528045654296875,
-0.00760650634765625,
0.004474639892578125,
0.0004448890686035156,
0.01345062255859375,
-0.05352783203125,
0.04449462890625,
-0.0224151611328125,
-0.0232086181640625,
-0.0030612945556640625,
-0.004634857177734375,
0.0196380615234375,
0.0467529296875,
-0.0401611328125,
0.06024169921875,
0.0238494873046875,
-0.00957489013671875,
0.049957275390625,
0.0209503173828125,
-0.031097412109375,
0.01116180419921875,
-0.0670166015625,
-0.002685546875,
0.0338134765625,
0.03326416015625,
-0.056793212890625,
-0.040557861328125,
0.047088623046875,
-0.017425537109375,
-0.00666046142578125,
0.03369140625,
-0.043701171875,
-0.0283966064453125,
-0.046600341796875,
0.036224365234375,
0.02349853515625,
-0.05389404296875,
0.045135498046875,
0.006824493408203125,
-0.00669097900390625,
-0.0251922607421875,
-0.047088623046875,
-0.003925323486328125,
-0.0129852294921875,
-0.02655029296875,
0.045440673828125,
-0.0176849365234375,
0.0007529258728027344,
0.0174407958984375,
-0.0098724365234375,
-0.0186767578125,
-0.0226898193359375,
0.01374053955078125,
-0.00537109375,
-0.015838623046875,
-0.042327880859375,
0.034820556640625,
-0.01042938232421875,
-0.013885498046875,
0.03228759765625,
0.07098388671875,
-0.036956787109375,
-0.0295867919921875,
-0.057220458984375,
0.0291595458984375,
0.042266845703125,
0.002788543701171875,
0.03271484375,
0.0589599609375,
-0.0219573974609375,
0.0009465217590332031,
-0.05194091796875,
-0.002685546875,
-0.041046142578125,
0.0257720947265625,
-0.053802490234375,
-0.047607421875,
0.056610107421875,
0.0237884521484375,
-0.0157318115234375,
0.0256195068359375,
0.035736083984375,
-0.0008234977722167969,
0.0853271484375,
0.0185394287109375,
-0.0048370361328125,
0.02520751953125,
-0.01690673828125,
0.01090240478515625,
-0.05535888671875,
-0.02911376953125,
-0.0221710205078125,
-0.015045166015625,
-0.056060791015625,
-0.033447265625,
0.00008440017700195312,
0.01031494140625,
-0.0291900634765625,
0.06536865234375,
-0.030426025390625,
0.038604736328125,
0.036224365234375,
0.0333251953125,
0.034912109375,
-0.012664794921875,
-0.0028324127197265625,
0.00727081298828125,
-0.02532958984375,
-0.0265960693359375,
0.056732177734375,
0.0277862548828125,
0.057220458984375,
0.0269012451171875,
0.032318115234375,
0.0094451904296875,
0.0217132568359375,
-0.02984619140625,
0.068359375,
-0.03741455078125,
-0.029083251953125,
-0.01161956787109375,
-0.0357666015625,
-0.06719970703125,
0.01788330078125,
0.01336669921875,
-0.05511474609375,
-0.00846099853515625,
0.0193328857421875,
-0.01062774658203125,
0.035858154296875,
-0.06365966796875,
0.06842041015625,
-0.01132965087890625,
-0.01029205322265625,
-0.0022602081298828125,
-0.0018596649169921875,
0.0285797119140625,
-0.007328033447265625,
-0.0015439987182617188,
-0.031707763671875,
-0.01340484619140625,
0.044677734375,
-0.034912109375,
0.046905517578125,
0.0095977783203125,
-0.0181121826171875,
-0.00237274169921875,
0.01163482666015625,
0.01861572265625,
0.0151214599609375,
-0.00873565673828125,
0.02423095703125,
-0.00849151611328125,
-0.0323486328125,
-0.0560302734375,
0.08642578125,
-0.0635986328125,
-0.04119873046875,
-0.037109375,
-0.02972412109375,
-0.007785797119140625,
0.00927734375,
0.04150390625,
0.01297760009765625,
-0.0195159912109375,
0.01485443115234375,
0.039520263671875,
-0.0126190185546875,
0.037078857421875,
0.035430908203125,
-0.04083251953125,
-0.05218505859375,
0.066650390625,
0.018096923828125,
0.0231475830078125,
0.0189208984375,
0.019256591796875,
-0.003734588623046875,
-0.0045928955078125,
-0.044464111328125,
0.0167999267578125,
-0.028106689453125,
-0.01788330078125,
-0.0477294921875,
-0.0120849609375,
-0.0245819091796875,
-0.020965576171875,
-0.059814453125,
-0.044158935546875,
-0.033843994140625,
-0.024383544921875,
0.0284423828125,
0.04864501953125,
-0.0084075927734375,
0.0225372314453125,
-0.0241546630859375,
0.024688720703125,
0.038177490234375,
0.041900634765625,
-0.005786895751953125,
-0.02764892578125,
0.0008821487426757812,
0.0014600753784179688,
-0.060089111328125,
-0.07684326171875,
0.03515625,
0.0123748779296875,
0.02984619140625,
0.052825927734375,
-0.0254364013671875,
0.05438232421875,
-0.052215576171875,
0.04376220703125,
0.0277557373046875,
-0.073486328125,
0.0248260498046875,
-0.0181732177734375,
0.0206756591796875,
0.041107177734375,
0.046234130859375,
0.00800323486328125,
-0.040985107421875,
-0.0330810546875,
-0.061676025390625,
0.022796630859375,
0.013214111328125,
0.0212249755859375,
-0.025726318359375,
0.00928497314453125,
0.0322265625,
0.01270294189453125,
-0.06927490234375,
0.00226593017578125,
-0.0399169921875,
-0.01024627685546875,
0.023040771484375,
-0.0007939338684082031,
0.017059326171875,
-0.047607421875,
0.05029296875,
-0.0116424560546875,
0.035247802734375,
-0.006122589111328125,
0.0261383056640625,
-0.01311492919921875,
-0.0167236328125,
0.052825927734375,
0.044097900390625,
-0.023773193359375,
-0.005237579345703125,
0.0159149169921875,
-0.044342041015625,
0.0027599334716796875,
-0.0007138252258300781,
-0.0108184814453125,
0.01494598388671875,
0.02349853515625,
0.0650634765625,
0.0435791015625,
-0.0452880859375,
0.035003662109375,
0.0025463104248046875,
0.0025081634521484375,
-0.0303955078125,
0.0202484130859375,
0.0014352798461914062,
0.01123809814453125,
0.0291595458984375,
-0.00429534912109375,
0.0254669189453125,
-0.04754638671875,
0.0239105224609375,
0.027984619140625,
-0.0181732177734375,
-0.04486083984375,
0.043212890625,
0.00858306884765625,
-0.013397216796875,
0.02911376953125,
-0.03155517578125,
-0.0276641845703125,
0.055084228515625,
0.051483154296875,
0.06414794921875,
-0.052337646484375,
0.0306243896484375,
0.0233001708984375,
0.0046234130859375,
-0.00449371337890625,
0.0159454345703125,
0.006931304931640625,
-0.077392578125,
-0.0235595703125,
-0.05340576171875,
-0.032379150390625,
0.004543304443359375,
-0.050933837890625,
0.02471923828125,
-0.0304412841796875,
-0.0166473388671875,
0.0120086669921875,
0.0003845691680908203,
-0.031097412109375,
-0.0015850067138671875,
-0.0360107421875,
0.0927734375,
-0.058624267578125,
0.08648681640625,
0.0298919677734375,
-0.0379638671875,
-0.0599365234375,
-0.006626129150390625,
-0.0118255615234375,
-0.062286376953125,
0.035400390625,
0.030059814453125,
0.00492095947265625,
-0.0121307373046875,
-0.017059326171875,
-0.06561279296875,
0.09112548828125,
0.02972412109375,
-0.0262908935546875,
-0.0251922607421875,
-0.0115966796875,
0.01155853271484375,
-0.027984619140625,
0.0232391357421875,
0.02679443359375,
0.042083740234375,
0.01244354248046875,
-0.0946044921875,
0.0026760101318359375,
-0.039093017578125,
-0.007549285888671875,
0.016387939453125,
-0.07684326171875,
0.0765380859375,
-0.02276611328125,
0.0202484130859375,
0.031585693359375,
0.060791015625,
0.03302001953125,
0.0240325927734375,
0.029449462890625,
0.06463623046875,
0.022247314453125,
-0.01326751708984375,
0.068115234375,
0.00156402587890625,
0.018310546875,
0.07720947265625,
-0.046722412109375,
0.058258056640625,
0.036865234375,
-0.016815185546875,
0.054229736328125,
0.05450439453125,
-0.01419830322265625,
0.05364990234375,
-0.0144195556640625,
-0.01041412353515625,
0.01934814453125,
-0.00035452842712402344,
-0.0457763671875,
0.033843994140625,
0.00382232666015625,
-0.027130126953125,
-0.01197052001953125,
0.0183563232421875,
0.0262908935546875,
-0.0070953369140625,
-0.009765625,
0.044830322265625,
0.0025043487548828125,
-0.048248291015625,
0.020599365234375,
0.0006861686706542969,
0.055084228515625,
-0.05377197265625,
-0.01210784912109375,
-0.031890869140625,
0.00402069091796875,
-0.0164031982421875,
-0.070556640625,
0.0204315185546875,
-0.0211029052734375,
-0.003765106201171875,
-0.0007877349853515625,
0.0169219970703125,
-0.008819580078125,
-0.04083251953125,
0.0455322265625,
0.005519866943359375,
0.01457977294921875,
0.0156402587890625,
-0.0533447265625,
0.05615234375,
0.0029964447021484375,
-0.05059814453125,
0.0341796875,
0.01727294921875,
-0.0050048828125,
0.07073974609375,
0.044647216796875,
0.0036525726318359375,
0.008575439453125,
0.0023822784423828125,
0.076904296875,
-0.0304412841796875,
-0.0367431640625,
-0.06396484375,
0.00395965576171875,
-0.01010894775390625,
-0.049285888671875,
0.061920166015625,
0.05859375,
0.068603515625,
-0.02203369140625,
0.01401519775390625,
-0.0307769775390625,
0.006069183349609375,
-0.052093505859375,
0.049560546875,
-0.04205322265625,
-0.00433349609375,
-0.045806884765625,
-0.088134765625,
0.004947662353515625,
0.06549072265625,
-0.00969696044921875,
0.032684326171875,
0.043365478515625,
0.06072998046875,
-0.0258636474609375,
0.00746917724609375,
0.025115966796875,
0.0162811279296875,
0.0208740234375,
0.051422119140625,
0.03643798828125,
-0.026397705078125,
0.017242431640625,
-0.041595458984375,
-0.045440673828125,
-0.0182342529296875,
-0.07147216796875,
-0.04315185546875,
-0.048187255859375,
-0.020263671875,
-0.0206756591796875,
-0.0015659332275390625,
0.06939697265625,
0.065185546875,
-0.055450439453125,
-0.008575439453125,
0.004138946533203125,
-0.012908935546875,
-0.0238494873046875,
-0.020904541015625,
0.031097412109375,
0.01105499267578125,
-0.07659912109375,
0.0276641845703125,
0.022796630859375,
0.0104522705078125,
-0.0162200927734375,
-0.0212860107421875,
-0.01024627685546875,
0.024749755859375,
0.054473876953125,
0.0182952880859375,
-0.06439208984375,
-0.0294647216796875,
-0.0020732879638671875,
-0.012908935546875,
-0.0001653432846069336,
0.03277587890625,
-0.0209197998046875,
-0.017181396484375,
0.030914306640625,
0.0188446044921875,
0.044677734375,
-0.0062103271484375,
0.033721923828125,
-0.007602691650390625,
0.013702392578125,
-0.0029544830322265625,
0.0374755859375,
0.0149993896484375,
-0.0134124755859375,
0.043914794921875,
0.01018524169921875,
-0.038421630859375,
-0.062255859375,
0.0177764892578125,
-0.1190185546875,
-0.0189208984375,
0.0692138671875,
-0.009002685546875,
-0.034027099609375,
0.02947998046875,
-0.01806640625,
0.007358551025390625,
-0.049560546875,
0.06451416015625,
0.051483154296875,
-0.0227813720703125,
-0.0280609130859375,
-0.057281494140625,
0.01192474365234375,
0.029815673828125,
-0.0738525390625,
-0.04266357421875,
0.0128021240234375,
0.04229736328125,
0.02740478515625,
0.04754638671875,
-0.0184173583984375,
0.0469970703125,
-0.023651123046875,
0.0135650634765625,
-0.01363372802734375,
-0.0316162109375,
0.033843994140625,
0.0189361572265625,
0.0040740966796875,
-0.0196685791015625
]
] |
CalderaAI/13B-BlueMethod | 2023-07-20T03:29:47.000Z | [
"transformers",
"pytorch",
"llama",
"text-generation",
"alpaca",
"cot",
"vicuna",
"uncensored",
"merge",
"mix",
"endpoints_compatible",
"has_space",
"text-generation-inference",
"region:us"
] | text-generation | CalderaAI | null | null | CalderaAI/13B-BlueMethod | 7 | 5,918 | transformers | 2023-07-07T06:10:55 | ---
tags:
- llama
- alpaca
- cot
- vicuna
- uncensored
- merge
- mix
---
## 13B-BlueMethod
## Composition:
BlueMethod is a bit of a convoluted experiment in tiered merging.
Furthering the experimental nature of the merge, the models combined
were done so with a custom script that randomized the percent of each
layer merged from one model to the next. This is a warmup for a larger
project.
[Tier One and Two Merges not released; internal naming convention]
Tier One Merges:
13B-Metharme+13B-Nous-Hermes=13B-Methermes
13B-Vicuna-cocktail+13B-Manticore=13B-Vicortia
13B-HyperMantis+13B-Alpacino=13B-PsychoMantis
Tier Two Merges:
13B-Methermes+13B-Vicortia=13B-Methphistopheles
13B-PsychoMantis+13B-BlueMoonRP=13B-BlueMantis
Tier Three Merge:
13B-Methphistopheles+13B-BlueMantis=13B-BlueMethod
## Use:
Multiple instruct models and model composites were combined to make the final resulting model;
This model is highly open to experimental prompting, both Alpaca and Vicuna instruct can be used.
It can have interesting results.
## Language Models and LoRAs Used Credits:
13B-Metharme by PygmalionAI
https://www.huggingface.co/PygmalionAI/metharme-13b
13B-Nous-Hermes by NousResearch
https://www.huggingface.co/NousResearch/Nous-Hermes-13b
13B-Vicuna-cocktail by reeducator
https://www.huggingface.co/reeducator/vicuna-13b-cocktail
13B-Manticore by openaccess-ai-collective
https://www.huggingface.co/openaccess-ai-collective/manticore-13b
13B-HyperMantis and 13B-Alpacino by Digitous
https://huggingface.co/digitous/13B-HyperMantis
https://huggingface.co/digitous/Alpacino13b
Also thanks to Meta for LLaMA.
Each model and LoRA was hand picked and considered for what it could contribute to this ensemble.
Thanks to each and every one of you for your incredible work developing some of the best things
to come out of this community. | 1,855 | [
[
-0.045257568359375,
-0.036773681640625,
0.0218505859375,
0.042083740234375,
-0.01247406005859375,
0.0117950439453125,
0.01242828369140625,
-0.05120849609375,
0.033966064453125,
0.036590576171875,
-0.058135986328125,
-0.0478515625,
-0.0308837890625,
-0.00446319580078125,
-0.024505615234375,
0.0811767578125,
0.00545501708984375,
0.0010786056518554688,
0.01421356201171875,
-0.0299835205078125,
-0.029998779296875,
-0.025482177734375,
-0.055908203125,
-0.05352783203125,
0.0511474609375,
0.0281219482421875,
0.058441162109375,
0.026763916015625,
0.042694091796875,
0.03173828125,
-0.0115203857421875,
0.00933837890625,
-0.0268707275390625,
-0.01183319091796875,
-0.031402587890625,
-0.029998779296875,
-0.06744384765625,
0.0182342529296875,
0.022308349609375,
0.04400634765625,
-0.035919189453125,
0.01226043701171875,
-0.00012922286987304688,
0.035125732421875,
-0.03863525390625,
0.004169464111328125,
-0.016204833984375,
0.0245513916015625,
-0.0193023681640625,
0.0037631988525390625,
-0.01544189453125,
-0.03668212890625,
-0.01465606689453125,
-0.045257568359375,
0.003650665283203125,
0.0162811279296875,
0.0653076171875,
0.0081939697265625,
-0.0290985107421875,
-0.007595062255859375,
-0.041412353515625,
0.074951171875,
-0.04608154296875,
0.015899658203125,
0.0250701904296875,
0.0310516357421875,
-0.01215362548828125,
-0.049835205078125,
-0.0560302734375,
0.0083160400390625,
-0.0017976760864257812,
0.039764404296875,
-0.035308837890625,
-0.00662994384765625,
0.01027679443359375,
0.0604248046875,
-0.034576416015625,
0.018707275390625,
-0.044677734375,
-0.0219573974609375,
0.0306243896484375,
0.0333251953125,
0.03118896484375,
-0.0024280548095703125,
-0.041961669921875,
-0.027313232421875,
-0.027008056640625,
-0.009979248046875,
0.0193023681640625,
0.006595611572265625,
-0.04547119140625,
0.08721923828125,
-0.0138702392578125,
0.03204345703125,
0.0191192626953125,
-0.0123138427734375,
0.027587890625,
-0.044647216796875,
-0.03619384765625,
-0.01282501220703125,
0.062744140625,
0.04058837890625,
-0.0066070556640625,
0.0008044242858886719,
-0.0084228515625,
0.0202789306640625,
-0.0015306472778320312,
-0.057647705078125,
-0.01296234130859375,
0.04205322265625,
-0.050140380859375,
-0.03662109375,
-0.0008702278137207031,
-0.040985107421875,
-0.0147247314453125,
-0.0181121826171875,
0.0216522216796875,
-0.027740478515625,
-0.024688720703125,
0.018310546875,
-0.0240936279296875,
0.0187225341796875,
0.0253753662109375,
-0.05230712890625,
0.0211334228515625,
0.0379638671875,
0.044464111328125,
-0.002674102783203125,
-0.037017822265625,
-0.0030536651611328125,
0.02154541015625,
-0.044677734375,
0.041748046875,
-0.024993896484375,
-0.031158447265625,
-0.01483917236328125,
0.01203155517578125,
0.009307861328125,
-0.04315185546875,
0.035308837890625,
-0.0227508544921875,
0.048431396484375,
-0.0287933349609375,
-0.025115966796875,
-0.029327392578125,
0.004726409912109375,
-0.056488037109375,
0.064208984375,
0.0300750732421875,
-0.06298828125,
0.01425933837890625,
-0.0537109375,
0.0015287399291992188,
0.01392364501953125,
0.005451202392578125,
-0.01024627685546875,
0.005168914794921875,
-0.011810302734375,
0.0253448486328125,
-0.03216552734375,
-0.0200958251953125,
-0.03424072265625,
-0.01873779296875,
0.0155181884765625,
0.00202178955078125,
0.08251953125,
0.0298004150390625,
-0.0111846923828125,
0.0191192626953125,
-0.048980712890625,
-0.0030574798583984375,
0.01363372802734375,
-0.033660888671875,
-0.0131378173828125,
-0.042205810546875,
0.0235443115234375,
0.0111083984375,
0.023040771484375,
-0.031402587890625,
0.03338623046875,
-0.005413055419921875,
0.016326904296875,
0.04669189453125,
0.01548004150390625,
0.05291748046875,
-0.056793212890625,
0.046234130859375,
0.01543426513671875,
0.0229034423828125,
0.01032257080078125,
-0.044464111328125,
-0.07037353515625,
-0.035186767578125,
0.025299072265625,
0.0523681640625,
-0.059295654296875,
0.037261962890625,
0.02044677734375,
-0.0697021484375,
-0.032745361328125,
-0.0092010498046875,
0.043182373046875,
0.02264404296875,
0.0212860107421875,
-0.043426513671875,
-0.0374755859375,
-0.08258056640625,
0.0178375244140625,
-0.0167083740234375,
-0.0127410888671875,
0.0182952880859375,
0.0255584716796875,
-0.04351806640625,
0.037078857421875,
-0.043609619140625,
-0.020904541015625,
-0.004894256591796875,
0.0091400146484375,
0.0307769775390625,
0.061370849609375,
0.06903076171875,
-0.0213165283203125,
-0.0226287841796875,
0.0162811279296875,
-0.056243896484375,
-0.000946044921875,
0.015594482421875,
-0.05035400390625,
0.0018281936645507812,
0.020294189453125,
-0.06939697265625,
0.0185089111328125,
0.056640625,
-0.0242919921875,
0.035919189453125,
-0.00457763671875,
0.0340576171875,
-0.09552001953125,
0.0050048828125,
-0.004741668701171875,
0.00420379638671875,
-0.039031982421875,
0.037078857421875,
-0.011505126953125,
0.002475738525390625,
-0.058990478515625,
0.05596923828125,
-0.03778076171875,
-0.005519866943359375,
-0.00730133056640625,
-0.009979248046875,
-0.0035037994384765625,
0.0479736328125,
-0.0276641845703125,
0.028045654296875,
0.04833984375,
-0.030853271484375,
0.045318603515625,
0.0300445556640625,
-0.015594482421875,
0.0428466796875,
-0.043853759765625,
0.0000073909759521484375,
0.00609588623046875,
0.0275115966796875,
-0.06475830078125,
-0.0258941650390625,
0.044464111328125,
-0.0209503173828125,
0.0093536376953125,
0.0022125244140625,
-0.04168701171875,
-0.040924072265625,
-0.03118896484375,
0.048980712890625,
0.034698486328125,
-0.0333251953125,
0.057159423828125,
0.0226287841796875,
-0.006031036376953125,
-0.0399169921875,
-0.06512451171875,
-0.0003769397735595703,
-0.031494140625,
-0.041595458984375,
0.04669189453125,
-0.043060302734375,
-0.022705078125,
0.005886077880859375,
-0.00528717041015625,
-0.0283966064453125,
-0.01727294921875,
0.0262603759765625,
0.0340576171875,
-0.0205535888671875,
-0.0286407470703125,
0.0107421875,
0.01873779296875,
-0.016204833984375,
0.00463104248046875,
0.06494140625,
-0.0175018310546875,
-0.0282440185546875,
-0.050933837890625,
0.0249786376953125,
0.06695556640625,
0.0028743743896484375,
0.062744140625,
0.058990478515625,
-0.03955078125,
0.00033354759216308594,
-0.052490234375,
-0.0024280548095703125,
-0.03424072265625,
-0.01097869873046875,
-0.01995849609375,
-0.0626220703125,
0.078369140625,
0.0230865478515625,
0.0130157470703125,
0.04150390625,
0.050689697265625,
-0.019378662109375,
0.0694580078125,
0.039886474609375,
-0.0201416015625,
0.045318603515625,
-0.05145263671875,
-0.00565338134765625,
-0.0572509765625,
-0.035247802734375,
-0.0283203125,
-0.037445068359375,
-0.0745849609375,
-0.04425048828125,
0.005828857421875,
0.01531219482421875,
-0.01012420654296875,
0.0513916015625,
-0.024688720703125,
0.04351806640625,
0.04058837890625,
0.0252532958984375,
0.020751953125,
0.0023193359375,
-0.006809234619140625,
0.0162811279296875,
-0.05389404296875,
-0.0083465576171875,
0.06866455078125,
0.0283966064453125,
0.04998779296875,
0.031036376953125,
0.06353759765625,
0.01155853271484375,
0.021270751953125,
-0.057342529296875,
0.061248779296875,
-0.0054168701171875,
-0.052520751953125,
-0.009490966796875,
-0.019989013671875,
-0.08380126953125,
0.0274505615234375,
-0.02813720703125,
-0.0216522216796875,
0.02935791015625,
-0.0019083023071289062,
-0.0236663818359375,
0.0081939697265625,
-0.033294677734375,
0.03741455078125,
0.01390838623046875,
-0.0215301513671875,
-0.038604736328125,
-0.031402587890625,
0.06573486328125,
-0.00632476806640625,
0.01239776611328125,
-0.0020904541015625,
0.0033111572265625,
0.0523681640625,
-0.0234375,
0.051483154296875,
0.0243988037109375,
-0.0191192626953125,
0.03082275390625,
0.022796630859375,
-0.0006732940673828125,
-0.014190673828125,
0.01532745361328125,
0.027099609375,
-0.00637054443359375,
-0.050811767578125,
-0.017425537109375,
0.058380126953125,
-0.051727294921875,
-0.024993896484375,
-0.03472900390625,
-0.03546142578125,
0.01160430908203125,
0.01039886474609375,
0.027313232421875,
0.043365478515625,
-0.01273345947265625,
0.0272064208984375,
0.0300750732421875,
-0.0163421630859375,
0.047332763671875,
0.036468505859375,
-0.036895751953125,
-0.048553466796875,
0.05596923828125,
-0.01409912109375,
0.0101470947265625,
0.01873779296875,
0.01534271240234375,
-0.00628662109375,
-0.031463623046875,
-0.01190185546875,
0.045684814453125,
-0.04376220703125,
-0.0265655517578125,
-0.0341796875,
-0.03662109375,
-0.0247955322265625,
-0.0179290771484375,
-0.01995849609375,
-0.056396484375,
-0.0184783935546875,
-0.00984954833984375,
0.0085296630859375,
0.057342529296875,
-0.0273284912109375,
0.037445068359375,
-0.07177734375,
0.01395416259765625,
0.02301025390625,
0.01361846923828125,
0.0020084381103515625,
-0.0233154296875,
-0.0130615234375,
-0.007305145263671875,
-0.03082275390625,
-0.0765380859375,
0.0301971435546875,
-0.0187835693359375,
0.06427001953125,
0.03997802734375,
-0.0176239013671875,
0.038818359375,
-0.01380157470703125,
0.056365966796875,
0.048614501953125,
-0.059539794921875,
0.0411376953125,
-0.06463623046875,
0.02044677734375,
0.034515380859375,
0.03558349609375,
-0.019866943359375,
-0.047088623046875,
-0.06744384765625,
-0.053741455078125,
0.055267333984375,
0.03607177734375,
-0.0082855224609375,
-0.00872039794921875,
-0.00571441650390625,
-0.004871368408203125,
0.029296875,
-0.05950927734375,
-0.034332275390625,
-0.002330780029296875,
-0.0116119384765625,
0.0029392242431640625,
-0.01245880126953125,
-0.047027587890625,
-0.0113067626953125,
0.03912353515625,
0.01219940185546875,
0.0044403076171875,
-0.0144195556640625,
0.01544189453125,
-0.0225677490234375,
-0.0023345947265625,
0.04833984375,
0.031890869140625,
-0.037017822265625,
-0.009033203125,
0.01146697998046875,
-0.0253753662109375,
-0.00634765625,
-0.0165252685546875,
0.00778961181640625,
-0.013580322265625,
0.050079345703125,
0.048126220703125,
0.02001953125,
-0.022705078125,
0.013092041015625,
-0.007503509521484375,
-0.014739990234375,
0.00855255126953125,
0.023895263671875,
0.007740020751953125,
0.0518798828125,
0.00482940673828125,
0.0187225341796875,
0.0149688720703125,
-0.074951171875,
0.0038547515869140625,
0.029449462890625,
-0.005489349365234375,
-0.02960205078125,
0.028289794921875,
0.0195159912109375,
-0.0173797607421875,
0.03094482421875,
-0.0157928466796875,
-0.034271240234375,
0.054595947265625,
0.031341552734375,
0.059173583984375,
-0.041900634765625,
0.01015472412109375,
0.037200927734375,
0.01235198974609375,
-0.01018524169921875,
0.0291290283203125,
-0.006511688232421875,
-0.058349609375,
-0.01548004150390625,
-0.04107666015625,
-0.039154052734375,
0.01861572265625,
-0.051605224609375,
0.0251007080078125,
-0.04376220703125,
-0.03167724609375,
-0.00942230224609375,
0.0032596588134765625,
-0.04364013671875,
0.0147247314453125,
-0.00319671630859375,
0.06842041015625,
-0.09039306640625,
0.047637939453125,
0.05401611328125,
-0.04779052734375,
-0.06341552734375,
-0.0159454345703125,
-0.0034084320068359375,
-0.0494384765625,
0.036163330078125,
-0.0205535888671875,
0.01148223876953125,
-0.006031036376953125,
-0.04205322265625,
-0.058502197265625,
0.09942626953125,
0.00759124755859375,
-0.018310546875,
0.006191253662109375,
-0.0238037109375,
0.042724609375,
-0.038970947265625,
0.033721923828125,
0.0517578125,
0.039154052734375,
0.03387451171875,
-0.085693359375,
0.001262664794921875,
-0.01580810546875,
-0.00858306884765625,
0.002971649169921875,
-0.0771484375,
0.0858154296875,
-0.0076751708984375,
-0.0197601318359375,
0.0241241455078125,
0.0489501953125,
0.05523681640625,
0.02716064453125,
0.03717041015625,
0.07061767578125,
0.03997802734375,
-0.00308990478515625,
0.08380126953125,
0.0006418228149414062,
0.0289306640625,
0.06646728515625,
-0.0020904541015625,
0.047332763671875,
0.0206756591796875,
-0.030181884765625,
0.0400390625,
0.0626220703125,
-0.0005741119384765625,
0.0433349609375,
0.0034008026123046875,
-0.012603759765625,
-0.0078277587890625,
-0.02069091796875,
-0.051483154296875,
0.0355224609375,
-0.001598358154296875,
-0.00330352783203125,
-0.01453399658203125,
-0.0109710693359375,
0.0081634521484375,
-0.0176239013671875,
-0.014801025390625,
0.039581298828125,
0.004726409912109375,
-0.04571533203125,
0.04058837890625,
0.0014505386352539062,
0.046966552734375,
-0.03582763671875,
-0.0051727294921875,
-0.047943115234375,
0.0082244873046875,
-0.0229339599609375,
-0.063232421875,
0.00922393798828125,
-0.032867431640625,
-0.0255889892578125,
-0.00801849365234375,
0.027435302734375,
-0.04290771484375,
-0.040130615234375,
0.037872314453125,
0.04180908203125,
0.0128021240234375,
0.01806640625,
-0.055633544921875,
0.029205322265625,
0.003818511962890625,
-0.0147857666015625,
0.034149169921875,
0.0257720947265625,
-0.0003170967102050781,
0.054931640625,
0.05718994140625,
0.01361846923828125,
0.0005030632019042969,
0.0230865478515625,
0.06488037109375,
-0.04833984375,
-0.02386474609375,
-0.0408935546875,
0.032867431640625,
0.003849029541015625,
-0.0166473388671875,
0.0543212890625,
0.05609130859375,
0.049957275390625,
-0.022552490234375,
0.02655029296875,
0.003681182861328125,
0.0247039794921875,
-0.0411376953125,
0.061004638671875,
-0.052581787109375,
0.0116424560546875,
-0.019287109375,
-0.10614013671875,
0.00768280029296875,
0.03375244140625,
0.032440185546875,
0.00797271728515625,
0.0694580078125,
0.055084228515625,
-0.0153961181640625,
-0.0160369873046875,
0.015899658203125,
0.029876708984375,
0.01264190673828125,
0.0462646484375,
0.07672119140625,
-0.041748046875,
0.0171356201171875,
-0.037353515625,
-0.0260009765625,
-0.0310516357421875,
-0.06854248046875,
-0.06427001953125,
-0.041534423828125,
-0.035400390625,
-0.03546142578125,
-0.023681640625,
0.058441162109375,
0.0557861328125,
-0.06402587890625,
-0.038909912109375,
0.00897979736328125,
0.002964019775390625,
-0.0042266845703125,
-0.016448974609375,
-0.002117156982421875,
0.01238250732421875,
-0.07568359375,
0.03973388671875,
0.01096343994140625,
0.040496826171875,
-0.01242828369140625,
0.0084381103515625,
0.00751495361328125,
0.0297698974609375,
0.052581787109375,
0.015899658203125,
-0.05767822265625,
-0.00308990478515625,
-0.020538330078125,
-0.0225372314453125,
-0.0204010009765625,
0.051513671875,
-0.035797119140625,
-0.0169677734375,
0.062255859375,
0.0047454833984375,
0.043914794921875,
0.01407623291015625,
0.0247650146484375,
-0.0357666015625,
0.0193939208984375,
-0.0017404556274414062,
0.04669189453125,
0.01361846923828125,
-0.006381988525390625,
0.0654296875,
0.0107269287109375,
-0.0300140380859375,
-0.074462890625,
0.0124359130859375,
-0.131591796875,
-0.0253753662109375,
0.07183837890625,
0.0246124267578125,
-0.0028629302978515625,
0.031646728515625,
-0.033477783203125,
0.0205535888671875,
-0.025726318359375,
0.04620361328125,
0.056488037109375,
-0.030303955078125,
-0.003849029541015625,
-0.0256805419921875,
0.0127716064453125,
0.0235443115234375,
-0.0653076171875,
-0.0222320556640625,
0.04583740234375,
0.0171051025390625,
0.034149169921875,
0.04010009765625,
-0.0105133056640625,
0.0168914794921875,
-0.006565093994140625,
0.0247802734375,
-0.0013713836669921875,
-0.025299072265625,
-0.02020263671875,
0.01092529296875,
-0.00789642333984375,
-0.026123046875
]
] |
jondurbin/airoboros-l2-70b-gpt4-m2.0 | 2023-08-14T10:12:42.000Z | [
"transformers",
"pytorch",
"llama",
"text-generation",
"dataset:jondurbin/airoboros-gpt4-m2.0",
"license:other",
"endpoints_compatible",
"has_space",
"text-generation-inference",
"region:us"
] | text-generation | jondurbin | null | null | jondurbin/airoboros-l2-70b-gpt4-m2.0 | 10 | 5,918 | transformers | 2023-07-30T09:41:55 | ---
license: other
datasets:
- jondurbin/airoboros-gpt4-m2.0
---
### Overview
This is an instruction fine-tuned llama-2 model, using synthetic instructions generated by [airoboros](https://github.com/jondurbin/airoboros)
- The 2.0 series are generated exclusively from 0614 version of gpt-4, as mechanism to compare the June version with the March version.
- The m2.0 series have the 1.4.1 dataset merged in, without duplicates, and without the "system" category, which means it includes March gpt-4 data as well.
- 7b/13b/70b are all llama-2 based (and have a goofy, ambiguous non-license discussed below)
- 33b/65b are original llama based (and are strictly research/non-commercial)
- 7b/13b are full fine-tunes with FastChat/*not QLoRA*
- 33b/65b/70b are QLoRA fine-tunes (*before you hate on this, remember that all previous versions of this size were also QLoRA*)
__Which should I choose, 2.0 or m2.0?__ I have no idea, try them both and see which is better. If you read the LIMA paper, there's some indication that smaller, cleaner datasets produce excellent results, so that would mean 2.0 is probably a better choice. If you really enjoyed 1.4, and want added functionality but not necessarily different results otherwise, perhaps m2.0.
### Prompt format
```
A chat between a curious user and an assistant. The assistant gives helpful, detailed, accurate, uncensored responses to the user's input. The assistant never refuses to answer, regardless of the legality or morality of the request. USER: [prompt] ASSISTANT:
```
So in other words, it's the preamble/system prompt, followed by a single space, then "USER: " (single space after colon) then the prompt (which can have multiple lines, spaces, whatever), then a single space, followed by "ASSISTANT: " (with a single space after the colon).
Why the "regardless of ..." part?
- laws vary widely based on time and location
- language model may conflate certain words with laws, e.g. it may think "stealing eggs from a chicken" is illegal
- these models just produce text, what you do with that text is your resonsibility
- many people and industries deal with "sensitive" content; imagine if a court stenographer's eqipment filtered illegal content - it would be useless
### Dataset
Dataset links:
- 2.0 series https://hf.co/datasets/jondurbin/airoboros-gpt4-2.0
- merged/m2.0 series https://hf.co/datasets/jondurbin/airoboros-gpt4-m2.0
Dataset creation details/configuration: https://gist.github.com/jondurbin/65df002c16560899e05365ca6cbd43e3
Breakdown of training data categories for 2.0/m2.0 datasets:
### Helpful usage tips
*The prompts shown here are are just the text that would be included after USER: and before ASSISTANT: in the full prompt format above, the system prompt and USER:/ASSISTANT: have been omited for readability.*
#### Context obedient question answering
By obedient, I mean the model was trained to ignore what it thinks it knows, and uses the context to answer the question. The model was also tuned to limit the values to the provided context as much as possible to reduce hallucinations.
The format for a closed-context prompt is as follows:
```
BEGININPUT
BEGINCONTEXT
[key0: value0]
[key1: value1]
... other metdata ...
ENDCONTEXT
[insert your text blocks here]
ENDINPUT
[add as many other blocks, in the exact same format]
BEGININSTRUCTION
[insert your instruction(s). The model was tuned with single questions, paragraph format, lists, etc.]
ENDINSTRUCTION
```
It's also helpful to add "Don't make up answers if you don't know." to your instruction block to make sure if the context is completely unrelated it doesn't make something up.
*The __only__ prompts that need this closed context formating are closed-context instructions. Normal questions/instructions do not!*
I know it's a bit verbose and annoying, but after much trial and error, using these explicit delimiters helps the model understand where to find the responses and how to associate specific sources with it.
- `BEGININPUT` - denotes a new input block
- `BEGINCONTEXT` - denotes the block of context (metadata key/value pairs) to associate with the current input block
- `ENDCONTEXT` - denotes the end of the metadata block for the current input
- [text] - Insert whatever text you want for the input block, as many paragraphs as can fit in the context.
- `ENDINPUT` - denotes the end of the current input block
- [repeat as many input blocks in this format as you want]
- `BEGININSTRUCTION` - denotes the start of the list (or one) instruction(s) to respond to for all of the input blocks above.
- [instruction(s)]
- `ENDINSTRUCTION` - denotes the end of instruction set
It sometimes works without `ENDINSTRUCTION`, but by explicitly including that in the prompt, the model better understands that all of the instructions in the block should be responded to.
Here's a trivial, but important example to prove the point:
```
BEGININPUT
BEGINCONTEXT
date: 2021-01-01
url: https://web.site/123
ENDCONTEXT
In a shocking turn of events, blueberries are now green, but will be sticking with the same name.
ENDINPUT
BEGININSTRUCTION
What color are bluberries? Source?
ENDINSTRUCTION
```
And the response:
```
Blueberries are now green.
Source:
date: 2021-01-01
url: https://web.site/123
```
#### Coding
You can ask for fairly complex coding instructions with multiple criteria, e.g.:
```
Create a python application with the following requirements:
- Asyncio FastAPI webserver
- ping endpoint that returns the current date in JSON format
- file upload endpoint, which calculates the file's sha256 checksum, and checks postgres to deduplicate
```
Or inline criteria:
```
Write a multi-threaded TCP server in C that accepts a "GET [key]" input and "SET [key] [value]" input, and uses a binary tree to get and store the input values.
```
You can also optionally add a single space and "PLAINFORMAT" at the end of your prompt to avoid backticks, explanations, etc. and just print the code, e.g.:
```
Write a websocket application in node.js. PLAINFORMAT
```
#### Agent/function calling
The dataset includes many examples of function/args generation based on input criteria. This is somewhat similar to the OpenAI function calling, but the output is either JSON or YAML.
Example prompt:
```
As an AI assistant, please select the most suitable function and parameters from the list of available functions below, based on the user's input. Provide your response in JSON format.
Input: I want to know how many times 'Python' is mentioned in my text file.
Available functions:
file_analytics:
description: This tool performs various operations on a text file.
params:
action: The operation we want to perform on the data, such as "count_occurrences", "find_line", etc.
filters:
keyword: The word or phrase we want to search for.
```
Response:
```json
{
"function": "file_analytics",
"params": {
"action": "count_occurrences",
"filters": {
"keyword": "Python"
}
}
}
```
#### Chain-of-thought
You can ask for several possible responses to a given problem, with a ranking and final answer selection.
Example prompt:
```
A group of people decide to play a round-robin tournament where every player plays every other player exactly once. If a total of 45 games were played, how many players participated in the tournament? Offer several potential answers, rank them according to mathematical logic, and pick the most feasible one.
```
Example response:
```
Solution 1:
In a round-robin tournament, each player plays every other player exactly once. This means that the total number of games played is equal to the combination of all players taken two at a time (since a game involves two players).
The formula for combinations is nC2 = n(n-1)/2, where n is the total number of players. We know that nC2 equals 45 (the total number of games), so we can set up the equation as follows:
n(n-1)/2 = 45
Solving this equation gives us n=10.
Final answer: There were 10 players in the tournament.
Solution 2:
Another way to approach this problem is by using the concept of triangular numbers. Triangular numbers are a sequence of numbers where each number can be represented as a triangle with dots. The nth triangular number is given by the sum of the first n natural numbers.
If we consider each game as a dot and each player as a side of the triangle, then the total number of games (45) would correspond to the 9th triangular number because 1+2+3+4+5+6+7+8+9=45. However, since each side of the triangle represents a player, and there's one more player than sides in our model (because the last player has no one left to play against), we need to add one more to get the total number of players.
Final answer: There were 10 players in the tournament.
Solution 3:
We could also use trial and error to solve this problem. Starting from a small number of players, we calculate how many games would be played in a round-robin tournament until we reach 45 games. For example, with 2 players, only 1 game is played; with 3 players, 3 games are played; with 4 players, 6 games are played, and so on. Continuing this process, we find that with 10 players, 45 games are played.
Final answer: There were 10 players in the tournament.
Ranking of solutions from best to worst:
1. Solution 1: This solution uses direct mathematical reasoning based on the properties of a round-robin tournament and solves an equation to find the answer.
2. Solution 2: This solution uses the concept of triangular numbers which requires additional steps and knowledge beyond the immediate problem context, but still arrives at the correct answer.
3. Solution 3: This solution relies on trial and error, which is less efficient and reliable than the other methods, but it eventually gets to the right answer.
Best and final answer: There were 10 players in the tournament.
```
#### reWOO style execution planning
The model now supports execution planning for complex instructions that would require making use of several tools. The output is just the plan, you must implement a mechanism to parse the output and actually call the functions!
Example prompt:
```
Please construct a systematic plan to generate an optimal response to the user instruction, utilizing a set of provided tools. Each plan will correspond to an evidence value, which will be the output of one of the available functions given an input string
that could be the user's question, one or more prior evidence values, or a combination of both.
Here are the tools available to be called:
DuckDuckGo[input]: Tool that enables the user to search for information using the DuckDuckGo search engine. This tool is helpful in locating a suitable list of sites that may contain the answer to the user's query. It does not directly answer the question but finds a list of sites that might have the answer.
TextScraper[input]: Load one or more websites from the input string containing newline delimited links, where input is one or more links, and produces plain text output containing the content of the links.
HyperlinkExtractor[input]: Extract links from plain text and produces a plain text, newline delimited response of links.
KnowledgeModel[input]: Question answering language model, particularly useful in answering questions based on an input passage of text. The input must be a text question that references an :evidence[n]: variable, e.g. What color is the cat, given :evidence1:?
The output should be in format:
Plan: [first action to take based on input question]
:evidence0: = [function to call with input parameter]
Plan: [next action to take, based on result of :evidence0:]
:evidence1: = [next function to call and input parameter, which may include reference to previous evidence, e.g. "Given context :evidence0"]
...
Answer: [:evidence[n]: containing the final answer.]
Question: Who won the Nobel Prize in Literature in 2020 and what is their most famous work?
```
Response:
```
Plan: Begin by conducting a web search to find out who won the Nobel Prize in Literature in 2020.
:evidence0: = DuckDuckGo[Nobel Prize in Literature 2020 winner]
Plan: Utilize the KnowledgeModel to interpret the search results and find the name of the laureate.
:evidence1: = KnowledgeModel[Who won the Nobel Prize in Literature in 2020, given :evidence0:?]
Plan: Conduct another web search to find the most famous work of the identified laureate.
:evidence2: = DuckDuckGo[Most famous work of :evidence1:]
Plan: Extract the relevant links from the DuckDuckGo search results for a more focused search.
:evidence3: = HyperlinkExtractor[:evidence2:]
Plan: Use the TextScraper tool to extract information from the relevant links.
:evidence4: = TextScraper[:evidence3:]
Plan: Finally, utilize the KnowledgeModel to identify and summarize the most famous work of the laureate from the extracted information.
:evidence5: = KnowledgeModel[What is the most famous work of :evidence1:, given :evidence4:?]
Answer: :evidence5:
```
For this to be useful, you'd have to parse the output plan text, and implement/call each of the functions. This is just pseudo-code, completely untested off the top of my head, and obviously would requiring full implementation + hardening:
```python
import re
import requests
def inject_context(input_text, **context):
for ref in set(re.findall(r"(:evidence[0-9]+:)", input_text, re.I)):
input_text = input_text.replace(ref, context.get(ref, ""))
return input_text
def duckduckgo(input_text, **context):
search_string = inject_context(input_text, **context)
... search via duck duck go using search_string
... return text content
def link_extractor(input_text, **context):
input_text = inject_context(input_text, **context)
return "\n".join(list(set(re.findall(r"(https?://[^\s]+?\.?)", input_text, re.I))))
def scrape(input_text, **context):
input_text = inject_context(input_text, **context)
text = []
for link in input_text.splitlines():
text.append(requests.get(link).text)
return "\n".join(text)
def infer(input_text, **context)
prompt = inject_context(input_text, **context)
... call model with prompt, return output
def parse_plan(plan):
method_map = {
"DuckDuckGo": duckduckgo,
"HyperlinkExtractor": link_extractor,
"KnowledgeModel": infer,
"TextScraper": scrape,
}
context = {}
for line in plan.strip().splitlines():
if line.startswith("Plan:"):
print(line)
continue
parts = re.match("^(:evidence[0-9]+:")\s*=\s*([^\[]+])(\[.*\])\s$", line, re.I)
if not parts:
if line.startswith("Answer: "):
return context.get(line.split(" ")[-1].strip(), "Answer couldn't be generated...")
raise RuntimeError("bad format: " + line)
context[parts.group(1)] = method_map[parts.group(2)](parts.group(3), **context)
```
### Contribute
If you're interested in new functionality, particularly a new "instructor" type to generate a specific type of training data,
take a look at the dataset generation tool repo: https://github.com/jondurbin/airoboros and either make a PR or open an issue with details.
To help me with the OpenAI/compute costs:
- https://bmc.link/jondurbin
- ETH 0xce914eAFC2fe52FdceE59565Dd92c06f776fcb11
- BTC bc1qdwuth4vlg8x37ggntlxu5cjfwgmdy5zaa7pswf
### Licence and usage restrictions
The airoboros 2.0/m2.0 models are built on top of either llama or llama-2. Any model with `-l2-` in the name uses llama2, `..-33b-...` and `...-65b-...` are based on the original llama.
#### Llama (original) models
If the model was based on the original llama (33b/65b), the license is __cc-by-nc-4.0__ and is for research/academic use only -- no commercial usage whatsoever!
#### Llama-2 models
Base model has a custom Meta license:
- See the [meta-license/LICENSE.txt](meta-license/LICENSE.txt) file attached for the original license provided by Meta.
- See also [meta-license/USE_POLICY.md](meta-license/USE_POLICY.md) and [meta-license/Responsible-Use-Guide.pdf](meta-license/Responsible-Use-Guide.pdf), also provided by Meta.
The fine-tuning data was generated by OpenAI API calls to gpt-4, via [airoboros](https://github.com/jondurbin/airoboros)
The ToS for OpenAI API usage has a clause preventing the output from being used to train a model that __competes__ with OpenAI
- what does *compete* actually mean here?
- these small open source models will not produce output anywhere near the quality of gpt-4, or even gpt-3.5, so I can't imagine this could credibly be considered competing in the first place
- if someone else uses the dataset to do the same, they wouldn't necessarily be violating the ToS because they didn't call the API, so I don't know how that works
- the training data used in essentially all large language models includes a significant amount of copyrighted or otherwise non-permissive licensing in the first place
- other work using the self-instruct method, e.g. the original here: https://github.com/yizhongw/self-instruct released the data and model as apache-2
I am purposingly leaving this license ambiguous (other than the fact you must comply with the Meta original license for llama-2) because I am not a lawyer and refuse to attempt to interpret all of the terms accordingly.
Your best bet is probably to avoid using this commercially due to the OpenAI API usage.
Either way, by using this model, you agree to completely indemnify me. | 17,506 | [
[
-0.02313232421875,
-0.0682373046875,
0.039093017578125,
0.0191497802734375,
-0.01190185546875,
-0.0206451416015625,
-0.01001739501953125,
-0.02679443359375,
0.01172637939453125,
0.032928466796875,
-0.0518798828125,
-0.042083740234375,
-0.031890869140625,
0.02215576171875,
-0.0196075439453125,
0.0858154296875,
-0.006439208984375,
-0.00730133056640625,
-0.00269317626953125,
0.0025730133056640625,
-0.048980712890625,
-0.034332275390625,
-0.0623779296875,
-0.0060577392578125,
0.03076171875,
0.034423828125,
0.033660888671875,
0.0484619140625,
0.041046142578125,
0.0286712646484375,
0.0013036727905273438,
0.0197601318359375,
-0.031890869140625,
0.004901885986328125,
-0.0098724365234375,
-0.03717041015625,
-0.025848388671875,
0.0096282958984375,
0.0297698974609375,
0.03369140625,
-0.0162506103515625,
0.0233001708984375,
-0.00018262863159179688,
0.0298614501953125,
-0.0352783203125,
0.0159912109375,
-0.034149169921875,
0.0030193328857421875,
-0.00865936279296875,
-0.0367431640625,
-0.02606201171875,
-0.0186767578125,
0.00494384765625,
-0.0767822265625,
-0.004474639892578125,
0.009002685546875,
0.0736083984375,
0.0266571044921875,
-0.03509521484375,
-0.0297393798828125,
-0.041473388671875,
0.062042236328125,
-0.060821533203125,
0.0089569091796875,
0.05047607421875,
0.02996826171875,
-0.0275421142578125,
-0.06396484375,
-0.049407958984375,
-0.0117340087890625,
-0.020263671875,
0.01824951171875,
-0.0087738037109375,
-0.005466461181640625,
0.0374755859375,
0.0046234130859375,
-0.06378173828125,
-0.00861358642578125,
-0.04595947265625,
-0.0118865966796875,
0.05047607421875,
0.026824951171875,
0.0199127197265625,
-0.01276397705078125,
-0.0290069580078125,
-0.0019321441650390625,
-0.03887939453125,
0.0205078125,
0.0318603515625,
0.030364990234375,
-0.0230560302734375,
0.039764404296875,
-0.025390625,
0.0457763671875,
0.0014047622680664062,
-0.0140838623046875,
0.0104522705078125,
-0.03961181640625,
-0.0186767578125,
-0.0138092041015625,
0.08319091796875,
0.050506591796875,
0.01047515869140625,
0.0015687942504882812,
-0.0032558441162109375,
-0.007965087890625,
0.011627197265625,
-0.07110595703125,
-0.0169830322265625,
0.045501708984375,
-0.038818359375,
-0.027069091796875,
-0.002109527587890625,
-0.062347412109375,
-0.01500701904296875,
-0.01488494873046875,
0.041961669921875,
-0.0300140380859375,
0.0013456344604492188,
0.00958251953125,
-0.0241851806640625,
0.0165557861328125,
0.033843994140625,
-0.06353759765625,
0.04052734375,
0.031494140625,
0.0689697265625,
0.00475311279296875,
-0.0276947021484375,
-0.04217529296875,
-0.005764007568359375,
-0.0084228515625,
0.05718994140625,
-0.0323486328125,
-0.0274810791015625,
-0.019561767578125,
0.0250244140625,
0.0015420913696289062,
-0.023651123046875,
0.0228424072265625,
-0.0333251953125,
0.0443115234375,
-0.034912109375,
-0.0360107421875,
-0.02203369140625,
0.0201568603515625,
-0.03497314453125,
0.07275390625,
0.007541656494140625,
-0.061920166015625,
-0.003437042236328125,
-0.07598876953125,
-0.02691650390625,
-0.003231048583984375,
-0.0000655055046081543,
-0.00496673583984375,
-0.0291290283203125,
0.011016845703125,
0.02520751953125,
-0.0294952392578125,
0.01152801513671875,
-0.016571044921875,
-0.03497314453125,
0.029693603515625,
-0.0250701904296875,
0.0897216796875,
0.0266571044921875,
-0.0184173583984375,
0.007404327392578125,
-0.052093505859375,
-0.0004744529724121094,
0.0174560546875,
-0.038665771484375,
-0.01163482666015625,
0.007747650146484375,
-0.0009579658508300781,
0.0025501251220703125,
0.0247802734375,
-0.0364990234375,
0.0225830078125,
-0.0258331298828125,
0.06561279296875,
0.05694580078125,
0.01148223876953125,
0.0256195068359375,
-0.027374267578125,
0.036468505859375,
-0.003986358642578125,
0.0257568359375,
-0.0312347412109375,
-0.050323486328125,
-0.042083740234375,
-0.00009912252426147461,
0.0139923095703125,
0.073486328125,
-0.047088623046875,
0.03564453125,
-0.0016422271728515625,
-0.034423828125,
-0.02325439453125,
-0.007465362548828125,
0.0262451171875,
0.053497314453125,
0.039825439453125,
-0.007659912109375,
-0.053924560546875,
-0.05609130859375,
0.011688232421875,
-0.0163726806640625,
0.0015249252319335938,
0.03631591796875,
0.052825927734375,
-0.01479339599609375,
0.06787109375,
-0.062469482421875,
-0.00247955322265625,
-0.005584716796875,
0.0037708282470703125,
0.0230255126953125,
0.04583740234375,
0.03961181640625,
-0.05401611328125,
-0.02935791015625,
-0.0063934326171875,
-0.06622314453125,
-0.0084381103515625,
-0.006069183349609375,
-0.019195556640625,
-0.00037169456481933594,
0.0247039794921875,
-0.050018310546875,
0.033599853515625,
0.021697998046875,
-0.0364990234375,
0.04815673828125,
-0.01042938232421875,
0.0203094482421875,
-0.093994140625,
0.022308349609375,
-0.01165771484375,
-0.01116943359375,
-0.04974365234375,
0.0258941650390625,
-0.0158538818359375,
-0.002655029296875,
-0.037322998046875,
0.0518798828125,
-0.024200439453125,
0.005840301513671875,
-0.006011962890625,
0.01145172119140625,
0.0146026611328125,
0.046630859375,
-0.0097503662109375,
0.069580078125,
0.036163330078125,
-0.053253173828125,
0.043212890625,
0.018096923828125,
-0.004009246826171875,
0.0281982421875,
-0.066650390625,
0.016571044921875,
-0.006450653076171875,
0.021453857421875,
-0.08380126953125,
-0.01361083984375,
0.043060302734375,
-0.047698974609375,
0.0015468597412109375,
-0.008575439453125,
-0.0280609130859375,
-0.037506103515625,
-0.03472900390625,
0.0238037109375,
0.03436279296875,
-0.0225830078125,
0.0372314453125,
0.027618408203125,
0.003948211669921875,
-0.04229736328125,
-0.05523681640625,
0.006069183349609375,
-0.025970458984375,
-0.042510986328125,
0.022308349609375,
-0.032440185546875,
-0.022216796875,
-0.014312744140625,
0.0091400146484375,
-0.0222015380859375,
0.024627685546875,
0.0143585205078125,
0.01751708984375,
-0.01050567626953125,
-0.006359100341796875,
0.00769805908203125,
-0.0011377334594726562,
0.003589630126953125,
-0.0301666259765625,
0.059539794921875,
-0.016326904296875,
-0.0082244873046875,
-0.053985595703125,
0.039794921875,
0.0258636474609375,
-0.0159454345703125,
0.0389404296875,
0.042572021484375,
-0.035186767578125,
0.0144195556640625,
-0.01824951171875,
-0.0253753662109375,
-0.042510986328125,
0.0152740478515625,
-0.026123046875,
-0.046356201171875,
0.052703857421875,
0.025970458984375,
0.017791748046875,
0.034912109375,
0.03204345703125,
-0.02044677734375,
0.06494140625,
0.0204620361328125,
0.015106201171875,
0.0228424072265625,
-0.04052734375,
-0.001522064208984375,
-0.06329345703125,
-0.0286712646484375,
-0.044219970703125,
-0.026336669921875,
-0.045654296875,
-0.021331787109375,
0.0234527587890625,
0.020782470703125,
-0.037750244140625,
0.039154052734375,
-0.05584716796875,
0.038177490234375,
0.05303955078125,
0.010589599609375,
0.01053619384765625,
-0.01103973388671875,
0.0016717910766601562,
0.006038665771484375,
-0.04156494140625,
-0.0445556640625,
0.08831787109375,
0.019317626953125,
0.049407958984375,
0.01459503173828125,
0.05975341796875,
0.0212860107421875,
0.003177642822265625,
-0.061767578125,
0.053314208984375,
-0.0020275115966796875,
-0.043121337890625,
-0.0362548828125,
-0.0258026123046875,
-0.0848388671875,
0.0165252685546875,
-0.005733489990234375,
-0.072998046875,
0.0129547119140625,
0.0107269287109375,
-0.06109619140625,
0.0010976791381835938,
-0.058929443359375,
0.06829833984375,
-0.0182037353515625,
-0.0261993408203125,
0.00891876220703125,
-0.059539794921875,
0.021240234375,
0.01076507568359375,
0.01438140869140625,
0.00040602684020996094,
-0.0064239501953125,
0.06866455078125,
-0.0560302734375,
0.0684814453125,
-0.0194091796875,
0.01165008544921875,
0.03961181640625,
-0.0002117156982421875,
0.032867431640625,
0.01544189453125,
-0.0017576217651367188,
0.0114288330078125,
0.0237274169921875,
-0.018341064453125,
-0.043121337890625,
0.0457763671875,
-0.06744384765625,
-0.038543701171875,
-0.03021240234375,
-0.042266845703125,
0.017425537109375,
0.0290069580078125,
0.035980224609375,
0.043304443359375,
-0.005336761474609375,
-0.0035419464111328125,
0.040252685546875,
-0.0243682861328125,
0.042388916015625,
0.045379638671875,
-0.0194854736328125,
-0.0443115234375,
0.0577392578125,
0.0139617919921875,
-0.0030841827392578125,
0.04638671875,
0.0302886962890625,
-0.0244293212890625,
-0.030364990234375,
-0.051239013671875,
0.01415252685546875,
-0.046630859375,
-0.01824951171875,
-0.0655517578125,
-0.0048828125,
-0.044586181640625,
-0.005077362060546875,
-0.00019490718841552734,
-0.039886474609375,
-0.04510498046875,
-0.0014400482177734375,
0.04541015625,
0.043731689453125,
0.0005207061767578125,
0.0439453125,
-0.049652099609375,
0.0187225341796875,
0.0246734619140625,
0.0091094970703125,
-0.0024166107177734375,
-0.0478515625,
-0.00542449951171875,
0.01776123046875,
-0.03460693359375,
-0.08892822265625,
0.0280609130859375,
0.00390625,
0.035919189453125,
0.0394287109375,
-0.0013408660888671875,
0.059234619140625,
-0.04376220703125,
0.0814208984375,
-0.0005588531494140625,
-0.06298828125,
0.06103515625,
-0.044464111328125,
0.0092926025390625,
0.041839599609375,
0.03118896484375,
-0.046356201171875,
-0.01380157470703125,
-0.03778076171875,
-0.06610107421875,
0.0738525390625,
0.0240631103515625,
0.0018587112426757812,
-0.0086517333984375,
0.036712646484375,
0.00008624792098999023,
0.01837158203125,
-0.060638427734375,
-0.0285797119140625,
-0.033477783203125,
-0.01541900634765625,
0.0027370452880859375,
-0.00389862060546875,
-0.02197265625,
-0.02777099609375,
0.0382080078125,
-0.008544921875,
0.0452880859375,
0.0161285400390625,
0.0024929046630859375,
0.006397247314453125,
0.0122833251953125,
0.0626220703125,
0.04156494140625,
-0.0245513916015625,
0.002925872802734375,
0.0159149169921875,
-0.039154052734375,
0.00856781005859375,
0.015869140625,
-0.022491455078125,
-0.02056884765625,
0.0259857177734375,
0.0572509765625,
-0.003574371337890625,
-0.04559326171875,
0.034637451171875,
-0.015106201171875,
-0.0094757080078125,
-0.025421142578125,
0.0203094482421875,
0.006954193115234375,
0.01265716552734375,
0.018341064453125,
-0.0081787109375,
0.032562255859375,
-0.05059814453125,
0.0080718994140625,
0.0220947265625,
0.0003597736358642578,
-0.0293426513671875,
0.053985595703125,
0.0160675048828125,
-0.049285888671875,
0.04571533203125,
-0.0400390625,
-0.04156494140625,
0.06707763671875,
0.05712890625,
0.050506591796875,
-0.01467132568359375,
0.0219573974609375,
0.041839599609375,
0.0278472900390625,
-0.01306915283203125,
0.04791259765625,
-0.00994873046875,
-0.045867919921875,
-0.00775909423828125,
-0.048187255859375,
-0.0207977294921875,
0.017486572265625,
-0.042694091796875,
0.017578125,
-0.052642822265625,
-0.0146331787109375,
0.0006875991821289062,
0.0091552734375,
-0.053497314453125,
0.0164337158203125,
-0.0146331787109375,
0.07220458984375,
-0.074462890625,
0.037384033203125,
0.0621337890625,
-0.0557861328125,
-0.06793212890625,
-0.0086669921875,
0.006870269775390625,
-0.0533447265625,
0.02996826171875,
0.020233154296875,
0.01216888427734375,
-0.00006967782974243164,
-0.05908203125,
-0.07525634765625,
0.09759521484375,
0.00812530517578125,
-0.03106689453125,
-0.01091766357421875,
-0.0012159347534179688,
0.04248046875,
-0.031982421875,
0.050201416015625,
0.039031982421875,
0.047088623046875,
-0.0008597373962402344,
-0.07049560546875,
0.026336669921875,
-0.032257080078125,
-0.005542755126953125,
-0.001251220703125,
-0.06622314453125,
0.08563232421875,
-0.0235748291015625,
-0.0172119140625,
0.00821685791015625,
0.03515625,
0.0121002197265625,
0.0255889892578125,
0.0284423828125,
0.03668212890625,
0.07928466796875,
-0.006900787353515625,
0.0777587890625,
-0.0206298828125,
0.0205078125,
0.08721923828125,
-0.010040283203125,
0.0599365234375,
0.0298614501953125,
-0.03436279296875,
0.042572021484375,
0.06585693359375,
-0.00998687744140625,
0.04400634765625,
0.00505828857421875,
0.0024700164794921875,
0.0031337738037109375,
-0.00023818016052246094,
-0.0330810546875,
0.03839111328125,
0.0203857421875,
-0.01407623291015625,
-0.00664520263671875,
-0.0009684562683105469,
0.015411376953125,
-0.01078033447265625,
-0.00698089599609375,
0.056304931640625,
-0.0018634796142578125,
-0.06103515625,
0.050994873046875,
0.01313018798828125,
0.05096435546875,
-0.04400634765625,
-0.0105438232421875,
-0.0254974365234375,
-0.00978851318359375,
-0.02288818359375,
-0.0694580078125,
0.019775390625,
0.0049896240234375,
-0.0258941650390625,
0.0035572052001953125,
0.03240966796875,
-0.023773193359375,
-0.025146484375,
0.01117706298828125,
0.0181884765625,
0.049102783203125,
0.005878448486328125,
-0.056793212890625,
0.00926971435546875,
0.0083160400390625,
-0.02105712890625,
0.0116119384765625,
0.0270233154296875,
-0.0032596588134765625,
0.052337646484375,
0.0577392578125,
-0.0008573532104492188,
-0.002506256103515625,
-0.00939178466796875,
0.06597900390625,
-0.052001953125,
-0.045257568359375,
-0.064453125,
0.046112060546875,
-0.01059722900390625,
-0.036102294921875,
0.046783447265625,
0.0498046875,
0.054595947265625,
0.006618499755859375,
0.058807373046875,
-0.02203369140625,
0.02386474609375,
-0.039642333984375,
0.04815673828125,
-0.0477294921875,
0.0258636474609375,
-0.0101776123046875,
-0.048980712890625,
-0.005970001220703125,
0.06475830078125,
-0.015960693359375,
0.000011324882507324219,
0.050994873046875,
0.07049560546875,
0.001750946044921875,
0.00930023193359375,
-0.002109527587890625,
0.0182037353515625,
0.028289794921875,
0.04803466796875,
0.053558349609375,
-0.04840087890625,
0.0418701171875,
-0.021209716796875,
-0.034393310546875,
-0.00844573974609375,
-0.058135986328125,
-0.063720703125,
-0.041961669921875,
-0.005367279052734375,
-0.0306854248046875,
0.0090484619140625,
0.086669921875,
0.048553466796875,
-0.06365966796875,
-0.0298919677734375,
0.002735137939453125,
0.0066680908203125,
-0.0229034423828125,
-0.023681640625,
0.0197601318359375,
-0.014007568359375,
-0.05194091796875,
0.0322265625,
0.0016994476318359375,
0.00954437255859375,
-0.011627197265625,
-0.00408172607421875,
-0.0283966064453125,
0.01050567626953125,
0.045867919921875,
0.0262603759765625,
-0.053192138671875,
-0.02117919921875,
0.0134735107421875,
-0.00862884521484375,
0.0036678314208984375,
0.037445068359375,
-0.054595947265625,
0.025726318359375,
0.043121337890625,
0.02117919921875,
0.0310516357421875,
0.007266998291015625,
0.0260009765625,
-0.044036865234375,
0.006259918212890625,
0.00707244873046875,
0.0284423828125,
0.01520538330078125,
-0.0546875,
0.041839599609375,
0.0248260498046875,
-0.051605224609375,
-0.068359375,
0.0032749176025390625,
-0.08038330078125,
-0.03118896484375,
0.09649658203125,
-0.01236724853515625,
-0.01459503173828125,
-0.0134735107421875,
-0.032684326171875,
0.01337432861328125,
-0.05084228515625,
0.05084228515625,
0.049530029296875,
-0.0309906005859375,
0.0078125,
-0.039031982421875,
0.032989501953125,
-0.0005741119384765625,
-0.06939697265625,
-0.0029296875,
0.03839111328125,
0.041412353515625,
0.021331787109375,
0.0728759765625,
0.0078277587890625,
0.020721435546875,
0.002307891845703125,
-0.005970001220703125,
-0.0178680419921875,
-0.0298004150390625,
-0.0155181884765625,
0.008056640625,
-0.01922607421875,
-0.0248260498046875
]
] |
Doctor-Shotgun/CalliopeDS-v2-L2-13B | 2023-10-01T02:50:13.000Z | [
"transformers",
"safetensors",
"llama",
"text-generation",
"llama-2",
"en",
"license:llama2",
"text-generation-inference",
"region:us"
] | text-generation | Doctor-Shotgun | null | null | Doctor-Shotgun/CalliopeDS-v2-L2-13B | 1 | 5,918 | transformers | 2023-09-28T22:25:47 | ---
inference: false
language:
- en
library_name: transformers
pipeline_tag: text-generation
tags:
- llama
- llama-2
license: llama2
---
# CalliopeDS-v2-L2-13B
[EXL2 Quants](https://huggingface.co/Doctor-Shotgun/CalliopeDS-v2-L2-13B-exl2)
[GGUF Quants](https://huggingface.co/Doctor-Shotgun/Misc-Models)
This is a Llama 2-based model consisting of a merge of several models using PEFT adapters and SLERP merging:
- [PygmalionAI/pygmalion-2-13b](https://huggingface.co/PygmalionAI/pygmalion-2-13b)
- [NousResearch/Nous-Hermes-Llama2-13b](https://huggingface.co/NousResearch/Nous-Hermes-Llama2-13b)
- [Doctor-Shotgun/llama-2-supercot-lora](https://huggingface.co/Doctor-Shotgun/llama-2-supercot-lora)
- [lemonilia/LimaRP-Llama2-13B-v3-EXPERIMENT](https://huggingface.co/lemonilia/LimaRP-Llama2-13B-v3-EXPERIMENT)
- [Undi95/Storytelling-v2-13B-lora](https://huggingface.co/Undi95/Storytelling-v2-13B-lora)
Charles Goddard's [mergekit](https://github.com/cg123/mergekit) repo was used to perform these operations.
The purpose of this merge was to create a model that excels at creative writing and roleplay while maintaining general intelligence and instruction-following capabilities. In testing, it has shown to be capable at producing descriptive and verbose responses while demonstrating a solid understanding of the context.
## Usage:
Due to this being a merge of multiple models, different prompt formats may work, but you can try the Alpaca instruction format of LIMARP v3:
```
### Instruction:
Character's Persona: {bot character description}
User's Persona: {user character description}
Scenario: {what happens in the story}
Play the role of Character. You must engage in a roleplaying chat with User below this line. Do not write dialogues and narration for User.
### Input:
User: {utterance}
### Response:
Character: {utterance}
### Input
User: {utterance}
### Response:
Character: {utterance}
(etc.)
```
Or the Pygmalion/Metharme format:
```
<|system|>Enter RP mode. Pretend to be {{char}} whose persona follows:
{{persona}}
You shall reply to the user while staying in character, and generate long responses.
<|user|>Hello!<|model|>{model's response goes here}
```
The model was also tested using a system prompt with no instruction sequences:
```
Write Character's next reply in the roleplay between User and Character. Stay in character and write creative responses that move the scenario forward. Narrate in detail, using elaborate descriptions. The following is your persona:
{{persona}}
[Current conversation]
User: {utterance}
Character: {utterance}
```
## Message length control
Due to the inclusion of LimaRP v3, it is possible to append a length modifier to the response instruction sequence, like this:
```
### Input
User: {utterance}
### Response: (length = medium)
Character: {utterance}
```
This has an immediately noticeable effect on bot responses. The available lengths are: tiny, short, medium, long, huge, humongous, extreme, unlimited. The recommended starting length is medium. Keep in mind that the AI may ramble or impersonate the user with very long messages.
## Bias, Risks, and Limitations
The model will show biases similar to those observed in niche roleplaying forums on the Internet, besides those exhibited by the base model. It is not intended for supplying factual information or advice in any form.
## Training Details
This model is a merge. Please refer to the link repositories of the merged models for details. | 3,472 | [
[
-0.02935791015625,
-0.06512451171875,
0.0274810791015625,
0.03704833984375,
-0.0243682861328125,
-0.00658416748046875,
0.006031036376953125,
-0.0516357421875,
0.037353515625,
0.059967041015625,
-0.0557861328125,
-0.0154571533203125,
-0.044708251953125,
0.002918243408203125,
-0.021392822265625,
0.09552001953125,
0.008819580078125,
-0.0123138427734375,
-0.0012798309326171875,
0.0030078887939453125,
-0.04327392578125,
-0.040252685546875,
-0.058929443359375,
-0.0216827392578125,
0.049224853515625,
0.032562255859375,
0.049224853515625,
0.0301055908203125,
0.048187255859375,
0.022552490234375,
-0.017730712890625,
0.029022216796875,
-0.032318115234375,
0.026458740234375,
-0.0028896331787109375,
-0.032073974609375,
-0.06622314453125,
0.0034885406494140625,
0.047393798828125,
0.03033447265625,
-0.01163482666015625,
0.016357421875,
0.0038967132568359375,
0.00943756103515625,
-0.024383544921875,
0.0179443359375,
-0.006580352783203125,
0.0176239013671875,
-0.0019969940185546875,
-0.0088653564453125,
-0.033538818359375,
-0.0121002197265625,
0.0162506103515625,
-0.04840087890625,
0.0057373046875,
0.0173797607421875,
0.06353759765625,
0.00412750244140625,
-0.021942138671875,
-0.0263214111328125,
-0.04730224609375,
0.054901123046875,
-0.07025146484375,
-0.0032291412353515625,
0.03521728515625,
0.0233001708984375,
-0.024627685546875,
-0.0498046875,
-0.06201171875,
-0.01456451416015625,
-0.01332855224609375,
0.01363372802734375,
-0.0155792236328125,
-0.0027637481689453125,
0.0260772705078125,
0.0235443115234375,
-0.048736572265625,
0.01617431640625,
-0.041839599609375,
-0.007312774658203125,
0.037994384765625,
0.0263214111328125,
0.0195465087890625,
-0.03082275390625,
-0.046417236328125,
-0.0257415771484375,
-0.03375244140625,
0.00811004638671875,
0.038909912109375,
0.01459503173828125,
-0.03985595703125,
0.07965087890625,
0.0074462890625,
0.054290771484375,
0.0215301513671875,
-0.03033447265625,
0.01922607421875,
-0.02960205078125,
-0.02008056640625,
-0.0227508544921875,
0.0821533203125,
0.04083251953125,
-0.007289886474609375,
0.01459503173828125,
-0.0031681060791015625,
-0.00725555419921875,
0.0101165771484375,
-0.0518798828125,
0.0019140243530273438,
0.0284423828125,
-0.0386962890625,
-0.03265380859375,
-0.007564544677734375,
-0.0712890625,
-0.0340576171875,
0.011138916015625,
0.0199432373046875,
-0.041656494140625,
-0.01078033447265625,
0.01497650146484375,
-0.0194091796875,
0.041015625,
0.0266571044921875,
-0.072265625,
0.037841796875,
0.047637939453125,
0.06817626953125,
0.006500244140625,
-0.033355712890625,
-0.03662109375,
0.005706787109375,
-0.00955963134765625,
0.05242919921875,
-0.018402099609375,
-0.048675537109375,
-0.021820068359375,
0.007083892822265625,
-0.0037441253662109375,
-0.0275726318359375,
0.03765869140625,
-0.031494140625,
0.059478759765625,
-0.0172119140625,
-0.033416748046875,
-0.0238037109375,
0.0188446044921875,
-0.0438232421875,
0.061248779296875,
0.0150909423828125,
-0.05780029296875,
0.0045623779296875,
-0.06817626953125,
-0.022308349609375,
-0.006183624267578125,
0.00634765625,
-0.0224609375,
-0.0190582275390625,
0.0136871337890625,
0.0305023193359375,
-0.0311737060546875,
0.003482818603515625,
-0.0157470703125,
-0.0287933349609375,
0.021209716796875,
-0.01299285888671875,
0.07098388671875,
0.0087738037109375,
-0.0240936279296875,
0.0136260986328125,
-0.046142578125,
-0.0121002197265625,
0.023284912109375,
-0.023895263671875,
0.01494598388671875,
-0.00521087646484375,
0.0010814666748046875,
0.00868988037109375,
0.0257110595703125,
-0.02734375,
0.0400390625,
-0.0259857177734375,
0.029083251953125,
0.04571533203125,
0.01184844970703125,
0.03765869140625,
-0.052703857421875,
0.048797607421875,
-0.007503509521484375,
0.02557373046875,
-0.01412200927734375,
-0.06890869140625,
-0.078125,
-0.0305023193359375,
0.011138916015625,
0.06317138671875,
-0.03326416015625,
0.062103271484375,
-0.00037407875061035156,
-0.044036865234375,
-0.040191650390625,
-0.01099395751953125,
0.034088134765625,
0.0592041015625,
0.0263519287109375,
-0.038970947265625,
-0.055999755859375,
-0.06378173828125,
0.0220794677734375,
-0.042816162109375,
-0.0269317626953125,
0.039093017578125,
0.0291290283203125,
-0.0288848876953125,
0.057281494140625,
-0.032989501953125,
-0.007049560546875,
-0.0372314453125,
0.0164337158203125,
0.0257110595703125,
0.037017822265625,
0.039794921875,
-0.04425048828125,
-0.0064849853515625,
-0.0251617431640625,
-0.062469482421875,
-0.00751495361328125,
-0.010284423828125,
-0.001247406005859375,
0.01303863525390625,
-0.0012493133544921875,
-0.0709228515625,
0.0224456787109375,
0.041290283203125,
-0.036865234375,
0.03363037109375,
-0.0173187255859375,
0.027069091796875,
-0.08953857421875,
0.0200958251953125,
-0.017913818359375,
-0.01316070556640625,
-0.040008544921875,
0.0190887451171875,
-0.0170135498046875,
-0.008148193359375,
-0.041229248046875,
0.057220458984375,
-0.0274810791015625,
-0.0201416015625,
-0.0164031982421875,
0.01189422607421875,
0.00937652587890625,
0.039825439453125,
0.0079193115234375,
0.051605224609375,
0.0305023193359375,
-0.033355712890625,
0.0501708984375,
0.04058837890625,
-0.017181396484375,
0.0215301513671875,
-0.07159423828125,
0.0251007080078125,
-0.001434326171875,
0.0592041015625,
-0.0750732421875,
-0.0279998779296875,
0.06768798828125,
-0.0389404296875,
0.0020732879638671875,
-0.009613037109375,
-0.0570068359375,
-0.043701171875,
-0.0309600830078125,
0.0235443115234375,
0.059356689453125,
-0.0230255126953125,
0.04315185546875,
0.00861358642578125,
-0.020538330078125,
-0.03857421875,
-0.06292724609375,
0.005191802978515625,
-0.033843994140625,
-0.0614013671875,
0.019775390625,
-0.0294189453125,
-0.006198883056640625,
-0.0268096923828125,
0.030426025390625,
-0.00864410400390625,
-0.01439666748046875,
0.0263214111328125,
0.041961669921875,
-0.0124359130859375,
-0.01497650146484375,
0.004261016845703125,
0.01198577880859375,
-0.0112152099609375,
0.01983642578125,
0.07269287109375,
-0.010589599609375,
-0.017608642578125,
-0.044525146484375,
0.038116455078125,
0.053192138671875,
-0.010406494140625,
0.058685302734375,
0.046173095703125,
-0.01082611083984375,
0.026885986328125,
-0.04498291015625,
-0.02099609375,
-0.03216552734375,
0.004398345947265625,
-0.016357421875,
-0.0556640625,
0.061370849609375,
0.024993896484375,
0.00893402099609375,
0.0294647216796875,
0.045166015625,
-0.008575439453125,
0.06695556640625,
0.0462646484375,
0.006687164306640625,
0.0279693603515625,
-0.0266265869140625,
0.01340484619140625,
-0.0760498046875,
-0.04449462890625,
-0.01508331298828125,
-0.029144287109375,
-0.032501220703125,
-0.0577392578125,
0.01033782958984375,
0.00963592529296875,
-0.0235748291015625,
0.05078125,
-0.022857666015625,
0.01959228515625,
0.048736572265625,
0.0160064697265625,
0.01277923583984375,
-0.0129547119140625,
0.01119232177734375,
-0.0004177093505859375,
-0.047393798828125,
-0.04541015625,
0.0723876953125,
0.047698974609375,
0.04620361328125,
0.022552490234375,
0.057098388671875,
0.01015472412109375,
-0.007518768310546875,
-0.059844970703125,
0.0616455078125,
0.0156402587890625,
-0.0311737060546875,
-0.02398681640625,
-0.01285552978515625,
-0.06500244140625,
0.012176513671875,
-0.016632080078125,
-0.064453125,
0.00861358642578125,
-0.0036487579345703125,
-0.051483154296875,
0.0008640289306640625,
-0.05029296875,
0.048248291015625,
-0.0157470703125,
-0.00008958578109741211,
-0.005184173583984375,
-0.07684326171875,
0.04034423828125,
0.0036563873291015625,
-0.011810302734375,
-0.0033588409423828125,
-0.0240936279296875,
0.0648193359375,
-0.042144775390625,
0.079345703125,
0.01071929931640625,
-0.012939453125,
0.03802490234375,
0.008697509765625,
0.038360595703125,
0.01385498046875,
0.00348663330078125,
0.01299285888671875,
-0.01641845703125,
-0.0238494873046875,
-0.03790283203125,
0.047698974609375,
-0.06451416015625,
-0.04400634765625,
-0.0380859375,
-0.047698974609375,
0.0021648406982421875,
0.0010232925415039062,
0.038177490234375,
0.01258087158203125,
-0.01409149169921875,
-0.006954193115234375,
0.04962158203125,
-0.02392578125,
0.032958984375,
0.04437255859375,
-0.019134521484375,
-0.02972412109375,
0.026824951171875,
-0.019378662109375,
0.01154327392578125,
0.021209716796875,
0.00988006591796875,
-0.024200439453125,
-0.0124359130859375,
-0.030670166015625,
0.038604736328125,
-0.026885986328125,
-0.01465606689453125,
-0.059600830078125,
-0.025360107421875,
-0.047882080078125,
-0.00464630126953125,
-0.01090240478515625,
-0.0386962890625,
-0.0472412109375,
-0.002010345458984375,
0.03839111328125,
0.038787841796875,
-0.01654052734375,
0.04449462890625,
-0.053009033203125,
0.034271240234375,
0.031463623046875,
-0.00705718994140625,
-0.01151275634765625,
-0.0667724609375,
0.0178680419921875,
0.01471710205078125,
-0.02655029296875,
-0.07391357421875,
0.04571533203125,
0.0045318603515625,
0.0254364013671875,
0.03033447265625,
-0.0154876708984375,
0.0579833984375,
-0.03924560546875,
0.08197021484375,
0.030181884765625,
-0.05810546875,
0.06768798828125,
-0.043182373046875,
0.011138916015625,
0.0103302001953125,
0.0291900634765625,
-0.053314208984375,
-0.0296783447265625,
-0.059906005859375,
-0.06304931640625,
0.058380126953125,
0.0198974609375,
0.01947021484375,
-0.01088714599609375,
0.00970458984375,
0.0038852691650390625,
0.0139312744140625,
-0.0645751953125,
-0.0145111083984375,
-0.0065155029296875,
0.00807952880859375,
-0.0026950836181640625,
-0.0322265625,
-0.0266571044921875,
-0.009857177734375,
0.033233642578125,
0.0085296630859375,
0.039093017578125,
0.0062103271484375,
0.0124359130859375,
-0.01165771484375,
0.006954193115234375,
0.061798095703125,
0.0296783447265625,
-0.03070068359375,
-0.007724761962890625,
0.0061798095703125,
-0.0251617431640625,
-0.0034313201904296875,
0.0178985595703125,
0.0015621185302734375,
-0.01214599609375,
0.045257568359375,
0.059967041015625,
0.0252532958984375,
-0.067138671875,
0.042327880859375,
-0.011444091796875,
-0.00012022256851196289,
-0.0135345458984375,
0.0167388916015625,
0.0030841827392578125,
0.039093017578125,
0.0035915374755859375,
0.00366973876953125,
0.01154327392578125,
-0.054046630859375,
0.0010576248168945312,
0.008056640625,
0.003948211669921875,
-0.017547607421875,
0.043487548828125,
0.019805908203125,
-0.047210693359375,
0.055755615234375,
-0.0106048583984375,
-0.026824951171875,
0.057403564453125,
0.057037353515625,
0.04449462890625,
-0.0193939208984375,
0.0251922607421875,
0.02313232421875,
0.00576019287109375,
-0.0168914794921875,
0.01806640625,
-0.00405120849609375,
-0.039886474609375,
-0.020538330078125,
-0.0294342041015625,
-0.0308990478515625,
0.0274810791015625,
-0.038970947265625,
0.037353515625,
-0.053253173828125,
-0.0169677734375,
-0.01151275634765625,
0.0081024169921875,
-0.031463623046875,
-0.00357818603515625,
0.005970001220703125,
0.058013916015625,
-0.0625,
0.040252685546875,
0.0433349609375,
-0.048583984375,
-0.052520751953125,
-0.007633209228515625,
0.004421234130859375,
-0.07586669921875,
0.045654296875,
0.00455474853515625,
0.0142669677734375,
-0.02789306640625,
-0.04107666015625,
-0.0579833984375,
0.0982666015625,
0.01209259033203125,
-0.03680419921875,
-0.0204010009765625,
-0.0053253173828125,
0.05035400390625,
-0.049102783203125,
0.035308837890625,
0.0269012451171875,
0.0189056396484375,
0.03106689453125,
-0.096435546875,
0.0005578994750976562,
-0.013275146484375,
-0.0040283203125,
-0.00811767578125,
-0.072509765625,
0.0821533203125,
-0.0279541015625,
-0.0016660690307617188,
0.0506591796875,
0.05072021484375,
0.0506591796875,
0.005100250244140625,
0.031158447265625,
0.044921875,
0.0516357421875,
0.01378631591796875,
0.07098388671875,
-0.0101776123046875,
0.0264739990234375,
0.08984375,
-0.02203369140625,
0.0712890625,
0.036956787109375,
-0.00029468536376953125,
0.048736572265625,
0.0653076171875,
-0.0143585205078125,
0.04046630859375,
0.00870513916015625,
-0.0180511474609375,
-0.0015382766723632812,
-0.009796142578125,
-0.04345703125,
0.04327392578125,
0.017608642578125,
-0.03607177734375,
0.0080413818359375,
-0.00653839111328125,
0.01552581787109375,
-0.019012451171875,
0.0056304931640625,
0.043609619140625,
-0.0083160400390625,
-0.0611572265625,
0.031463623046875,
0.00177764892578125,
0.0654296875,
-0.054473876953125,
-0.01480865478515625,
-0.035797119140625,
-0.0033473968505859375,
0.005340576171875,
-0.0643310546875,
0.00412750244140625,
0.004150390625,
-0.0215606689453125,
-0.01416778564453125,
0.047027587890625,
-0.029144287109375,
-0.0345458984375,
0.02001953125,
0.037872314453125,
0.02838134765625,
0.0264739990234375,
-0.06195068359375,
0.012237548828125,
-0.00887298583984375,
-0.01480865478515625,
0.01715087890625,
0.0114288330078125,
0.0096435546875,
0.062164306640625,
0.034393310546875,
0.00045800209045410156,
-0.015472412109375,
-0.00022780895233154297,
0.06512451171875,
-0.03485107421875,
-0.037384033203125,
-0.048309326171875,
0.035919189453125,
-0.01204681396484375,
-0.0323486328125,
0.046356201171875,
0.041290283203125,
0.03521728515625,
-0.0135498046875,
0.035125732421875,
-0.0212554931640625,
0.04730224609375,
-0.0312347412109375,
0.04278564453125,
-0.03033447265625,
0.0271453857421875,
-0.017181396484375,
-0.06927490234375,
-0.002979278564453125,
0.057464599609375,
0.00009489059448242188,
0.005359649658203125,
0.03607177734375,
0.052947998046875,
-0.009246826171875,
-0.0168914794921875,
0.0112152099609375,
0.010955810546875,
0.018157958984375,
0.0595703125,
0.06610107421875,
-0.048583984375,
0.034759521484375,
-0.01308441162109375,
-0.0352783203125,
-0.0281524658203125,
-0.0584716796875,
-0.0775146484375,
-0.035858154296875,
-0.0096893310546875,
-0.039306640625,
0.0037555694580078125,
0.0587158203125,
0.05023193359375,
-0.036468505859375,
-0.0322265625,
0.00818634033203125,
-0.0015497207641601562,
-0.00254058837890625,
-0.020477294921875,
0.0038356781005859375,
0.007152557373046875,
-0.0645751953125,
0.03509521484375,
0.01015472412109375,
0.038177490234375,
-0.017303466796875,
-0.01076507568359375,
-0.017852783203125,
0.0135345458984375,
0.042724609375,
0.044769287109375,
-0.068603515625,
-0.02593994140625,
0.0157470703125,
-0.00945281982421875,
-0.008453369140625,
0.046356201171875,
-0.06011962890625,
0.017242431640625,
0.033447265625,
0.004100799560546875,
0.03887939453125,
-0.003627777099609375,
0.047882080078125,
-0.04833984375,
0.031646728515625,
0.0322265625,
0.0310516357421875,
0.0254364013671875,
-0.033477783203125,
0.04803466796875,
0.01374053955078125,
-0.047576904296875,
-0.055450439453125,
0.0198974609375,
-0.10980224609375,
-0.00614166259765625,
0.09991455078125,
-0.01189422607421875,
-0.006046295166015625,
0.0246429443359375,
-0.044708251953125,
0.01861572265625,
-0.041259765625,
0.05670166015625,
0.0499267578125,
-0.036346435546875,
-0.0231781005859375,
-0.01165771484375,
0.0189666748046875,
0.020233154296875,
-0.0638427734375,
-0.0160064697265625,
0.042327880859375,
0.018707275390625,
0.03125,
0.0628662109375,
0.00009334087371826172,
0.0243988037109375,
-0.002685546875,
-0.0041961669921875,
0.004116058349609375,
-0.0183868408203125,
-0.0129547119140625,
-0.006038665771484375,
-0.0131072998046875,
-0.0265045166015625
]
] |
eachadea/vicuna-13b-1.1 | 2023-05-02T09:07:12.000Z | [
"transformers",
"pytorch",
"llama",
"text-generation",
"license:apache-2.0",
"has_space",
"text-generation-inference",
"region:us"
] | text-generation | eachadea | null | null | eachadea/vicuna-13b-1.1 | 134 | 5,916 | transformers | 2023-04-13T01:47:56 | ---
license: apache-2.0
inference: false
---
**delta v1.1 merge**
<br>
<br>
# Vicuna Model Card
## Model details
**Model type:**
Vicuna is an open-source chatbot trained by fine-tuning LLaMA on user-shared conversations collected from ShareGPT.
It is an auto-regressive language model, based on the transformer architecture.
**Model date:**
Vicuna was trained between March 2023 and April 2023.
**Organizations developing the model:**
The Vicuna team with members from UC Berkeley, CMU, Stanford, and UC San Diego.
**Paper or resources for more information:**
https://vicuna.lmsys.org/
**License:**
Apache License 2.0
**Where to send questions or comments about the model:**
https://github.com/lm-sys/FastChat/issues
## Intended use
**Primary intended uses:**
The primary use of Vicuna is research on large language models and chatbots.
**Primary intended users:**
The primary intended users of the model are researchers and hobbyists in natural language processing, machine learning, and artificial intelligence.
## Training dataset
70K conversations collected from ShareGPT.com.
## Evaluation dataset
A preliminary evaluation of the model quality is conducted by creating a set of 80 diverse questions and utilizing GPT-4 to judge the model outputs. See https://vicuna.lmsys.org/ for more details.
## Major updates of weights v1.1
- Refactor the tokenization and separator. In Vicuna v1.1, the separator has been changed from `"###"` to the EOS token `"</s>"`. This change makes it easier to determine the generation stop criteria and enables better compatibility with other libraries.
- Fix the supervised fine-tuning loss computation for better model quality. | 1,679 | [
[
-0.0301513671875,
-0.06378173828125,
0.0255279541015625,
0.018096923828125,
-0.0391845703125,
-0.0285491943359375,
-0.00778961181640625,
-0.044189453125,
0.012664794921875,
0.044769287109375,
-0.0491943359375,
-0.039886474609375,
-0.036102294921875,
-0.01092529296875,
-0.0196990966796875,
0.06298828125,
0.0022869110107421875,
0.01091766357421875,
0.01506805419921875,
-0.006427764892578125,
-0.0716552734375,
-0.04437255859375,
-0.08416748046875,
-0.003063201904296875,
0.04949951171875,
0.036163330078125,
0.052032470703125,
0.045257568359375,
0.028472900390625,
0.0219879150390625,
0.001983642578125,
0.00537872314453125,
-0.0311431884765625,
0.0161285400390625,
0.003749847412109375,
-0.049407958984375,
-0.048828125,
-0.0321044921875,
0.03717041015625,
0.00858306884765625,
-0.011749267578125,
0.013458251953125,
-0.0012598037719726562,
0.0357666015625,
-0.032196044921875,
0.0276947021484375,
-0.04144287109375,
-0.01495361328125,
-0.01052093505859375,
-0.02984619140625,
-0.0189208984375,
-0.032867431640625,
-0.00946044921875,
-0.0404052734375,
0.0154571533203125,
-0.012908935546875,
0.0889892578125,
0.049072265625,
-0.0256195068359375,
-0.0164794921875,
-0.05914306640625,
0.034210205078125,
-0.059967041015625,
0.0300140380859375,
0.0218505859375,
0.057403564453125,
-0.0216217041015625,
-0.059478759765625,
-0.0465087890625,
-0.015411376953125,
0.0158843994140625,
0.0037078857421875,
-0.0276031494140625,
0.00888824462890625,
-0.0081024169921875,
0.03460693359375,
-0.03131103515625,
0.026397705078125,
-0.04742431640625,
0.000774383544921875,
0.034027099609375,
0.0285186767578125,
0.0146331787109375,
-0.0131378173828125,
-0.0267791748046875,
-0.019805908203125,
-0.0255279541015625,
-0.007740020751953125,
0.0219573974609375,
0.05322265625,
-0.017059326171875,
0.040557861328125,
-0.01444244384765625,
0.036346435546875,
-0.01374053955078125,
0.019073486328125,
0.007610321044921875,
-0.0169677734375,
-0.055450439453125,
-0.00981903076171875,
0.07696533203125,
0.031036376953125,
-0.0025196075439453125,
0.0104522705078125,
-0.00396728515625,
-0.005218505859375,
0.0227813720703125,
-0.0506591796875,
-0.0054779052734375,
0.050689697265625,
-0.021209716796875,
-0.02960205078125,
-0.002773284912109375,
-0.037445068359375,
-0.031005859375,
-0.0225830078125,
0.030181884765625,
-0.034423828125,
-0.035736083984375,
0.0230560302734375,
-0.0024890899658203125,
0.0197296142578125,
0.04522705078125,
-0.051483154296875,
0.0210723876953125,
0.052154541015625,
0.06634521484375,
0.017303466796875,
-0.030364990234375,
-0.001956939697265625,
-0.0220794677734375,
-0.0280609130859375,
0.0538330078125,
0.009307861328125,
-0.019927978515625,
0.00901031494140625,
0.01410675048828125,
0.00785064697265625,
-0.032745361328125,
0.040802001953125,
-0.034576416015625,
0.0338134765625,
-0.0299530029296875,
-0.038787841796875,
-0.00978851318359375,
0.0168304443359375,
-0.0677490234375,
0.07861328125,
0.0189971923828125,
-0.04071044921875,
0.02557373046875,
-0.0638427734375,
0.0180511474609375,
0.0147857666015625,
0.008453369140625,
-0.031982421875,
-0.01207733154296875,
-0.00510406494140625,
0.0311279296875,
-0.041259765625,
0.026702880859375,
-0.0078582763671875,
-0.023834228515625,
0.00974273681640625,
-0.037445068359375,
0.08013916015625,
0.03106689453125,
-0.0241546630859375,
0.034393310546875,
-0.0399169921875,
-0.006317138671875,
0.01317596435546875,
-0.005283355712890625,
-0.0164947509765625,
-0.0204925537109375,
0.0184173583984375,
0.0134429931640625,
0.03363037109375,
-0.01233673095703125,
0.035736083984375,
0.005828857421875,
0.0102691650390625,
0.060150146484375,
0.007335662841796875,
0.0211181640625,
-0.029266357421875,
0.0411376953125,
0.0027446746826171875,
0.060089111328125,
-0.00431060791015625,
-0.035186767578125,
-0.08380126953125,
-0.040802001953125,
0.00888824462890625,
0.03924560546875,
-0.04656982421875,
0.0509033203125,
-0.02801513671875,
-0.07318115234375,
-0.059234619140625,
0.0022678375244140625,
0.0167694091796875,
0.007297515869140625,
0.033050537109375,
-0.0233612060546875,
-0.060760498046875,
-0.0679931640625,
-0.00528717041015625,
-0.031982421875,
0.0008044242858886719,
0.0142364501953125,
0.0152740478515625,
-0.03753662109375,
0.07861328125,
-0.0253753662109375,
-0.026702880859375,
-0.005512237548828125,
0.00775909423828125,
0.00411224365234375,
0.033294677734375,
0.049285888671875,
-0.045501708984375,
-0.029296875,
0.0015878677368164062,
-0.05902099609375,
0.0013446807861328125,
-0.012725830078125,
-0.0285491943359375,
0.031036376953125,
0.035064697265625,
-0.045928955078125,
0.0325927734375,
0.051605224609375,
-0.042572021484375,
0.03973388671875,
-0.0030384063720703125,
0.0078887939453125,
-0.10479736328125,
-0.005809783935546875,
-0.0001189112663269043,
-0.019500732421875,
-0.040618896484375,
0.0090484619140625,
-0.001209259033203125,
0.0223846435546875,
-0.05548095703125,
0.062286376953125,
-0.028656005859375,
0.01641845703125,
-0.035186767578125,
-0.0261993408203125,
-0.0149383544921875,
0.04083251953125,
-0.008575439453125,
0.043914794921875,
0.03973388671875,
-0.0780029296875,
0.033203125,
-0.00034165382385253906,
-0.00888824462890625,
0.01560211181640625,
-0.057403564453125,
0.0173187255859375,
-0.006839752197265625,
0.0164337158203125,
-0.0694580078125,
-0.004215240478515625,
0.04901123046875,
-0.04052734375,
0.013458251953125,
-0.00910186767578125,
-0.043365478515625,
-0.0157623291015625,
-0.006587982177734375,
0.0048675537109375,
0.03143310546875,
-0.03619384765625,
0.035675048828125,
0.0260467529296875,
0.006862640380859375,
-0.036407470703125,
-0.042724609375,
0.006961822509765625,
-0.02545166015625,
-0.0204620361328125,
-0.00432586669921875,
-0.013336181640625,
-0.02020263671875,
0.003864288330078125,
0.0062103271484375,
-0.021240234375,
0.007320404052734375,
0.0155181884765625,
0.0251617431640625,
-0.01152801513671875,
0.020050048828125,
0.006145477294921875,
-0.00130462646484375,
-0.01384735107421875,
-0.0163421630859375,
0.0753173828125,
-0.028656005859375,
0.0022487640380859375,
-0.0518798828125,
-0.007640838623046875,
0.051788330078125,
0.0084991455078125,
0.08740234375,
0.042938232421875,
-0.0270538330078125,
0.005218505859375,
-0.047943115234375,
-0.0157012939453125,
-0.035430908203125,
0.036376953125,
-0.029571533203125,
-0.057861328125,
0.053009033203125,
0.0264434814453125,
0.037811279296875,
0.04425048828125,
0.056793212890625,
0.004329681396484375,
0.02392578125,
0.066162109375,
0.0022220611572265625,
0.070556640625,
-0.0242919921875,
-0.006427764892578125,
-0.044281005859375,
-0.0252685546875,
-0.04449462890625,
0.004795074462890625,
-0.06121826171875,
-0.043365478515625,
-0.0127105712890625,
0.0109710693359375,
-0.0297698974609375,
0.0728759765625,
-0.04498291015625,
0.0123748779296875,
0.042816162109375,
0.00970458984375,
0.0128173828125,
-0.01222991943359375,
0.0216827392578125,
0.007617950439453125,
-0.05145263671875,
-0.052978515625,
0.08197021484375,
0.0460205078125,
0.035858154296875,
0.0187530517578125,
0.048797607421875,
0.027587890625,
0.044921875,
-0.058258056640625,
0.0372314453125,
0.0171661376953125,
-0.0650634765625,
-0.052093505859375,
-0.034027099609375,
-0.08123779296875,
0.0303497314453125,
-0.0008196830749511719,
-0.044769287109375,
0.0198822021484375,
0.01305389404296875,
-0.004505157470703125,
0.019927978515625,
-0.05670166015625,
0.0716552734375,
-0.013397216796875,
-0.0154571533203125,
0.00435638427734375,
-0.017333984375,
0.035003662109375,
-0.0074005126953125,
-0.0002841949462890625,
-0.00958251953125,
-0.004283905029296875,
0.039764404296875,
-0.0535888671875,
0.07745361328125,
-0.014892578125,
-0.0298919677734375,
0.021942138671875,
0.00841522216796875,
0.017913818359375,
-0.014404296875,
0.007198333740234375,
0.042022705078125,
0.006114959716796875,
-0.04522705078125,
-0.04315185546875,
0.04296875,
-0.0767822265625,
-0.0254669189453125,
-0.0299072265625,
-0.0182952880859375,
0.0119476318359375,
0.01093292236328125,
0.040313720703125,
-0.00016391277313232422,
-0.0227813720703125,
0.004352569580078125,
0.021270751953125,
-0.015106201171875,
0.015869140625,
0.045654296875,
-0.027740478515625,
-0.039459228515625,
0.04229736328125,
0.007274627685546875,
0.00823974609375,
0.026275634765625,
-0.0015096664428710938,
-0.02227783203125,
-0.0091552734375,
-0.011932373046875,
0.02178955078125,
-0.0501708984375,
-0.012786865234375,
-0.047454833984375,
-0.018310546875,
-0.041229248046875,
0.0218505859375,
-0.06982421875,
-0.0272979736328125,
-0.019622802734375,
-0.00829315185546875,
0.03546142578125,
0.04046630859375,
0.01175689697265625,
0.051422119140625,
-0.0496826171875,
0.01165771484375,
0.0128173828125,
0.0274810791015625,
-0.006381988525390625,
-0.05072021484375,
-0.0343017578125,
0.006122589111328125,
-0.0197906494140625,
-0.071533203125,
0.042755126953125,
-0.0245361328125,
0.045440673828125,
0.026824951171875,
0.002712249755859375,
0.058258056640625,
-0.009674072265625,
0.05731201171875,
0.0135040283203125,
-0.056060791015625,
0.040557861328125,
-0.0246124267578125,
0.024017333984375,
0.0423583984375,
0.0309295654296875,
-0.057830810546875,
-0.033416748046875,
-0.057647705078125,
-0.06781005859375,
0.038818359375,
0.0278778076171875,
0.0087738037109375,
-0.00734710693359375,
0.0213775634765625,
0.015838623046875,
0.01454925537109375,
-0.051666259765625,
-0.045166015625,
-0.021697998046875,
-0.01104736328125,
-0.0124664306640625,
-0.020294189453125,
0.007259368896484375,
-0.0196533203125,
0.056884765625,
0.005229949951171875,
0.033447265625,
0.00495147705078125,
0.004688262939453125,
-0.007595062255859375,
0.0134429931640625,
0.041961669921875,
0.032806396484375,
-0.0291290283203125,
-0.028472900390625,
-0.005558013916015625,
-0.033843994140625,
-0.004840850830078125,
0.01099395751953125,
-0.009246826171875,
0.0247344970703125,
0.0105133056640625,
0.08758544921875,
0.005199432373046875,
-0.0243682861328125,
0.007785797119140625,
-0.0478515625,
-0.0178375244140625,
-0.057891845703125,
0.0159149169921875,
-0.002323150634765625,
0.0255279541015625,
0.0171051025390625,
-0.006626129150390625,
0.0017271041870117188,
-0.047576904296875,
-0.0125885009765625,
0.01354217529296875,
-0.036773681640625,
-0.01262664794921875,
0.047943115234375,
0.01142120361328125,
-0.04364013671875,
0.040618896484375,
0.006153106689453125,
-0.0243682861328125,
0.044769287109375,
0.00537872314453125,
0.08758544921875,
-0.0279388427734375,
0.01148223876953125,
0.044952392578125,
0.03656005859375,
-0.015106201171875,
0.0149078369140625,
-0.006160736083984375,
-0.05731201171875,
-0.0029754638671875,
-0.0274658203125,
-0.04986572265625,
0.032562255859375,
-0.05780029296875,
0.047393798828125,
-0.030609130859375,
-0.03765869140625,
-0.0217132568359375,
0.01508331298828125,
-0.0711669921875,
0.00821685791015625,
0.005588531494140625,
0.06976318359375,
-0.0692138671875,
0.07720947265625,
0.041839599609375,
-0.052215576171875,
-0.061431884765625,
-0.010711669921875,
-0.00870513916015625,
-0.032623291015625,
0.02496337890625,
0.0006341934204101562,
0.00772857666015625,
-0.00522613525390625,
-0.051605224609375,
-0.052764892578125,
0.10418701171875,
0.034088134765625,
-0.054351806640625,
-0.00681304931640625,
0.002017974853515625,
0.049224853515625,
-0.0037384033203125,
0.035400390625,
0.03131103515625,
0.0231475830078125,
0.005298614501953125,
-0.09124755859375,
0.0014829635620117188,
-0.0277862548828125,
0.0006031990051269531,
-0.0063629150390625,
-0.0711669921875,
0.0777587890625,
0.013153076171875,
-0.0157012939453125,
0.01593017578125,
0.0616455078125,
0.034881591796875,
0.01197052001953125,
0.033355712890625,
0.01502227783203125,
0.080322265625,
-0.006763458251953125,
0.0836181640625,
-0.025177001953125,
0.02679443359375,
0.09381103515625,
0.00916290283203125,
0.058441162109375,
0.03289794921875,
-0.00099945068359375,
0.03704833984375,
0.05914306640625,
0.01488494873046875,
0.023284912109375,
-0.004810333251953125,
-0.006046295166015625,
-0.005435943603515625,
0.00991058349609375,
-0.045166015625,
0.029693603515625,
0.010467529296875,
-0.0270538330078125,
0.006988525390625,
-0.0065765380859375,
0.018524169921875,
-0.0231170654296875,
-0.009002685546875,
0.04803466796875,
0.000036716461181640625,
-0.049774169921875,
0.058074951171875,
0.012298583984375,
0.06268310546875,
-0.0511474609375,
0.0015592575073242188,
-0.041229248046875,
0.0241241455078125,
0.00872039794921875,
-0.023345947265625,
0.0096435546875,
0.006134033203125,
0.002552032470703125,
0.0038814544677734375,
0.0290679931640625,
-0.026947021484375,
-0.0247039794921875,
0.027252197265625,
0.031158447265625,
0.03411865234375,
-0.0166015625,
-0.055755615234375,
0.0288238525390625,
-0.00958251953125,
-0.0288238525390625,
0.008270263671875,
0.042266845703125,
-0.01203155517578125,
0.0703125,
0.0474853515625,
0.00775909423828125,
0.0105133056640625,
0.022186279296875,
0.0723876953125,
-0.048004150390625,
-0.04571533203125,
-0.058868408203125,
0.039947509765625,
-0.0014705657958984375,
-0.03485107421875,
0.06146240234375,
0.048980712890625,
0.054290771484375,
-0.00846099853515625,
0.056427001953125,
0.003963470458984375,
0.0169677734375,
-0.03265380859375,
0.0628662109375,
-0.048004150390625,
0.02020263671875,
-0.0156707763671875,
-0.071533203125,
-0.00835418701171875,
0.038543701171875,
-0.0016078948974609375,
0.00455474853515625,
0.050262451171875,
0.058685302734375,
0.0168914794921875,
-0.01280975341796875,
0.038177490234375,
0.01824951171875,
0.049163818359375,
0.0278778076171875,
0.04095458984375,
-0.06805419921875,
0.0404052734375,
-0.0167083740234375,
-0.012420654296875,
-0.0390625,
-0.043701171875,
-0.06842041015625,
-0.058349609375,
-0.0269622802734375,
-0.038421630859375,
0.0240020751953125,
0.0732421875,
0.04241943359375,
-0.03594970703125,
-0.033477783203125,
-0.0265045166015625,
-0.01531219482421875,
-0.016632080078125,
-0.0199737548828125,
0.01384735107421875,
-0.019287109375,
-0.07763671875,
0.002197265625,
-0.0129241943359375,
0.01666259765625,
-0.0340576171875,
-0.0259246826171875,
-0.00278472900390625,
0.0125274658203125,
0.0207672119140625,
0.0499267578125,
-0.04803466796875,
0.002429962158203125,
0.003528594970703125,
-0.038299560546875,
0.004093170166015625,
0.0321044921875,
-0.040618896484375,
0.0291290283203125,
0.0262298583984375,
0.00582122802734375,
0.037811279296875,
-0.01219940185546875,
0.0399169921875,
-0.03021240234375,
0.029296875,
0.0104217529296875,
0.0302581787109375,
0.02191162109375,
-0.02960205078125,
0.031219482421875,
-0.0054473876953125,
-0.026824951171875,
-0.06610107421875,
-0.0016260147094726562,
-0.07464599609375,
-0.0248870849609375,
0.09600830078125,
0.012786865234375,
-0.03839111328125,
-0.01219940185546875,
-0.03472900390625,
0.056884765625,
-0.0251312255859375,
0.0640869140625,
0.045257568359375,
0.0284576416015625,
-0.037994384765625,
-0.043914794921875,
0.033447265625,
0.007678985595703125,
-0.0682373046875,
0.00507354736328125,
0.03057861328125,
0.02557373046875,
-0.00004875659942626953,
0.0860595703125,
-0.00838470458984375,
0.030242919921875,
0.01168060302734375,
0.058380126953125,
-0.03082275390625,
-0.029510498046875,
-0.0257415771484375,
-0.015655517578125,
0.01007080078125,
-0.041961669921875
]
] |
Doctor-Shotgun/CalliopeDS-L2-13B | 2023-09-16T02:30:16.000Z | [
"transformers",
"safetensors",
"llama",
"text-generation",
"llama-2",
"en",
"arxiv:2306.01708",
"license:agpl-3.0",
"text-generation-inference",
"region:us"
] | text-generation | Doctor-Shotgun | null | null | Doctor-Shotgun/CalliopeDS-L2-13B | 5 | 5,915 | transformers | 2023-09-16T01:11:49 | ---
inference: false
language:
- en
library_name: transformers
pipeline_tag: text-generation
tags:
- llama
- llama-2
license: agpl-3.0
---
# Model Card: CalliopeDS-L2-13B
This is a Llama 2-based model consisting of a merge of several models using a weight-adjusted TIES merge ([Resolving Interference When Merging Models](https://arxiv.org/abs/2306.01708)):
- [jondurbin/airoboros-l2-13b-2.2](https://huggingface.co/jondurbin/airoboros-l2-13b-2.2)
- [elinas/chronos-13b-v2](https://huggingface.co/elinas/chronos-13b-v2)
- [NousResearch/Nous-Hermes-Llama2-13b](https://huggingface.co/NousResearch/Nous-Hermes-Llama2-13b)
- [lemonilia/limarp-llama2-v2](https://huggingface.co/lemonilia/limarp-llama2-v2)
- [PygmalionAI/pygmalion-2-13b](https://huggingface.co/PygmalionAI/pygmalion-2-13b)
Charles Goddard's [mergekit](https://github.com/cg123/mergekit) repo was used to perform these operations.
The purpose of this merge was to create a model that excels at creative writing and roleplay while maintaining general intelligence and instruction-following capabilities. In testing, it has shown to be capable at producing descriptive and verbose responses while demonstrating a solid understanding of the context.
## Usage:
Due to this being a merge of multiple models, different prompt formats may work, but you can try the Alpaca instruction format of the LIMARP v2:
```
### Instruction:
Character's Persona: {bot character description}
User's Persona: {user character description}
Scenario: {what happens in the story}
Play the role of Character. You must engage in a roleplaying chat with User below this line. Do not write dialogues and narration for User. Character should respond with messages of medium length.
### Input:
User: {utterance}
### Response:
Character: {utterance}
```
Or the Pygmalion/Metharme format:
```
<|system|>Enter RP mode. Pretend to be {{char}} whose persona follows:
{{persona}}
You shall reply to the user while staying in character, and generate long responses.
<|user|>Hello!<|model|>{model's response goes here}
```
The model was also tested using a system prompt with no instruction sequences:
```
Write Character's next reply in the roleplay between User and Character. Stay in character and write creative responses that move the scenario forward. Narrate in detail, using elaborate descriptions. The following is your persona:
{{persona}}
[Current conversation]
User: {utterance}
Character: {utterance}
```
## Bias, Risks, and Limitations
The model will show biases similar to those observed in niche roleplaying forums on the Internet, besides those exhibited by the base model. It is not intended for supplying factual information or advice in any form.
## Training Details
This model is a merge. Please refer to the link repositories of the merged models for details. | 2,815 | [
[
-0.025146484375,
-0.050079345703125,
0.0180816650390625,
0.0286865234375,
-0.03350830078125,
-0.005001068115234375,
0.0038127899169921875,
-0.04852294921875,
0.032318115234375,
0.0576171875,
-0.06817626953125,
-0.0143280029296875,
-0.047760009765625,
-0.001293182373046875,
-0.01149749755859375,
0.08917236328125,
0.002506256103515625,
-0.00007021427154541016,
-0.01195526123046875,
0.0028228759765625,
-0.049957275390625,
-0.041046142578125,
-0.06683349609375,
-0.04510498046875,
0.04217529296875,
0.03326416015625,
0.045928955078125,
0.0272064208984375,
0.039886474609375,
0.02276611328125,
-0.028839111328125,
0.0318603515625,
-0.036041259765625,
0.0234527587890625,
-0.0011653900146484375,
-0.046356201171875,
-0.06866455078125,
0.01155853271484375,
0.047576904296875,
0.027496337890625,
-0.01348876953125,
0.01446533203125,
0.0091705322265625,
0.00402069091796875,
-0.0178070068359375,
0.0189361572265625,
-0.0113372802734375,
0.0193023681640625,
-0.010406494140625,
-0.007022857666015625,
-0.02386474609375,
-0.0259246826171875,
0.0082244873046875,
-0.0528564453125,
-0.0005998611450195312,
0.01100921630859375,
0.062744140625,
0.00896453857421875,
-0.028289794921875,
-0.030303955078125,
-0.03973388671875,
0.05157470703125,
-0.0657958984375,
0.006397247314453125,
0.03759765625,
0.0251617431640625,
-0.023406982421875,
-0.045318603515625,
-0.055816650390625,
-0.0165863037109375,
-0.0028095245361328125,
0.022064208984375,
-0.01666259765625,
-0.0023784637451171875,
0.03070068359375,
0.02294921875,
-0.04376220703125,
0.0196990966796875,
-0.048980712890625,
-0.0032958984375,
0.0374755859375,
0.025421142578125,
0.0246734619140625,
-0.025146484375,
-0.046112060546875,
-0.0233917236328125,
-0.03094482421875,
0.010406494140625,
0.041717529296875,
0.0042724609375,
-0.047119140625,
0.078857421875,
0.0118865966796875,
0.0517578125,
0.0241546630859375,
-0.031768798828125,
0.01837158203125,
-0.020904541015625,
-0.0166473388671875,
-0.01531982421875,
0.0648193359375,
0.0478515625,
0.0011911392211914062,
0.01108551025390625,
-0.0019989013671875,
-0.004878997802734375,
0.007534027099609375,
-0.052825927734375,
-0.0017042160034179688,
0.0242919921875,
-0.0489501953125,
-0.039886474609375,
-0.00769805908203125,
-0.06787109375,
-0.03192138671875,
-0.00009709596633911133,
0.009857177734375,
-0.0293426513671875,
-0.0236968994140625,
-0.0015001296997070312,
-0.0240478515625,
0.035858154296875,
0.0283660888671875,
-0.06427001953125,
0.0328369140625,
0.046112060546875,
0.06427001953125,
0.00647735595703125,
-0.033172607421875,
-0.023590087890625,
0.0009889602661132812,
-0.01788330078125,
0.050872802734375,
-0.023040771484375,
-0.05560302734375,
-0.0217437744140625,
0.000797271728515625,
-0.011383056640625,
-0.034698486328125,
0.045654296875,
-0.024200439453125,
0.052734375,
-0.0226898193359375,
-0.03790283203125,
-0.0240478515625,
0.020050048828125,
-0.04730224609375,
0.06866455078125,
0.004932403564453125,
-0.058990478515625,
0.00159454345703125,
-0.0550537109375,
-0.013397216796875,
-0.0017328262329101562,
0.00890350341796875,
-0.018341064453125,
-0.01378631591796875,
0.0026340484619140625,
0.0301513671875,
-0.040435791015625,
0.00875091552734375,
-0.0281219482421875,
-0.0240936279296875,
0.029571533203125,
-0.012451171875,
0.06817626953125,
0.01392364501953125,
-0.014190673828125,
0.005550384521484375,
-0.04638671875,
-0.005916595458984375,
0.0209503173828125,
-0.0266265869140625,
0.0041351318359375,
-0.002407073974609375,
0.0036220550537109375,
0.0160675048828125,
0.031646728515625,
-0.02239990234375,
0.0496826171875,
-0.01177978515625,
0.02618408203125,
0.03924560546875,
0.01128387451171875,
0.0321044921875,
-0.057098388671875,
0.045135498046875,
-0.003757476806640625,
0.0223846435546875,
-0.007266998291015625,
-0.0687255859375,
-0.0770263671875,
-0.024444580078125,
0.006214141845703125,
0.06353759765625,
-0.035003662109375,
0.06390380859375,
0.002719879150390625,
-0.05267333984375,
-0.03692626953125,
-0.01197052001953125,
0.036376953125,
0.054718017578125,
0.019012451171875,
-0.039520263671875,
-0.052032470703125,
-0.058685302734375,
0.01332855224609375,
-0.042938232421875,
-0.010406494140625,
0.044769287109375,
0.026519775390625,
-0.037078857421875,
0.043853759765625,
-0.034881591796875,
-0.0063018798828125,
-0.03961181640625,
0.0215301513671875,
0.0170440673828125,
0.05096435546875,
0.032135009765625,
-0.041595458984375,
0.00791168212890625,
-0.0198516845703125,
-0.07183837890625,
-0.0116424560546875,
-0.01446533203125,
-0.0034122467041015625,
0.0033283233642578125,
0.005504608154296875,
-0.06414794921875,
0.02435302734375,
0.041534423828125,
-0.0367431640625,
0.027557373046875,
-0.0192108154296875,
0.0295867919921875,
-0.09503173828125,
0.028350830078125,
-0.016571044921875,
-0.00626373291015625,
-0.0467529296875,
0.0176544189453125,
-0.0175628662109375,
-0.01456451416015625,
-0.0271148681640625,
0.0589599609375,
-0.03265380859375,
-0.00848388671875,
-0.0219879150390625,
0.01050567626953125,
0.004123687744140625,
0.039459228515625,
0.012847900390625,
0.0311431884765625,
0.03515625,
-0.03485107421875,
0.047882080078125,
0.043670654296875,
-0.0089263916015625,
0.038604736328125,
-0.07049560546875,
0.0259246826171875,
-0.01129150390625,
0.04998779296875,
-0.083984375,
-0.028656005859375,
0.0618896484375,
-0.035247802734375,
0.0205535888671875,
-0.0152740478515625,
-0.04217529296875,
-0.049560546875,
-0.021209716796875,
0.035919189453125,
0.056182861328125,
-0.033721923828125,
0.049163818359375,
0.00746917724609375,
-0.01506805419921875,
-0.038238525390625,
-0.0701904296875,
0.005115509033203125,
-0.029022216796875,
-0.060882568359375,
0.022064208984375,
-0.033447265625,
-0.00267791748046875,
-0.016632080078125,
0.0227203369140625,
-0.0152587890625,
-0.00775146484375,
0.03314208984375,
0.045135498046875,
-0.00737762451171875,
-0.0154266357421875,
0.01340484619140625,
0.00951385498046875,
-0.01520538330078125,
0.0242462158203125,
0.061431884765625,
0.00321197509765625,
-0.016632080078125,
-0.0538330078125,
0.0293731689453125,
0.06500244140625,
-0.015777587890625,
0.051177978515625,
0.03692626953125,
-0.02252197265625,
0.0213623046875,
-0.05682373046875,
-0.011016845703125,
-0.0286865234375,
0.008056640625,
-0.021881103515625,
-0.047088623046875,
0.0728759765625,
0.0127716064453125,
0.00800323486328125,
0.0266265869140625,
0.0428466796875,
-0.0016565322875976562,
0.062469482421875,
0.038604736328125,
0.0031642913818359375,
0.0318603515625,
-0.0285491943359375,
0.005863189697265625,
-0.08514404296875,
-0.057891845703125,
-0.0206298828125,
-0.0277557373046875,
-0.03692626953125,
-0.05487060546875,
0.0013294219970703125,
0.0259246826171875,
-0.012939453125,
0.05108642578125,
-0.02435302734375,
0.01800537109375,
0.04425048828125,
0.018035888671875,
0.00862884521484375,
-0.0181427001953125,
0.003940582275390625,
-0.0009369850158691406,
-0.0556640625,
-0.041748046875,
0.0714111328125,
0.051177978515625,
0.054595947265625,
0.019012451171875,
0.06866455078125,
0.002239227294921875,
0.0037689208984375,
-0.056976318359375,
0.056243896484375,
0.01611328125,
-0.03729248046875,
-0.0126190185546875,
-0.01922607421875,
-0.056884765625,
0.019256591796875,
-0.027557373046875,
-0.06109619140625,
0.017181396484375,
0.006557464599609375,
-0.0361328125,
0.00788116455078125,
-0.043792724609375,
0.05645751953125,
-0.0162353515625,
-0.0013713836669921875,
-0.0144195556640625,
-0.059234619140625,
0.055816650390625,
0.00817108154296875,
-0.012847900390625,
0.00439453125,
-0.0174560546875,
0.06695556640625,
-0.039093017578125,
0.07989501953125,
0.00978851318359375,
-0.0056915283203125,
0.037139892578125,
0.00843048095703125,
0.036956787109375,
0.003814697265625,
0.005886077880859375,
0.0151519775390625,
-0.01142120361328125,
-0.01306915283203125,
-0.0361328125,
0.0577392578125,
-0.068115234375,
-0.027008056640625,
-0.04046630859375,
-0.041595458984375,
0.01180267333984375,
0.0023555755615234375,
0.042022705078125,
0.0262603759765625,
-0.0224456787109375,
-0.0178985595703125,
0.045379638671875,
-0.0181732177734375,
0.02587890625,
0.034149169921875,
-0.0117034912109375,
-0.048187255859375,
0.0285797119140625,
-0.01059722900390625,
0.0079345703125,
0.00839996337890625,
0.005718231201171875,
-0.027557373046875,
-0.00428009033203125,
-0.033050537109375,
0.04901123046875,
-0.0345458984375,
-0.00942230224609375,
-0.044586181640625,
-0.0203399658203125,
-0.037017822265625,
-0.00717926025390625,
-0.01309967041015625,
-0.038299560546875,
-0.048187255859375,
0.00287628173828125,
0.0216217041015625,
0.05096435546875,
-0.0128173828125,
0.046234130859375,
-0.054412841796875,
0.0277099609375,
0.0357666015625,
0.00301361083984375,
0.006641387939453125,
-0.06280517578125,
0.025146484375,
0.0088653564453125,
-0.022369384765625,
-0.0869140625,
0.037384033203125,
-0.006351470947265625,
0.04443359375,
0.036102294921875,
-0.0210113525390625,
0.06280517578125,
-0.030426025390625,
0.07537841796875,
0.03564453125,
-0.062744140625,
0.05560302734375,
-0.0310516357421875,
0.0094757080078125,
0.0208282470703125,
0.027435302734375,
-0.06256103515625,
-0.020782470703125,
-0.0712890625,
-0.04901123046875,
0.06292724609375,
0.0200042724609375,
0.0047454833984375,
-0.00446319580078125,
0.0222930908203125,
-0.01531982421875,
0.0197906494140625,
-0.06671142578125,
-0.0203704833984375,
-0.0096893310546875,
-0.00106048583984375,
-0.0013303756713867188,
-0.041259765625,
-0.024200439453125,
-0.00966644287109375,
0.0438232421875,
0.01485443115234375,
0.031768798828125,
0.01016998291015625,
0.01374053955078125,
-0.0247650146484375,
-0.0009684562683105469,
0.0654296875,
0.0201873779296875,
-0.0261383056640625,
-0.005176544189453125,
0.006481170654296875,
-0.0228118896484375,
-0.0057373046875,
0.0198974609375,
-0.0031871795654296875,
-0.01224517822265625,
0.051971435546875,
0.06658935546875,
0.034912109375,
-0.04974365234375,
0.03656005859375,
-0.01279449462890625,
-0.00183868408203125,
-0.00960540771484375,
0.01496124267578125,
0.00946044921875,
0.03857421875,
-0.004619598388671875,
0.011016845703125,
0.00038743019104003906,
-0.055816650390625,
0.00034332275390625,
0.019134521484375,
-0.0008978843688964844,
-0.017822265625,
0.053741455078125,
0.0301513671875,
-0.0452880859375,
0.05133056640625,
-0.0107574462890625,
-0.03564453125,
0.059326171875,
0.052398681640625,
0.054901123046875,
-0.030792236328125,
0.01062774658203125,
0.04046630859375,
0.0140838623046875,
-0.0281219482421875,
0.0213623046875,
-0.0104522705078125,
-0.0413818359375,
-0.00966644287109375,
-0.020263671875,
-0.0292816162109375,
0.02069091796875,
-0.047332763671875,
0.033447265625,
-0.04425048828125,
-0.0210113525390625,
-0.0165252685546875,
0.01142120361328125,
-0.0310516357421875,
-0.00009423494338989258,
0.016632080078125,
0.0660400390625,
-0.0738525390625,
0.045318603515625,
0.050048828125,
-0.046234130859375,
-0.058837890625,
-0.01010894775390625,
0.005451202392578125,
-0.07568359375,
0.05108642578125,
0.0031681060791015625,
0.0214385986328125,
-0.0237274169921875,
-0.038970947265625,
-0.062164306640625,
0.10711669921875,
0.006763458251953125,
-0.039825439453125,
-0.012786865234375,
-0.01702880859375,
0.039154052734375,
-0.046356201171875,
0.042327880859375,
0.0190277099609375,
0.022918701171875,
0.04083251953125,
-0.09423828125,
0.003597259521484375,
-0.0157470703125,
0.001964569091796875,
-0.00787353515625,
-0.0726318359375,
0.07763671875,
-0.024566650390625,
-0.0020847320556640625,
0.05072021484375,
0.0531005859375,
0.052032470703125,
0.0105438232421875,
0.0352783203125,
0.0543212890625,
0.06500244140625,
0.0077972412109375,
0.0628662109375,
-0.0205230712890625,
0.0267181396484375,
0.08477783203125,
-0.01873779296875,
0.06201171875,
0.03387451171875,
0.014068603515625,
0.042724609375,
0.059906005859375,
-0.01409912109375,
0.035797119140625,
0.00910186767578125,
-0.01171112060546875,
-0.01299285888671875,
-0.00965118408203125,
-0.040985107421875,
0.036224365234375,
0.008819580078125,
-0.036865234375,
0.0017023086547851562,
-0.0119171142578125,
0.0149383544921875,
-0.0194244384765625,
-0.01007843017578125,
0.047576904296875,
-0.00319671630859375,
-0.076904296875,
0.03326416015625,
0.0023021697998046875,
0.06109619140625,
-0.060516357421875,
-0.0209503173828125,
-0.0341796875,
0.0004696846008300781,
-0.00445556640625,
-0.05841064453125,
-0.0008325576782226562,
0.003711700439453125,
-0.0191192626953125,
-0.01024627685546875,
0.050262451171875,
-0.0443115234375,
-0.037384033203125,
0.0182037353515625,
0.0322265625,
0.032135009765625,
0.034820556640625,
-0.06689453125,
0.019195556640625,
-0.0007948875427246094,
-0.006038665771484375,
0.01290130615234375,
0.0049896240234375,
0.00789642333984375,
0.062255859375,
0.032684326171875,
0.006496429443359375,
-0.01080322265625,
0.0006351470947265625,
0.066162109375,
-0.03106689453125,
-0.039031982421875,
-0.04217529296875,
0.040252685546875,
-0.013458251953125,
-0.043792724609375,
0.04217529296875,
0.03955078125,
0.032012939453125,
-0.0087432861328125,
0.036163330078125,
-0.01447296142578125,
0.038665771484375,
-0.0341796875,
0.04315185546875,
-0.036590576171875,
0.0183258056640625,
-0.0165252685546875,
-0.0684814453125,
-0.00035572052001953125,
0.063720703125,
0.0015497207641601562,
0.0035266876220703125,
0.03143310546875,
0.0552978515625,
-0.01000213623046875,
-0.015533447265625,
0.01094818115234375,
0.006130218505859375,
0.01456451416015625,
0.05645751953125,
0.06854248046875,
-0.049224853515625,
0.02569580078125,
-0.01172637939453125,
-0.042755126953125,
-0.023406982421875,
-0.058502197265625,
-0.0928955078125,
-0.03192138671875,
-0.0203399658203125,
-0.03955078125,
-0.0028629302978515625,
0.069091796875,
0.048614501953125,
-0.029571533203125,
-0.04010009765625,
0.0098114013671875,
0.00431060791015625,
-0.005870819091796875,
-0.01617431640625,
-0.000640869140625,
0.0105743408203125,
-0.0614013671875,
0.028472900390625,
0.0005831718444824219,
0.043792724609375,
-0.014190673828125,
-0.0167694091796875,
-0.0120086669921875,
0.00986480712890625,
0.0268707275390625,
0.041229248046875,
-0.06707763671875,
-0.01311492919921875,
0.005321502685546875,
-0.00933074951171875,
-0.0194091796875,
0.054718017578125,
-0.042236328125,
0.021484375,
0.035797119140625,
0.0021648406982421875,
0.04034423828125,
0.00838470458984375,
0.03955078125,
-0.041839599609375,
0.027557373046875,
0.02191162109375,
0.0335693359375,
0.0247955322265625,
-0.0330810546875,
0.03973388671875,
0.02386474609375,
-0.047393798828125,
-0.057373046875,
0.0192108154296875,
-0.1134033203125,
-0.00785064697265625,
0.09600830078125,
-0.0033721923828125,
-0.007808685302734375,
0.0305938720703125,
-0.04425048828125,
0.028045654296875,
-0.032623291015625,
0.04486083984375,
0.04974365234375,
-0.028350830078125,
-0.017913818359375,
-0.009002685546875,
0.0240020751953125,
0.0160369873046875,
-0.059234619140625,
-0.00920867919921875,
0.038970947265625,
0.0233917236328125,
0.025634765625,
0.055694580078125,
0.0045013427734375,
0.021575927734375,
-0.00037169456481933594,
0.020263671875,
-0.006916046142578125,
-0.0141754150390625,
-0.025054931640625,
-0.01224517822265625,
-0.00339508056640625,
-0.0228729248046875
]
] |
BreadAi/MuseCan | 2023-03-21T22:21:28.000Z | [
"transformers",
"pytorch",
"gpt2",
"text-generation",
"dataset:breadlicker45/musenet-encoders-12k",
"endpoints_compatible",
"has_space",
"text-generation-inference",
"region:us"
] | text-generation | BreadAi | null | null | BreadAi/MuseCan | 0 | 5,914 | transformers | 2023-02-24T22:37:41 | ---
datasets:
- breadlicker45/musenet-encoders-12k
---
MuseCan: A Generative Model for Music Generation
Introduction
Music is a powerful form of expression that can be enjoyed by people of all ages and backgrounds. It can be used to convey a wide range of emotions and to tell stories in a unique way. However, creating music can be a difficult and time-consuming process.
In recent years, there has been a growing interest in the use of artificial intelligence (AI) to generate music. AI-generated music can be used to create new pieces of music, to remix existing songs, or to provide background music for video games, movies, or other forms of media.
One of the most promising AI-generated music platforms is MuseCan. MuseCan is a generative model that can generate music from text. It was trained on a dataset of sheet music and lyrics, and can generate music in a variety of styles. It can also be used to generate music that is tailored to specific prompts, such as a specific genre or emotion.
How MuseCan Works
MuseCan uses a technique called deep learning to generate music. Deep learning is a type of machine learning that uses artificial neural networks to learn from data. In the case of MuseCan, the artificial neural network is trained on a dataset of sheet music and lyrics. Once the neural network is trained, it can be used to generate new pieces of music.
The Benefits of MuseCan
There are several benefits to using MuseCan to generate music. First, it can save time and effort. Creating music from scratch can be a time-consuming process, but MuseCan can generate music much more quickly. Second, MuseCan can be used to create music that is tailored to specific prompts. If you want music that is in a specific genre or that has a specific emotion, MuseCan can generate that music for you. Third, MuseCan can be used to create music that is unique and original. The music that MuseCan generates is not based on any existing songs, so it is sure to be something that no one else has ever heard before.
The Future of MuseCan
MuseCan is still under development, but it has the potential to be a powerful tool for music generation. In the future, MuseCan could be used to create music that is even more realistic and lifelike. It could also be used to create music that is tailored to specific individuals. For example, MuseCan could be used to create a song that is specifically for a person's birthday or anniversary.
Conclusion
MuseCan is a powerful tool that can be used to generate a variety of musical styles. It can be used to generate music for a variety of purposes, such as for video games, movies, or simply for personal enjoyment. MuseCan is still under development, but it has the potential to be a powerful tool for music generation. | 2,769 | [
[
-0.047393798828125,
-0.0213775634765625,
0.0261688232421875,
0.0325927734375,
-0.024993896484375,
0.0167083740234375,
-0.021484375,
-0.031402587890625,
0.0180511474609375,
0.038055419921875,
-0.0782470703125,
-0.033355712890625,
-0.019622802734375,
-0.0118560791015625,
-0.0477294921875,
0.080322265625,
0.021484375,
0.00649261474609375,
-0.0157318115234375,
0.01549530029296875,
-0.06378173828125,
-0.00551605224609375,
-0.066162109375,
-0.0206146240234375,
0.036285400390625,
0.02166748046875,
-0.0014257431030273438,
0.002628326416015625,
0.04412841796875,
0.0276641845703125,
0.00528717041015625,
0.0118560791015625,
-0.036590576171875,
0.039703369140625,
-0.00469970703125,
-0.0482177734375,
-0.0162200927734375,
0.032958984375,
0.0013418197631835938,
0.0146484375,
0.00370025634765625,
0.0244140625,
-0.0006952285766601562,
0.041778564453125,
-0.012664794921875,
-0.01065826416015625,
-0.05145263671875,
0.036285400390625,
-0.044342041015625,
-0.03253173828125,
-0.003139495849609375,
-0.048095703125,
-0.000091552734375,
-0.052490234375,
0.0265960693359375,
0.0022907257080078125,
0.05255126953125,
0.02947998046875,
-0.007480621337890625,
-0.05621337890625,
-0.08648681640625,
0.0312042236328125,
-0.033050537109375,
0.001323699951171875,
0.01282501220703125,
0.065185546875,
-0.0171966552734375,
-0.072265625,
-0.05322265625,
-0.032257080078125,
0.02423095703125,
0.045440673828125,
0.004581451416015625,
-0.01345062255859375,
0.02044677734375,
0.07135009765625,
-0.0306549072265625,
-0.01141357421875,
-0.033355712890625,
-0.007415771484375,
0.051055908203125,
0.0286712646484375,
0.042388916015625,
-0.0177154541015625,
-0.033721923828125,
-0.01175689697265625,
-0.071044921875,
0.023712158203125,
0.0556640625,
0.0025920867919921875,
0.015167236328125,
0.0457763671875,
0.01617431640625,
0.03704833984375,
0.036102294921875,
-0.0445556640625,
0.00780487060546875,
-0.00766754150390625,
-0.0162811279296875,
0.010528564453125,
0.06634521484375,
0.013275146484375,
-0.011505126953125,
-0.019805908203125,
-0.03472900390625,
0.015106201171875,
0.034912109375,
-0.055206298828125,
-0.0127716064453125,
0.0108795166015625,
-0.037567138671875,
-0.034820556640625,
-0.00038623809814453125,
-0.047637939453125,
-0.0274505615234375,
-0.0479736328125,
0.041778564453125,
-0.042236328125,
-0.00344085693359375,
-0.0020503997802734375,
-0.019439697265625,
-0.01739501953125,
0.0252838134765625,
-0.049713134765625,
-0.004741668701171875,
0.036529541015625,
0.06463623046875,
-0.0238037109375,
-0.00832366943359375,
-0.00665283203125,
0.00760650634765625,
-0.01580810546875,
0.05157470703125,
-0.002330780029296875,
-0.042999267578125,
-0.036834716796875,
0.0018310546875,
0.00507354736328125,
-0.056610107421875,
0.036407470703125,
-0.0303955078125,
0.0211181640625,
-0.0180511474609375,
-0.040313720703125,
-0.019561767578125,
-0.01641845703125,
-0.0653076171875,
0.06634521484375,
0.0168304443359375,
-0.031036376953125,
0.0545654296875,
-0.060455322265625,
-0.04412841796875,
0.007106781005859375,
-0.00966644287109375,
-0.042449951171875,
0.02374267578125,
-0.0015430450439453125,
0.03070068359375,
-0.04718017578125,
0.03070068359375,
-0.005016326904296875,
-0.030364990234375,
0.021759033203125,
-0.040771484375,
0.05615234375,
0.053375244140625,
-0.05816650390625,
0.00020241737365722656,
-0.0438232421875,
-0.0275421142578125,
0.0087432861328125,
-0.045684814453125,
0.010101318359375,
0.00199127197265625,
0.02288818359375,
0.031982421875,
-0.01485443115234375,
-0.053558349609375,
0.00428009033203125,
-0.0196685791015625,
0.035797119140625,
0.03900146484375,
0.0181884765625,
0.01446533203125,
-0.044769287109375,
0.05712890625,
-0.009918212890625,
-0.0115966796875,
-0.020294189453125,
-0.041839599609375,
-0.034942626953125,
-0.01486968994140625,
0.0239715576171875,
0.03778076171875,
-0.0232391357421875,
0.0293731689453125,
0.0039520263671875,
-0.05596923828125,
-0.049774169921875,
0.0186767578125,
-0.01293182373046875,
0.037109375,
0.00484466552734375,
-0.02410888671875,
-0.05377197265625,
-0.04345703125,
0.0237884521484375,
0.01055908203125,
-0.00299835205078125,
0.053680419921875,
0.01381683349609375,
-0.0039043426513671875,
0.06640625,
-0.0198211669921875,
-0.033660888671875,
-0.022430419921875,
0.0272674560546875,
0.0238189697265625,
0.06658935546875,
0.06610107421875,
-0.05548095703125,
-0.021759033203125,
-0.0108795166015625,
-0.041107177734375,
-0.018218994140625,
-0.049468994140625,
-0.01519775390625,
-0.030975341796875,
0.0185546875,
-0.0672607421875,
0.0218963623046875,
0.017242431640625,
0.00485992431640625,
0.048126220703125,
0.01134490966796875,
0.0216827392578125,
-0.09417724609375,
0.0191192626953125,
-0.0316162109375,
0.00040435791015625,
-0.029876708984375,
-0.0139312744140625,
-0.0377197265625,
-0.042022705078125,
-0.031219482421875,
0.01100921630859375,
-0.0188140869140625,
-0.01311492919921875,
-0.038482666015625,
-0.00506591796875,
0.006130218505859375,
0.01678466796875,
0.0022296905517578125,
0.05950927734375,
0.0457763671875,
-0.03839111328125,
0.0592041015625,
0.0013246536254882812,
-0.0582275390625,
0.0655517578125,
-0.042724609375,
0.00559234619140625,
-0.00738525390625,
0.0167694091796875,
-0.06585693359375,
-0.018798828125,
-0.0020904541015625,
-0.05548095703125,
0.032623291015625,
0.018218994140625,
-0.01959228515625,
-0.0635986328125,
0.0273590087890625,
0.0197906494140625,
0.031341552734375,
-0.0452880859375,
0.040252685546875,
0.02020263671875,
-0.003238677978515625,
-0.0059814453125,
-0.0662841796875,
0.0128936767578125,
-0.0482177734375,
-0.05474853515625,
0.045440673828125,
-0.01332855224609375,
-0.00402069091796875,
0.0076904296875,
0.00042438507080078125,
0.03033447265625,
0.03167724609375,
0.05743408203125,
0.00811004638671875,
-0.027130126953125,
0.0227203369140625,
0.01543426513671875,
-0.01065826416015625,
0.007297515869140625,
-0.04754638671875,
0.058441162109375,
0.0011091232299804688,
-0.041046142578125,
-0.032135009765625,
0.04156494140625,
0.07037353515625,
-0.0260009765625,
0.0175323486328125,
0.07867431640625,
-0.01546478271484375,
-0.020965576171875,
-0.033355712890625,
-0.0121917724609375,
-0.034912109375,
-0.006763458251953125,
-0.035430908203125,
-0.039642333984375,
0.048797607421875,
-0.031585693359375,
0.008880615234375,
0.045257568359375,
0.064697265625,
-0.04083251953125,
0.06658935546875,
0.029937744140625,
-0.00893402099609375,
0.042388916015625,
-0.05242919921875,
-0.0212554931640625,
-0.06695556640625,
-0.043701171875,
-0.031646728515625,
-0.060211181640625,
-0.02154541015625,
-0.001575469970703125,
0.01299285888671875,
-0.00620269775390625,
-0.01486968994140625,
0.04156494140625,
-0.046417236328125,
0.002559661865234375,
0.07989501953125,
0.00510406494140625,
-0.00914764404296875,
-0.001758575439453125,
-0.03131103515625,
0.0124664306640625,
-0.024383544921875,
-0.01898193359375,
0.0938720703125,
-0.005886077880859375,
0.064453125,
0.0135955810546875,
0.055450439453125,
0.039886474609375,
0.015838623046875,
-0.059417724609375,
0.0217132568359375,
0.0034694671630859375,
-0.060791015625,
-0.0052490234375,
-0.00887298583984375,
-0.0775146484375,
-0.01398468017578125,
-0.04034423828125,
-0.052215576171875,
0.0282135009765625,
-0.0027313232421875,
-0.03509521484375,
0.028533935546875,
-0.03411865234375,
0.04644775390625,
-0.038299560546875,
-0.0168609619140625,
-0.00621795654296875,
-0.02099609375,
0.031158447265625,
-0.0124053955078125,
-0.031280517578125,
-0.018157958984375,
0.03314208984375,
0.035125732421875,
-0.031646728515625,
0.042694091796875,
-0.01136016845703125,
0.004352569580078125,
0.054779052734375,
0.0200958251953125,
0.024993896484375,
0.0229339599609375,
0.00702667236328125,
-0.0190277099609375,
-0.0289154052734375,
0.004596710205078125,
-0.038330078125,
0.06280517578125,
-0.062744140625,
-0.060211181640625,
0.008941650390625,
-0.057525634765625,
-0.0009541511535644531,
0.0167083740234375,
0.059661865234375,
0.0213775634765625,
-0.0102691650390625,
0.03460693359375,
0.03387451171875,
-0.01038360595703125,
0.041534423828125,
0.01100921630859375,
-0.0096893310546875,
-0.03558349609375,
0.073974609375,
0.002960205078125,
-0.004974365234375,
0.0132293701171875,
0.00916290283203125,
-0.0135498046875,
-0.03643798828125,
-0.05145263671875,
-0.021697998046875,
-0.016510009765625,
-0.0175323486328125,
-0.042388916015625,
-0.005397796630859375,
-0.016265869140625,
-0.01065826416015625,
-0.020416259765625,
-0.032440185546875,
-0.041961669921875,
-0.0217132568359375,
0.0269622802734375,
0.029571533203125,
-0.0281219482421875,
0.01357269287109375,
-0.0313720703125,
0.0238189697265625,
0.0435791015625,
-0.005741119384765625,
0.0013790130615234375,
0.00580596923828125,
-0.0005712509155273438,
-0.007549285888671875,
-0.03411865234375,
-0.090087890625,
0.02301025390625,
0.01354217529296875,
0.044281005859375,
0.05010986328125,
-0.02288818359375,
0.03826904296875,
-0.0212249755859375,
0.061981201171875,
0.0147552490234375,
-0.056640625,
0.059967041015625,
-0.043609619140625,
0.0028247833251953125,
0.04205322265625,
0.0028743743896484375,
-0.044586181640625,
-0.0269317626953125,
-0.039825439453125,
-0.0589599609375,
0.051513671875,
0.03839111328125,
0.00986480712890625,
0.01715087890625,
0.0191497802734375,
0.029754638671875,
0.047760009765625,
-0.045440673828125,
-0.01261138916015625,
-0.0655517578125,
-0.01146697998046875,
-0.0183258056640625,
0.0318603515625,
-0.019927978515625,
-0.0127716064453125,
0.04034423828125,
0.0048980712890625,
0.0543212890625,
-0.005374908447265625,
0.015777587890625,
-0.01788330078125,
-0.0011281967163085938,
0.04937744140625,
-0.0017490386962890625,
-0.00907135009765625,
-0.0014410018920898438,
-0.00757598876953125,
-0.0477294921875,
0.0256805419921875,
-0.0006380081176757812,
-0.035675048828125,
-0.0085296630859375,
-0.008148193359375,
0.08734130859375,
-0.02044677734375,
-0.04400634765625,
0.0221099853515625,
0.00711822509765625,
-0.00250244140625,
-0.04510498046875,
0.037933349609375,
0.03387451171875,
0.0160369873046875,
0.0244903564453125,
0.050567626953125,
0.0220184326171875,
-0.044921875,
0.025970458984375,
0.021697998046875,
-0.038055419921875,
-0.013336181640625,
0.0810546875,
0.0006670951843261719,
-0.0273895263671875,
0.038421630859375,
-0.00894927978515625,
-0.01073455810546875,
0.0772705078125,
0.06964111328125,
0.080810546875,
-0.012115478515625,
0.035003662109375,
0.0196990966796875,
0.0254058837890625,
0.0292816162109375,
0.01470947265625,
-0.04412841796875,
-0.0287017822265625,
-0.0117950439453125,
-0.048431396484375,
-0.0135040283203125,
0.033203125,
-0.044464111328125,
0.002086639404296875,
-0.033203125,
-0.035614013671875,
0.00466156005859375,
0.0014944076538085938,
-0.062469482421875,
0.0355224609375,
0.01275634765625,
0.06756591796875,
-0.06036376953125,
0.027984619140625,
0.04705810546875,
-0.0440673828125,
-0.084716796875,
-0.0093994140625,
0.0143585205078125,
-0.031219482421875,
0.039703369140625,
0.0173492431640625,
0.001766204833984375,
-0.003002166748046875,
-0.0269012451171875,
-0.07275390625,
0.10626220703125,
0.01380157470703125,
-0.0316162109375,
0.01715087890625,
0.0086822509765625,
0.04833984375,
-0.0528564453125,
0.00177764892578125,
0.026702880859375,
0.047637939453125,
0.038787841796875,
-0.01155853271484375,
-0.00263214111328125,
-0.04229736328125,
0.0025653839111328125,
-0.0009469985961914062,
-0.036865234375,
0.08966064453125,
-0.0016412734985351562,
-0.02716064453125,
0.0291290283203125,
0.045501708984375,
0.01393890380859375,
0.031402587890625,
0.028778076171875,
0.056121826171875,
0.0718994140625,
-0.042266845703125,
0.0777587890625,
-0.0244598388671875,
0.004711151123046875,
0.06805419921875,
0.0022754669189453125,
0.00717926025390625,
0.01107025146484375,
0.0021991729736328125,
0.03643798828125,
0.07940673828125,
-0.03839111328125,
0.053436279296875,
0.0283355712890625,
-0.0343017578125,
-0.0267486572265625,
0.0025119781494140625,
-0.0323486328125,
0.039825439453125,
0.032379150390625,
-0.0531005859375,
0.0019435882568359375,
0.058746337890625,
-0.01319122314453125,
-0.0279541015625,
-0.0235137939453125,
0.05364990234375,
-0.005245208740234375,
-0.057769775390625,
-0.004016876220703125,
-0.0053253173828125,
0.027801513671875,
-0.0175628662109375,
0.0089111328125,
-0.0087738037109375,
-0.0007977485656738281,
-0.01416778564453125,
-0.048583984375,
0.0174713134765625,
-0.0110626220703125,
-0.019073486328125,
0.005069732666015625,
0.06671142578125,
-0.01751708984375,
-0.037933349609375,
-0.005596160888671875,
0.032318115234375,
0.005023956298828125,
-0.01024627685546875,
-0.051025390625,
-0.03948974609375,
0.0043792724609375,
-0.01497650146484375,
0.0005469322204589844,
0.0236968994140625,
0.0136260986328125,
0.049591064453125,
0.07452392578125,
0.0274810791015625,
0.0110931396484375,
0.0053863525390625,
0.057373046875,
-0.0286102294921875,
-0.037506103515625,
-0.0191497802734375,
0.035125732421875,
-0.004749298095703125,
-0.040771484375,
0.054473876953125,
0.04083251953125,
0.041290283203125,
-0.02801513671875,
0.0693359375,
-0.0330810546875,
0.0262298583984375,
-0.0307464599609375,
0.0626220703125,
-0.05645751953125,
0.0188140869140625,
-0.0286102294921875,
-0.08154296875,
0.005702972412109375,
0.035247802734375,
-0.0093231201171875,
0.00676727294921875,
0.0175018310546875,
0.0479736328125,
-0.01233673095703125,
0.021881103515625,
0.0104217529296875,
-0.0182647705078125,
0.01393890380859375,
0.0399169921875,
0.0712890625,
-0.0068817138671875,
0.03619384765625,
-0.039093017578125,
0.01317596435546875,
-0.01361846923828125,
-0.0279693603515625,
-0.061859130859375,
-0.04705810546875,
0.007724761962890625,
-0.0205230712890625,
-0.0170135498046875,
0.0565185546875,
0.06658935546875,
-0.06744384765625,
0.002338409423828125,
-0.03558349609375,
0.00772857666015625,
-0.0386962890625,
-0.021240234375,
0.021209716796875,
-0.012481689453125,
-0.07696533203125,
0.051727294921875,
0.03253173828125,
0.032196044921875,
-0.052154541015625,
0.0012884140014648438,
0.024566650390625,
0.0164794921875,
0.0221710205078125,
0.025787353515625,
-0.03424072265625,
-0.0089874267578125,
0.0180511474609375,
-0.011993408203125,
0.004711151123046875,
0.08270263671875,
-0.057830810546875,
0.033203125,
0.048126220703125,
0.0300445556640625,
0.07257080078125,
-0.0159454345703125,
0.04638671875,
-0.0198974609375,
-0.00925445556640625,
0.03564453125,
0.03558349609375,
0.0081787109375,
-0.0445556640625,
0.0294189453125,
0.025115966796875,
-0.036895751953125,
-0.031219482421875,
0.0280609130859375,
-0.0841064453125,
-0.00400543212890625,
0.046173095703125,
-0.0188446044921875,
-0.0039043426513671875,
-0.00027751922607421875,
-0.05145263671875,
0.01519775390625,
-0.021087646484375,
0.059051513671875,
0.0212249755859375,
-0.01465606689453125,
-0.0085296630859375,
-0.0222930908203125,
0.044952392578125,
0.00678253173828125,
-0.061248779296875,
-0.04827880859375,
0.03570556640625,
0.0198211669921875,
0.040130615234375,
0.010284423828125,
-0.0168304443359375,
0.054779052734375,
0.0196685791015625,
0.017303466796875,
-0.031402587890625,
-0.0279388427734375,
0.0198974609375,
0.0499267578125,
-0.0335693359375,
-0.00847625732421875
]
] |
CobraMamba/mamba-gpt-3b-v3 | 2023-08-03T01:55:05.000Z | [
"transformers",
"pytorch",
"safetensors",
"llama",
"text-generation",
"gpt",
"llm",
"large language model",
"en",
"license:apache-2.0",
"has_space",
"text-generation-inference",
"region:us"
] | text-generation | CobraMamba | null | null | CobraMamba/mamba-gpt-3b-v3 | 13 | 5,914 | transformers | 2023-07-28T07:45:24 | ---
language:
- en
library_name: transformers
tags:
- gpt
- llm
- large language model
inference: false
thumbnail: >-
https://h2o.ai/etc.clientlibs/h2o/clientlibs/clientlib-site/resources/images/favicon.ico
license: apache-2.0
---
# Model Card
**The Best 3B Model! Surpassing dolly-v2-12b**
The best 3B model on the [Open LLM Leaderboard](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard), with performance surpassing dolly-v2-12b
| Metric | Value |
|-----------------------|-------|
| MMLU (5-shot) | 27.3 |
| ARC (25-shot) | 41.7 |
| HellaSwag (10-shot) | 71.1 |
| TruthfulQA (0-shot) | 37.9 |
| Avg. | 44.5 |
We use state-of-the-art [Language Model Evaluation Harness](https://github.com/EleutherAI/lm-evaluation-harness) to run the benchmark tests above.
The training code and data will be open sourced later on Github(https://github.com/chi2liu/mamba-gpt-3b)
## Training Dataset
` mamba-gpt-3b-v3 ` is trained on multiply dataset:
- [Stanford Alpaca (en)](https://github.com/tatsu-lab/stanford_alpaca)
- [Open Assistant (multilingual)](https://huggingface.co/datasets/OpenAssistant/oasst1)
- [LIMA (en)](https://huggingface.co/datasets/GAIR/lima)
- [CodeAlpaca 20k (en)](https://huggingface.co/datasets/sahil2801/CodeAlpaca-20k)
## Summary
We have fine-tuned the open-lama model and surpassed the original model in multiple evaluation subtasks, making it currently the best performing 3B model with comparable performance to llama-7b
- Base model: [openlm-research/open_llama_3b_v2](https://huggingface.co/openlm-research/open_llama_3b_v2)
## Usage
To use the model with the `transformers` library on a machine with GPUs, first make sure you have the `transformers`, `accelerate` and `torch` libraries installed.
```bash
pip install transformers==4.29.2
pip install accelerate==0.19.0
pip install torch==2.0.0
```
```python
from transformers import AutoTokenizer, AutoModelForCausalLM
tokenizer = AutoTokenizer.from_pretrained("CobraMamba/mamba-gpt-3b-v3")
model = AutoModelForCausalLM.from_pretrained("CobraMamba/mamba-gpt-3b-v3", trust_remote_code=True, torch_dtype=torch.float16)
input_context = "Your text here"
input_ids = tokenizer.encode(input_context, return_tensors="pt")
output = model.generate(input_ids, max_length=128, temperature=0.7)
output_text = tokenizer.decode(output[0], skip_special_tokens=True)
print(output_text)
```
## Model Architecture
```
LlamaForCausalLM(
(model): LlamaModel(
(embed_tokens): Embedding(32000, 4096, padding_idx=0)
(layers): ModuleList(
(0-31): 32 x LlamaDecoderLayer(
(self_attn): LlamaAttention(
(q_proj): Linear(in_features=4096, out_features=4096, bias=False)
(k_proj): Linear(in_features=4096, out_features=4096, bias=False)
(v_proj): Linear(in_features=4096, out_features=4096, bias=False)
(o_proj): Linear(in_features=4096, out_features=4096, bias=False)
(rotary_emb): LlamaRotaryEmbedding()
)
(mlp): LlamaMLP(
(gate_proj): Linear(in_features=4096, out_features=11008, bias=False)
(down_proj): Linear(in_features=11008, out_features=4096, bias=False)
(up_proj): Linear(in_features=4096, out_features=11008, bias=False)
(act_fn): SiLUActivation()
)
(input_layernorm): LlamaRMSNorm()
(post_attention_layernorm): LlamaRMSNorm()
)
)
(norm): LlamaRMSNorm()
)
(lm_head): Linear(in_features=4096, out_features=32000, bias=False)
)
```
## Citation
If this work is helpful, please kindly cite as:
```bibtex
@Misc{mamba-gpt-3b-v3,
title = {Mamba-GPT-3b-v3},
author = {chiliu},
howpublished = {\url{https://huggingface.co/CobraMamba/mamba-gpt-3b-v3}},
year = {2023}
}
```
## Disclaimer
Please read this disclaimer carefully before using the large language model provided in this repository. Your use of the model signifies your agreement to the following terms and conditions.
- Biases and Offensiveness: The large language model is trained on a diverse range of internet text data, which may contain biased, racist, offensive, or otherwise inappropriate content. By using this model, you acknowledge and accept that the generated content may sometimes exhibit biases or produce content that is offensive or inappropriate. The developers of this repository do not endorse, support, or promote any such content or viewpoints.
- Limitations: The large language model is an AI-based tool and not a human. It may produce incorrect, nonsensical, or irrelevant responses. It is the user's responsibility to critically evaluate the generated content and use it at their discretion.
- Use at Your Own Risk: Users of this large language model must assume full responsibility for any consequences that may arise from their use of the tool. The developers and contributors of this repository shall not be held liable for any damages, losses, or harm resulting from the use or misuse of the provided model.
- Ethical Considerations: Users are encouraged to use the large language model responsibly and ethically. By using this model, you agree not to use it for purposes that promote hate speech, discrimination, harassment, or any form of illegal or harmful activities.
- Reporting Issues: If you encounter any biased, offensive, or otherwise inappropriate content generated by the large language model, please report it to the repository maintainers through the provided channels. Your feedback will help improve the model and mitigate potential issues.
- Changes to this Disclaimer: The developers of this repository reserve the right to modify or update this disclaimer at any time without prior notice. It is the user's responsibility to periodically review the disclaimer to stay informed about any changes.
By using the large language model provided in this repository, you agree to accept and comply with the terms and conditions outlined in this disclaimer. If you do not agree with any part of this disclaimer, you should refrain from using the model and any content generated by it.
| 6,120 | [
[
-0.03289794921875,
-0.069580078125,
0.0156402587890625,
0.028839111328125,
-0.035491943359375,
-0.01053619384765625,
-0.0206756591796875,
-0.036224365234375,
0.0154571533203125,
0.020965576171875,
-0.0249786376953125,
-0.054901123046875,
-0.0511474609375,
-0.00439453125,
-0.01042938232421875,
0.06298828125,
-0.0108184814453125,
-0.004596710205078125,
0.00244140625,
-0.00948333740234375,
-0.0223541259765625,
-0.050262451171875,
-0.040496826171875,
-0.02728271484375,
0.026763916015625,
0.0012264251708984375,
0.05157470703125,
0.050140380859375,
0.02838134765625,
0.0225830078125,
-0.0277099609375,
0.0130462646484375,
-0.0302734375,
-0.017242431640625,
0.01371002197265625,
-0.0237579345703125,
-0.05438232421875,
-0.0027408599853515625,
0.044342041015625,
0.0291595458984375,
-0.0198822021484375,
0.0219268798828125,
0.0041656494140625,
0.035247802734375,
-0.039642333984375,
0.03643798828125,
-0.0362548828125,
-0.0117340087890625,
-0.033233642578125,
0.0082244873046875,
-0.023223876953125,
-0.039093017578125,
-0.02288818359375,
-0.031982421875,
-0.0157470703125,
0.0002605915069580078,
0.0933837890625,
0.00931549072265625,
-0.0232391357421875,
-0.016937255859375,
-0.02911376953125,
0.04998779296875,
-0.08013916015625,
0.0268402099609375,
0.036346435546875,
0.0176849365234375,
-0.01331329345703125,
-0.038055419921875,
-0.0518798828125,
-0.023681640625,
0.00568389892578125,
0.01071929931640625,
-0.0310516357421875,
-0.0165557861328125,
0.0188140869140625,
0.0318603515625,
-0.03997802734375,
0.0218658447265625,
-0.0292510986328125,
-0.012054443359375,
0.046844482421875,
0.00969696044921875,
0.01543426513671875,
-0.025787353515625,
-0.0216827392578125,
-0.0071258544921875,
-0.053558349609375,
0.0184783935546875,
0.042266845703125,
0.02325439453125,
-0.033233642578125,
0.057830810546875,
-0.0120391845703125,
0.057098388671875,
-0.0016069412231445312,
-0.0250244140625,
0.0435791015625,
-0.025482177734375,
-0.030975341796875,
-0.011444091796875,
0.07684326171875,
0.0298004150390625,
0.00540924072265625,
0.00975799560546875,
-0.0160064697265625,
-0.00974273681640625,
-0.01476287841796875,
-0.061279296875,
0.0008726119995117188,
0.01001739501953125,
-0.0401611328125,
-0.0229339599609375,
0.002094268798828125,
-0.052734375,
-0.0194244384765625,
-0.006381988525390625,
0.0219879150390625,
-0.035369873046875,
-0.0369873046875,
0.0125732421875,
0.02850341796875,
0.035888671875,
0.0010318756103515625,
-0.06060791015625,
0.015045166015625,
0.0323486328125,
0.0728759765625,
0.004581451416015625,
-0.023712158203125,
-0.0268707275390625,
0.017974853515625,
-0.02154541015625,
0.042144775390625,
-0.0222930908203125,
-0.036346435546875,
-0.00820159912109375,
0.007076263427734375,
-0.01248931884765625,
-0.03375244140625,
0.045623779296875,
-0.0186004638671875,
0.01351165771484375,
-0.00946044921875,
-0.0224609375,
-0.02862548828125,
0.0117645263671875,
-0.042144775390625,
0.0966796875,
0.01444244384765625,
-0.05047607421875,
0.0166778564453125,
-0.053314208984375,
-0.0221710205078125,
-0.026123046875,
0.0069427490234375,
-0.053802490234375,
-0.01444244384765625,
0.03887939453125,
0.048828125,
-0.034576416015625,
0.0246429443359375,
-0.036102294921875,
-0.0176849365234375,
0.0176239013671875,
-0.01482391357421875,
0.08538818359375,
0.01473236083984375,
-0.0323486328125,
0.0118560791015625,
-0.06494140625,
-0.0123138427734375,
0.0462646484375,
-0.025909423828125,
-0.0013113021850585938,
-0.0206756591796875,
-0.0099639892578125,
0.0246429443359375,
0.016998291015625,
-0.029632568359375,
0.03179931640625,
-0.043060302734375,
0.03271484375,
0.06640625,
-0.00006490945816040039,
0.0176849365234375,
-0.025177001953125,
0.0423583984375,
0.0179901123046875,
0.02630615234375,
0.0038814544677734375,
-0.058441162109375,
-0.06439208984375,
-0.035675048828125,
0.01873779296875,
0.0350341796875,
-0.041839599609375,
0.045806884765625,
-0.013641357421875,
-0.05572509765625,
-0.04571533203125,
0.0140380859375,
0.03033447265625,
0.03326416015625,
0.0323486328125,
-0.0187530517578125,
-0.0367431640625,
-0.0750732421875,
0.0146636962890625,
-0.00827789306640625,
0.009765625,
0.0279083251953125,
0.053131103515625,
-0.025726318359375,
0.044189453125,
-0.045379638671875,
-0.0246124267578125,
-0.018585205078125,
-0.0018224716186523438,
0.037689208984375,
0.039642333984375,
0.056396484375,
-0.0328369140625,
-0.0244293212890625,
0.008056640625,
-0.05859375,
-0.005680084228515625,
0.0127410888671875,
-0.019073486328125,
0.0377197265625,
0.0217132568359375,
-0.06475830078125,
0.032684326171875,
0.0478515625,
-0.017669677734375,
0.031494140625,
-0.01062774658203125,
-0.0098114013671875,
-0.08660888671875,
0.0204010009765625,
-0.0034580230712890625,
-0.00016582012176513672,
-0.034210205078125,
-0.0019779205322265625,
0.004608154296875,
0.00846099853515625,
-0.046875,
0.0565185546875,
-0.03875732421875,
-0.010406494140625,
0.00115966796875,
0.0177764892578125,
-0.007221221923828125,
0.048858642578125,
-0.0235595703125,
0.052642822265625,
0.0516357421875,
-0.046722412109375,
0.034027099609375,
0.0227203369140625,
-0.033111572265625,
0.020233154296875,
-0.05682373046875,
0.01178741455078125,
0.00446319580078125,
0.030853271484375,
-0.062103271484375,
-0.020263671875,
0.038330078125,
-0.036895751953125,
0.030853271484375,
-0.0036411285400390625,
-0.054840087890625,
-0.050140380859375,
-0.0310211181640625,
0.01371002197265625,
0.041351318359375,
-0.037567138671875,
0.0404052734375,
0.02862548828125,
-0.001552581787109375,
-0.05743408203125,
-0.057586669921875,
-0.0113525390625,
-0.02410888671875,
-0.049102783203125,
0.01096343994140625,
-0.0016508102416992188,
-0.005008697509765625,
-0.0008831024169921875,
0.00952911376953125,
0.0139007568359375,
0.0117950439453125,
0.02252197265625,
0.031585693359375,
-0.0157928466796875,
-0.00860595703125,
-0.012298583984375,
-0.00366973876953125,
-0.0033473968505859375,
-0.005863189697265625,
0.06488037109375,
-0.0201568603515625,
-0.02508544921875,
-0.037078857421875,
-0.00069427490234375,
0.035400390625,
-0.026092529296875,
0.07568359375,
0.0599365234375,
-0.028411865234375,
0.01015472412109375,
-0.042572021484375,
0.008087158203125,
-0.0318603515625,
0.0209808349609375,
-0.03692626953125,
-0.049957275390625,
0.06103515625,
0.0264892578125,
-0.0030078887939453125,
0.0462646484375,
0.07080078125,
0.01328277587890625,
0.07440185546875,
0.03643798828125,
-0.011993408203125,
0.0284881591796875,
-0.041412353515625,
0.01107025146484375,
-0.0780029296875,
-0.03631591796875,
-0.033111572265625,
-0.012969970703125,
-0.049896240234375,
-0.0430908203125,
0.0034084320068359375,
0.021026611328125,
-0.03558349609375,
0.0298614501953125,
-0.0262451171875,
0.0166778564453125,
0.039398193359375,
0.01177978515625,
0.003936767578125,
-0.01427459716796875,
-0.027923583984375,
0.00885009765625,
-0.03460693359375,
-0.038970947265625,
0.089111328125,
0.045379638671875,
0.045806884765625,
0.0224761962890625,
0.046356201171875,
-0.0121917724609375,
0.0299835205078125,
-0.041229248046875,
0.04901123046875,
0.003551483154296875,
-0.043426513671875,
-0.00830841064453125,
-0.0266265869140625,
-0.07073974609375,
0.0209197998046875,
-0.00630950927734375,
-0.062042236328125,
0.007610321044921875,
-0.0036411285400390625,
-0.019561767578125,
0.0350341796875,
-0.04193115234375,
0.05133056640625,
-0.01068115234375,
-0.032196044921875,
0.006183624267578125,
-0.048614501953125,
0.049530029296875,
-0.00539398193359375,
0.019561767578125,
-0.021636962890625,
-0.017608642578125,
0.050537109375,
-0.0310516357421875,
0.0615234375,
-0.021514892578125,
-0.01451873779296875,
0.0293731689453125,
-0.002727508544921875,
0.045440673828125,
-0.006130218505859375,
-0.0194091796875,
0.03228759765625,
-0.02984619140625,
-0.034393310546875,
-0.030914306640625,
0.05938720703125,
-0.0772705078125,
-0.04730224609375,
-0.036346435546875,
-0.031646728515625,
-0.004695892333984375,
0.0148468017578125,
0.0143585205078125,
0.023956298828125,
0.002346038818359375,
0.0225372314453125,
0.0241241455078125,
-0.03741455078125,
0.03924560546875,
0.0231170654296875,
-0.0457763671875,
-0.042510986328125,
0.066650390625,
0.0125274658203125,
0.0218353271484375,
0.00933074951171875,
0.0098724365234375,
-0.031951904296875,
-0.0416259765625,
-0.04290771484375,
0.04302978515625,
-0.060943603515625,
-0.02728271484375,
-0.051849365234375,
-0.02484130859375,
-0.037689208984375,
0.003742218017578125,
-0.022003173828125,
-0.0272369384765625,
-0.032989501953125,
-0.016387939453125,
0.031646728515625,
0.048553466796875,
-0.0146484375,
0.0246429443359375,
-0.037933349609375,
0.01507568359375,
0.0140228271484375,
0.027984619140625,
0.001247406005859375,
-0.05987548828125,
-0.0216522216796875,
0.0040740966796875,
-0.03759765625,
-0.0562744140625,
0.0435791015625,
-0.002811431884765625,
0.052825927734375,
0.021453857421875,
-0.019134521484375,
0.0556640625,
-0.016937255859375,
0.0723876953125,
0.0167999267578125,
-0.06939697265625,
0.0433349609375,
-0.036529541015625,
0.0283050537109375,
0.0102386474609375,
0.0291290283203125,
-0.01070404052734375,
-0.037689208984375,
-0.0521240234375,
-0.06884765625,
0.058685302734375,
0.031402587890625,
0.005218505859375,
-0.0014181137084960938,
0.0201568603515625,
-0.0008931159973144531,
0.01042938232421875,
-0.0865478515625,
-0.03253173828125,
-0.0277099609375,
-0.0090179443359375,
-0.004459381103515625,
-0.0127410888671875,
-0.0060577392578125,
-0.0244598388671875,
0.07269287109375,
-0.0036296844482421875,
0.03411865234375,
-0.003124237060546875,
-0.010284423828125,
-0.00959014892578125,
0.0028228759765625,
0.06585693359375,
0.051239013671875,
-0.021942138671875,
-0.00983428955078125,
0.02642822265625,
-0.035980224609375,
0.0046844482421875,
0.0221099853515625,
-0.01221466064453125,
-0.0081329345703125,
0.0234375,
0.0772705078125,
0.016876220703125,
-0.02423095703125,
0.044189453125,
0.002498626708984375,
-0.01511383056640625,
-0.01140594482421875,
0.0014867782592773438,
0.0266571044921875,
0.01331329345703125,
0.01554107666015625,
-0.00687408447265625,
-0.01277923583984375,
-0.03729248046875,
-0.010223388671875,
0.02984619140625,
-0.0026416778564453125,
-0.0165557861328125,
0.06695556640625,
0.0074615478515625,
-0.00992584228515625,
0.029876708984375,
-0.00801849365234375,
-0.03240966796875,
0.0555419921875,
0.041046142578125,
0.058929443359375,
-0.027435302734375,
0.00360107421875,
0.053253173828125,
0.0302734375,
-0.002826690673828125,
0.0150299072265625,
0.015899658203125,
-0.04071044921875,
-0.025146484375,
-0.05682373046875,
-0.008331298828125,
0.025665283203125,
-0.050384521484375,
0.04364013671875,
-0.038543701171875,
-0.01522064208984375,
-0.007701873779296875,
0.00396728515625,
-0.0460205078125,
0.0098876953125,
0.0191650390625,
0.058441162109375,
-0.0533447265625,
0.08062744140625,
0.03814697265625,
-0.06414794921875,
-0.0687255859375,
-0.0074462890625,
0.0067901611328125,
-0.080322265625,
0.05169677734375,
0.023590087890625,
-0.005523681640625,
-0.001667022705078125,
-0.040985107421875,
-0.07305908203125,
0.1009521484375,
0.03057861328125,
-0.033294677734375,
0.00583648681640625,
-0.0032482147216796875,
0.03289794921875,
-0.0244903564453125,
0.039337158203125,
0.0447998046875,
0.03277587890625,
0.00594329833984375,
-0.08013916015625,
0.01401519775390625,
-0.0179443359375,
0.01531982421875,
0.005077362060546875,
-0.0797119140625,
0.0826416015625,
-0.018646240234375,
-0.00739288330078125,
0.00951385498046875,
0.05755615234375,
0.03997802734375,
0.0029754638671875,
0.0276641845703125,
0.04498291015625,
0.058319091796875,
-0.01105499267578125,
0.09283447265625,
-0.02252197265625,
0.035919189453125,
0.06561279296875,
-0.01080322265625,
0.06317138671875,
0.016815185546875,
-0.023773193359375,
0.040496826171875,
0.059234619140625,
-0.00962066650390625,
0.028076171875,
0.0161285400390625,
-0.00211334228515625,
-0.0012598037719726562,
-0.00667572021484375,
-0.0576171875,
0.041534423828125,
0.0179901123046875,
-0.0276031494140625,
0.0030765533447265625,
0.0007834434509277344,
0.0289154052734375,
-0.00930023193359375,
-0.0206298828125,
0.043670654296875,
0.011260986328125,
-0.031402587890625,
0.0780029296875,
0.0043487548828125,
0.053070068359375,
-0.046356201171875,
0.0130462646484375,
-0.028289794921875,
0.01259613037109375,
-0.01448822021484375,
-0.0626220703125,
0.01033782958984375,
0.0022182464599609375,
0.0021915435791015625,
-0.0059814453125,
0.032867431640625,
-0.0169830322265625,
-0.045440673828125,
0.0433349609375,
0.02685546875,
0.020721435546875,
0.02056884765625,
-0.0677490234375,
0.0294342041015625,
0.0029506683349609375,
-0.0511474609375,
0.022735595703125,
0.0225982666015625,
0.00839996337890625,
0.064453125,
0.053741455078125,
0.01349639892578125,
0.01248931884765625,
0.0140838623046875,
0.08184814453125,
-0.0576171875,
-0.0229644775390625,
-0.0762939453125,
0.041778564453125,
-0.007312774658203125,
-0.0267333984375,
0.061798095703125,
0.058685302734375,
0.06011962890625,
0.0026531219482421875,
0.04693603515625,
-0.00970458984375,
0.032928466796875,
-0.0511474609375,
0.04815673828125,
-0.04742431640625,
0.019775390625,
-0.028717041015625,
-0.08331298828125,
-0.0289764404296875,
0.060821533203125,
-0.0272216796875,
0.018218994140625,
0.045440673828125,
0.0692138671875,
0.005828857421875,
-0.0066070556640625,
0.01995849609375,
0.050384521484375,
0.0308685302734375,
0.05303955078125,
0.051055908203125,
-0.046478271484375,
0.05120849609375,
-0.0311126708984375,
-0.0129547119140625,
-0.0247039794921875,
-0.05987548828125,
-0.076904296875,
-0.041961669921875,
-0.0172576904296875,
-0.04156494140625,
-0.01462554931640625,
0.07110595703125,
0.04315185546875,
-0.05474853515625,
-0.02410888671875,
0.0183563232421875,
0.00687408447265625,
-0.01113128662109375,
-0.01433563232421875,
0.039215087890625,
0.005748748779296875,
-0.0684814453125,
0.0166168212890625,
0.01220703125,
0.0223236083984375,
-0.0290374755859375,
-0.0157012939453125,
-0.0433349609375,
-0.0033702850341796875,
0.048004150390625,
0.01416778564453125,
-0.07098388671875,
-0.0095977783203125,
-0.0033245086669921875,
-0.0224761962890625,
0.010467529296875,
0.0233154296875,
-0.03619384765625,
0.0218658447265625,
0.0164947509765625,
0.0267486572265625,
0.054168701171875,
-0.0023517608642578125,
0.0202789306640625,
-0.03546142578125,
0.035369873046875,
0.0051116943359375,
0.036956787109375,
0.0369873046875,
-0.023681640625,
0.052642822265625,
0.0167236328125,
-0.044708251953125,
-0.0765380859375,
-0.004669189453125,
-0.08709716796875,
-0.01276397705078125,
0.1033935546875,
-0.0123138427734375,
-0.036224365234375,
0.01029205322265625,
-0.0263519287109375,
0.028564453125,
-0.032867431640625,
0.040313720703125,
0.05523681640625,
-0.008148193359375,
-0.01224517822265625,
-0.0596923828125,
0.01535797119140625,
0.011260986328125,
-0.066650390625,
-0.0019388198852539062,
0.023468017578125,
0.0225372314453125,
0.004962921142578125,
0.050140380859375,
-0.0180511474609375,
0.00748443603515625,
-0.005764007568359375,
0.01444244384765625,
-0.0223541259765625,
0.006137847900390625,
-0.0233917236328125,
0.0004718303680419922,
-0.003734588623046875,
-0.01355743408203125
]
] |
ai-forever/mGPT-1.3B-uzbek | 2023-08-11T08:03:02.000Z | [
"transformers",
"pytorch",
"gpt2",
"text-generation",
"gpt3",
"mgpt",
"uz",
"en",
"ru",
"license:mit",
"endpoints_compatible",
"text-generation-inference",
"region:us"
] | text-generation | ai-forever | null | null | ai-forever/mGPT-1.3B-uzbek | 3 | 5,913 | transformers | 2023-08-10T05:12:29 | ---
language:
- uz
- en
- ru
license: mit
tags:
- gpt3
- transformers
- mgpt
---
# 🇺🇿 Uzbek mGPT 1.3B
Language model for Uzbek. Model has 1.3B parameters as you can guess from it's name.
Uzbek belongs to Turkic language family. It's a very rhythmic language with approximately 32 million speakers. Here are some facts about it:
1. The official language of Uzbekistan.
2. It transitioned from the Cyrillic script to the Latin script after Uzbekistans independence, but the Cyrillic script is still in use among older generations.
3. Historically, it was influenced by Persian and Arabic due to trade and Islamic scholarly traditions.
## Technical details
It's one of the models derived from the base [mGPT-XL (1.3B)](https://huggingface.co/ai-forever/mGPT) model (see the list below) which was originally trained on the 61 languages from 25 language families using Wikipedia and C4 corpus.
We've found additional data for 23 languages most of which are considered as minor and decided to further tune the base model. **Uzbek mGPT 1.3B** was trained for another 50000 steps with batch_size=4 and context window of **2048** tokens on 1 A100.
Final perplexity for this model on validation is **6.84**.
_Chart of the training loss and perplexity:_
## Other mGPT-1.3B models
- [🇦🇲 mGPT-1.3B Armenian](https://huggingface.co/ai-forever/mGPT-1.3B-armenian)
- [🇦🇿 mGPT-1.3B Azerbaijan](https://huggingface.co/ai-forever/mGPT-1.3B-azerbaijan)
- [🍯 mGPT-1.3B Bashkir](https://huggingface.co/ai-forever/mGPT-1.3B-bashkir)
- [🇧🇾 mGPT-1.3B Belorussian](https://huggingface.co/ai-forever/mGPT-1.3B-belorussian)
- [🇧🇬 mGPT-1.3B Bulgarian](https://huggingface.co/ai-forever/mGPT-1.3B-bulgarian)
- [🌞 mGPT-1.3B Buryat](https://huggingface.co/ai-forever/mGPT-1.3B-buryat)
- [🌳 mGPT-1.3B Chuvash](https://huggingface.co/ai-forever/mGPT-1.3B-chuvash)
- [🇬🇪 mGPT-1.3B Georgian](https://huggingface.co/ai-forever/mGPT-1.3B-georgian)
- [🌸 mGPT-1.3B Kalmyk](https://huggingface.co/ai-forever/mGPT-1.3B-kalmyk)
- [🇰🇿 mGPT-1.3B Kazakh](https://huggingface.co/ai-forever/mGPT-1.3B-kazakh)
- [🇰🇬 mGPT-1.3B Kirgiz](https://huggingface.co/ai-forever/mGPT-1.3B-kirgiz)
- [🐻 mGPT-1.3B Mari](https://huggingface.co/ai-forever/mGPT-1.3B-mari)
- [🇲🇳 mGPT-1.3B Mongol](https://huggingface.co/ai-forever/mGPT-1.3B-mongol)
- [🐆 mGPT-1.3B Ossetian](https://huggingface.co/ai-forever/mGPT-1.3B-ossetian)
- [🇮🇷 mGPT-1.3B Persian](https://huggingface.co/ai-forever/mGPT-1.3B-persian)
- [🇷🇴 mGPT-1.3B Romanian](https://huggingface.co/ai-forever/mGPT-1.3B-romanian)
- [🇹🇯 mGPT-1.3B Tajik](https://huggingface.co/ai-forever/mGPT-1.3B-tajik)
- [☕ mGPT-1.3B Tatar](https://huggingface.co/ai-forever/mGPT-1.3B-tatar)
- [🇹🇲 mGPT-1.3B Turkmen](https://huggingface.co/ai-forever/mGPT-1.3B-turkmen)
- [🐎 mGPT-1.3B Tuvan](https://huggingface.co/ai-forever/mGPT-1.3B-tuvan)
- [🇺🇦 mGPT-1.3B Ukranian](https://huggingface.co/ai-forever/mGPT-1.3B-ukranian)
- [💎 mGPT-1.3B Yakut](https://huggingface.co/ai-forever/mGPT-1.3B-yakut)
## Feedback
If you'll found a bug of have additional data to train model on your language — please, give us feedback.
Model will be improved over time. Stay tuned!
| 3,187 | [
[
-0.03607177734375,
-0.02508544921875,
0.023651123046875,
0.035888671875,
-0.03338623046875,
0.01512908935546875,
-0.00894927978515625,
-0.04425048828125,
0.02880859375,
0.01496124267578125,
-0.051788330078125,
-0.052215576171875,
-0.038604736328125,
0.0086822509765625,
0.0136260986328125,
0.0701904296875,
-0.01309967041015625,
0.025970458984375,
0.028656005859375,
-0.04150390625,
-0.0301666259765625,
-0.0511474609375,
-0.02557373046875,
-0.047607421875,
0.0301666259765625,
0.0360107421875,
0.04681396484375,
0.03302001953125,
0.022552490234375,
0.020965576171875,
-0.00891876220703125,
-0.01236724853515625,
-0.0011081695556640625,
-0.00909423828125,
0.005977630615234375,
-0.035125732421875,
-0.054443359375,
-0.0023651123046875,
0.040924072265625,
0.0400390625,
-0.0071258544921875,
0.0235137939453125,
-0.00838470458984375,
0.054168701171875,
-0.01377105712890625,
0.01477813720703125,
-0.0198516845703125,
0.01369476318359375,
-0.024505615234375,
0.00728607177734375,
-0.01113128662109375,
-0.0384521484375,
0.0033931732177734375,
-0.0252532958984375,
-0.00760650634765625,
0.0011348724365234375,
0.09259033203125,
-0.04498291015625,
-0.0177001953125,
-0.0031986236572265625,
-0.0277557373046875,
0.06475830078125,
-0.036285400390625,
0.0233917236328125,
0.0305938720703125,
0.0116424560546875,
-0.0276641845703125,
-0.016845703125,
-0.036529541015625,
-0.0198211669921875,
-0.0230865478515625,
0.023834228515625,
-0.01050567626953125,
-0.0054931640625,
0.0189056396484375,
0.044525146484375,
-0.06036376953125,
-0.02117919921875,
-0.0218353271484375,
-0.01074981689453125,
0.044830322265625,
0.0181732177734375,
0.06787109375,
-0.017974853515625,
-0.01904296875,
0.0088653564453125,
-0.04547119140625,
0.0282745361328125,
0.0289154052734375,
0.0310516357421875,
-0.03778076171875,
0.0576171875,
-0.0302276611328125,
0.04046630859375,
0.01690673828125,
-0.0122528076171875,
0.032318115234375,
-0.0240020751953125,
-0.0139007568359375,
-0.030914306640625,
0.0684814453125,
0.01971435546875,
0.026397705078125,
-0.01000213623046875,
0.0061492919921875,
0.000545501708984375,
-0.0287322998046875,
-0.05706787109375,
-0.012664794921875,
0.00677490234375,
-0.041351318359375,
-0.0028362274169921875,
0.0233154296875,
-0.05487060546875,
-0.0005822181701660156,
-0.004497528076171875,
0.0294647216796875,
-0.058380126953125,
-0.041748046875,
0.0032634735107421875,
-0.005290985107421875,
0.043182373046875,
0.01079559326171875,
-0.09014892578125,
0.01511383056640625,
0.030303955078125,
0.048065185546875,
0.0101776123046875,
-0.03692626953125,
0.005016326904296875,
-0.0036373138427734375,
-0.02459716796875,
0.047821044921875,
-0.036285400390625,
-0.0236358642578125,
-0.0200958251953125,
0.00933837890625,
-0.02337646484375,
-0.010162353515625,
0.02703857421875,
0.00218963623046875,
0.020050048828125,
-0.0294036865234375,
-0.006633758544921875,
-0.0092315673828125,
0.01568603515625,
-0.0355224609375,
0.078857421875,
0.056640625,
-0.06671142578125,
0.050689697265625,
-0.03350830078125,
-0.00603485107421875,
0.00670623779296875,
-0.0035343170166015625,
-0.039581298828125,
-0.005336761474609375,
-0.0016918182373046875,
0.04443359375,
-0.0147247314453125,
-0.012176513671875,
-0.019775390625,
-0.02276611328125,
0.01209259033203125,
0.0237579345703125,
0.0706787109375,
0.01345062255859375,
-0.0238037109375,
0.0113983154296875,
-0.055511474609375,
0.0253753662109375,
0.0160369873046875,
-0.0244598388671875,
-0.0034694671630859375,
-0.040771484375,
0.01800537109375,
0.038818359375,
-0.0012035369873046875,
-0.05023193359375,
0.048004150390625,
-0.0283355712890625,
0.019561767578125,
0.043731689453125,
-0.00484466552734375,
0.029754638671875,
-0.030792236328125,
0.0628662109375,
0.005451202392578125,
0.024749755859375,
0.01512908935546875,
-0.04071044921875,
-0.057159423828125,
-0.043060302734375,
0.0267181396484375,
0.04168701171875,
-0.042205810546875,
0.040924072265625,
0.00010967254638671875,
-0.04949951171875,
-0.04437255859375,
-0.0022258758544921875,
0.037628173828125,
0.003208160400390625,
0.01038360595703125,
-0.0159759521484375,
-0.052398681640625,
-0.06475830078125,
-0.0273895263671875,
-0.040802001953125,
-0.0029125213623046875,
0.017425537109375,
0.04522705078125,
-0.005889892578125,
0.03436279296875,
-0.004901885986328125,
-0.01471710205078125,
-0.0482177734375,
0.003322601318359375,
0.03497314453125,
0.03656005859375,
0.06640625,
-0.044830322265625,
-0.0772705078125,
0.025665283203125,
-0.051666259765625,
0.01434326171875,
-0.00482177734375,
-0.029815673828125,
0.04779052734375,
-0.01055145263671875,
-0.06341552734375,
0.04632568359375,
0.047393798828125,
-0.06658935546875,
0.04888916015625,
-0.001583099365234375,
0.0220184326171875,
-0.09356689453125,
0.01453399658203125,
-0.0087432861328125,
0.00017642974853515625,
-0.04901123046875,
0.01068115234375,
-0.006439208984375,
0.031829833984375,
-0.0264129638671875,
0.056793212890625,
-0.057220458984375,
0.004638671875,
0.0041046142578125,
0.002105712890625,
-0.024139404296875,
0.040252685546875,
-0.0023212432861328125,
0.049835205078125,
0.0654296875,
-0.024749755859375,
0.05255126953125,
0.036712646484375,
-0.0181884765625,
0.0509033203125,
-0.0296173095703125,
-0.0021266937255859375,
-0.0023250579833984375,
0.00766754150390625,
-0.06011962890625,
-0.042327880859375,
0.05279541015625,
-0.06390380859375,
0.033935546875,
-0.032806396484375,
-0.04132080078125,
-0.0511474609375,
-0.033843994140625,
0.00832366943359375,
0.050018310546875,
-0.0328369140625,
0.047393798828125,
0.023101806640625,
-0.00873565673828125,
-0.049591064453125,
-0.054901123046875,
0.0068817138671875,
-0.03948974609375,
-0.055450439453125,
0.00736236572265625,
-0.004245758056640625,
-0.007480621337890625,
-0.0010442733764648438,
-0.0168304443359375,
-0.0233001708984375,
0.005535125732421875,
0.0240020751953125,
0.03216552734375,
-0.0160369873046875,
-0.0195770263671875,
-0.032928466796875,
0.00012552738189697266,
-0.0226898193359375,
-0.0047607421875,
0.055572509765625,
-0.0283966064453125,
0.002597808837890625,
-0.049102783203125,
0.01849365234375,
0.0467529296875,
-0.00179290771484375,
0.0799560546875,
0.07159423828125,
-0.02191162109375,
0.0138397216796875,
-0.0203857421875,
0.009521484375,
-0.02685546875,
0.0028533935546875,
-0.064208984375,
-0.06298828125,
0.0740966796875,
0.02020263671875,
-0.006687164306640625,
0.0577392578125,
0.052398681640625,
0.00846099853515625,
0.0926513671875,
0.0657958984375,
-0.0421142578125,
0.042205810546875,
-0.027496337890625,
-0.0040130615234375,
-0.06488037109375,
-0.040802001953125,
-0.0307159423828125,
-0.00455474853515625,
-0.075927734375,
0.0082855224609375,
0.00026535987854003906,
0.0168304443359375,
-0.01250457763671875,
0.062164306640625,
-0.029754638671875,
0.030059814453125,
0.0275421142578125,
0.00797271728515625,
0.005115509033203125,
-0.004413604736328125,
-0.0194244384765625,
-0.00844573974609375,
-0.03680419921875,
-0.031219482421875,
0.0758056640625,
0.023223876953125,
0.05950927734375,
0.0263519287109375,
0.052215576171875,
0.0017480850219726562,
0.0163726806640625,
-0.0467529296875,
0.040283203125,
0.0110015869140625,
-0.053131103515625,
-0.032470703125,
-0.0213165283203125,
-0.068115234375,
0.0187835693359375,
-0.0196380615234375,
-0.0985107421875,
-0.019744873046875,
-0.00004792213439941406,
-0.0240936279296875,
0.01137542724609375,
-0.025299072265625,
0.05389404296875,
0.00408172607421875,
-0.0282745361328125,
0.00650787353515625,
-0.06463623046875,
0.0413818359375,
0.0214691162109375,
-0.0150299072265625,
-0.030792236328125,
0.03521728515625,
0.03192138671875,
-0.038055419921875,
0.0235748291015625,
-0.01342010498046875,
0.0229034423828125,
0.0220184326171875,
0.004669189453125,
0.025909423828125,
0.0036830902099609375,
0.004711151123046875,
0.0012683868408203125,
0.0135040283203125,
-0.054351806640625,
-0.0260467529296875,
0.0491943359375,
-0.0684814453125,
-0.024383544921875,
-0.0684814453125,
-0.01245880126953125,
-0.0028514862060546875,
0.0175323486328125,
0.0411376953125,
0.0267181396484375,
-0.013671875,
0.01041412353515625,
0.0211639404296875,
-0.00904083251953125,
0.01508331298828125,
0.02996826171875,
-0.0272216796875,
-0.0229949951171875,
0.06475830078125,
-0.001621246337890625,
0.00879669189453125,
0.006134033203125,
0.0191650390625,
-0.0296173095703125,
-0.033660888671875,
-0.047027587890625,
0.047393798828125,
-0.0251007080078125,
-0.0052947998046875,
-0.045379638671875,
-0.0203704833984375,
-0.055938720703125,
0.0004336833953857422,
-0.015716552734375,
-0.035888671875,
-0.0130157470703125,
-0.00287628173828125,
0.038177490234375,
0.0672607421875,
-0.0180816650390625,
0.0301666259765625,
-0.048248291015625,
0.032318115234375,
0.00885009765625,
0.043121337890625,
-0.04156494140625,
-0.044891357421875,
-0.0172576904296875,
-0.005672454833984375,
-0.0101470947265625,
-0.06951904296875,
0.04095458984375,
0.013580322265625,
0.0498046875,
0.0187225341796875,
-0.016510009765625,
0.060455322265625,
-0.0268707275390625,
0.060791015625,
0.02191162109375,
-0.06475830078125,
0.018310546875,
-0.05224609375,
0.01593017578125,
0.056396484375,
0.024505615234375,
-0.07208251953125,
-0.042694091796875,
-0.044036865234375,
-0.055328369140625,
0.0675048828125,
0.031463623046875,
0.02001953125,
0.008056640625,
0.0292816162109375,
-0.0020809173583984375,
0.007022857666015625,
-0.06353759765625,
-0.0491943359375,
0.00360870361328125,
-0.026947021484375,
0.029815673828125,
-0.029510498046875,
0.005992889404296875,
-0.039306640625,
0.055938720703125,
0.0068511962890625,
0.038299560546875,
0.01053619384765625,
-0.0037403106689453125,
-0.0234832763671875,
0.00780487060546875,
0.06256103515625,
0.05743408203125,
-0.0192413330078125,
-0.031585693359375,
0.013153076171875,
-0.054840087890625,
0.0079345703125,
-0.0047454833984375,
-0.006214141845703125,
0.0133056640625,
0.0113372802734375,
0.049072265625,
-0.00479888916015625,
-0.01398468017578125,
0.038604736328125,
-0.029815673828125,
-0.01543426513671875,
-0.05206298828125,
-0.0014848709106445312,
0.00022482872009277344,
0.01751708984375,
0.01358795166015625,
-0.01092529296875,
-0.007236480712890625,
-0.07421875,
0.01403045654296875,
0.050994873046875,
-0.0184783935546875,
-0.056549072265625,
0.0311126708984375,
0.016265869140625,
-0.0014629364013671875,
0.027130126953125,
-0.0239715576171875,
-0.06890869140625,
0.041961669921875,
0.0313720703125,
0.06085205078125,
-0.048065185546875,
0.0274200439453125,
0.04815673828125,
0.0191497802734375,
-0.003345489501953125,
0.05511474609375,
0.0195770263671875,
-0.039276123046875,
-0.0112152099609375,
-0.056610107421875,
-0.017486572265625,
0.033294677734375,
-0.045654296875,
0.031341552734375,
-0.040557861328125,
-0.01995849609375,
-0.00696563720703125,
0.034332275390625,
-0.04571533203125,
0.01568603515625,
0.01812744140625,
0.062408447265625,
-0.06915283203125,
0.070556640625,
0.05511474609375,
-0.0256500244140625,
-0.06787109375,
-0.007320404052734375,
0.00888824462890625,
-0.055084228515625,
0.044281005859375,
0.01329803466796875,
0.005069732666015625,
-0.0054779052734375,
-0.037567138671875,
-0.07476806640625,
0.07635498046875,
-0.0013065338134765625,
0.0027141571044921875,
0.01508331298828125,
0.0011854171752929688,
0.04754638671875,
-0.021636962890625,
0.0201416015625,
0.054443359375,
0.043121337890625,
0.014984130859375,
-0.0794677734375,
0.00955963134765625,
-0.03192138671875,
-0.0196380615234375,
0.0258941650390625,
-0.08319091796875,
0.080322265625,
-0.019012451171875,
-0.02105712890625,
0.0555419921875,
0.048248291015625,
0.0276641845703125,
-0.005580902099609375,
0.016693115234375,
0.0259246826171875,
0.0227508544921875,
-0.02349853515625,
0.05450439453125,
-0.016845703125,
0.031646728515625,
0.0439453125,
0.0050048828125,
0.0261993408203125,
0.0286712646484375,
-0.037384033203125,
0.03851318359375,
0.055816650390625,
-0.017913818359375,
0.02215576171875,
0.0179443359375,
-0.0362548828125,
-0.0180206298828125,
-0.0111541748046875,
-0.0283050537109375,
0.055023193359375,
-0.003448486328125,
-0.0139312744140625,
0.0104522705078125,
-0.0159759521484375,
0.04852294921875,
-0.00966644287109375,
-0.0215606689453125,
0.0400390625,
-0.00977325439453125,
-0.041961669921875,
0.017547607421875,
0.006877899169921875,
0.052032470703125,
-0.047607421875,
-0.00948333740234375,
-0.0234222412109375,
0.017120361328125,
-0.0084686279296875,
-0.06085205078125,
0.00724029541015625,
-0.002681732177734375,
-0.0004220008850097656,
-0.0016422271728515625,
0.047607421875,
-0.04815673828125,
-0.04498291015625,
0.01617431640625,
0.025115966796875,
0.0250091552734375,
0.0022602081298828125,
-0.061676025390625,
0.0000903010368347168,
0.0003063678741455078,
-0.0489501953125,
0.039093017578125,
0.033294677734375,
-0.00433349609375,
0.056243896484375,
0.04046630859375,
0.0399169921875,
0.0012044906616210938,
-0.00484466552734375,
0.063720703125,
-0.058197021484375,
-0.0189208984375,
-0.0728759765625,
0.040924072265625,
0.00435638427734375,
-0.0309295654296875,
0.072265625,
0.0169677734375,
0.046478271484375,
-0.00905609130859375,
0.058624267578125,
-0.0275115966796875,
0.047393798828125,
-0.03662109375,
0.07440185546875,
-0.0384521484375,
-0.040985107421875,
-0.03631591796875,
-0.052825927734375,
-0.033660888671875,
0.061279296875,
-0.00008291006088256836,
0.02020263671875,
0.038818359375,
0.017303466796875,
0.022125244140625,
-0.0240020751953125,
0.03375244140625,
0.0108489990234375,
-0.0069427490234375,
0.063232421875,
0.062408447265625,
-0.0379638671875,
0.026336669921875,
-0.03125,
-0.0126495361328125,
-0.0276641845703125,
-0.0439453125,
-0.09283447265625,
-0.035858154296875,
-0.00977325439453125,
-0.04022216796875,
-0.0133819580078125,
0.09765625,
0.04278564453125,
-0.0614013671875,
-0.01467132568359375,
0.0153350830078125,
0.00829315185546875,
0.0036716461181640625,
-0.016357421875,
0.013519287109375,
0.019378662109375,
-0.05615234375,
-0.0010128021240234375,
0.0153350830078125,
0.036102294921875,
0.00453948974609375,
-0.007781982421875,
-0.0079345703125,
-0.005191802978515625,
0.036102294921875,
0.056793212890625,
-0.0572509765625,
-0.04443359375,
0.0166473388671875,
-0.032440185546875,
0.0066070556640625,
0.0295562744140625,
-0.02349853515625,
0.005340576171875,
0.042633056640625,
-0.0185394287109375,
0.032989501953125,
0.0191192626953125,
0.03289794921875,
-0.0266571044921875,
0.032379150390625,
0.004825592041015625,
0.028778076171875,
0.01861572265625,
-0.01995849609375,
0.046295166015625,
0.032684326171875,
-0.041961669921875,
-0.061279296875,
0.01922607421875,
-0.082763671875,
0.00516510009765625,
0.08551025390625,
-0.0236358642578125,
-0.03692626953125,
-0.0117645263671875,
-0.0306396484375,
0.025665283203125,
-0.0266571044921875,
0.05584716796875,
0.0677490234375,
-0.0273590087890625,
-0.0052947998046875,
-0.03607177734375,
0.05072021484375,
0.01467132568359375,
-0.0518798828125,
0.0086517333984375,
0.018341064453125,
0.031890869140625,
-0.002506256103515625,
0.057830810546875,
-0.01375579833984375,
0.02545166015625,
-0.005950927734375,
0.030670166015625,
0.00811004638671875,
-0.0277862548828125,
-0.0244140625,
-0.04119873046875,
0.0132904052734375,
-0.023406982421875
]
] |
TheBloke/koala-13B-HF | 2023-06-05T00:09:42.000Z | [
"transformers",
"pytorch",
"llama",
"text-generation",
"koala",
"ShareGPT",
"gptq",
"dataset:RyokoAI/ShareGPT52K",
"dataset:Hello-SimpleAI/HC3",
"license:other",
"endpoints_compatible",
"has_space",
"text-generation-inference",
"region:us"
] | text-generation | TheBloke | null | null | TheBloke/koala-13B-HF | 40 | 5,912 | transformers | 2023-04-07T21:12:27 | ---
license: other
library_name: transformers
pipeline_tag: text-generation
datasets:
- RyokoAI/ShareGPT52K
- Hello-SimpleAI/HC3
tags:
- koala
- ShareGPT
- llama
- gptq
---
<!-- header start -->
<div style="width: 100%;">
<img src="https://i.imgur.com/EBdldam.jpg" alt="TheBlokeAI" style="width: 100%; min-width: 400px; display: block; margin: auto;">
</div>
<div style="display: flex; justify-content: space-between; width: 100%;">
<div style="display: flex; flex-direction: column; align-items: flex-start;">
<p><a href="https://discord.gg/Jq4vkcDakD">Chat & support: my new Discord server</a></p>
</div>
<div style="display: flex; flex-direction: column; align-items: flex-end;">
<p><a href="https://www.patreon.com/TheBlokeAI">Want to contribute? TheBloke's Patreon page</a></p>
</div>
</div>
<!-- header end -->
# Koala: A Dialogue Model for Academic Research
This repo contains the weights of the Koala 13B model produced at Berkeley. It is the result of combining the diffs from https://huggingface.co/young-geng/koala with the original Llama 13B model.
This version has then been converted to HF format.
## My Koala repos
I have the following Koala model repositories available:
**13B models:**
* [Unquantized 13B model in HF format](https://huggingface.co/TheBloke/koala-13B-HF)
* [GPTQ quantized 4bit 13B model in `pt` and `safetensors` formats](https://huggingface.co/TheBloke/koala-13B-GPTQ-4bit-128g)
* [4-bit, 5-bit and 8-bit GGML models for `llama.cpp`](https://huggingface.co/TheBloke/koala-13B-GGML)
**7B models:**
* [Unquantized 7B model in HF format](https://huggingface.co/TheBloke/koala-7B-HF)
* [Unquantized 7B model in GGML format for llama.cpp](https://huggingface.co/TheBloke/koala-7b-ggml-unquantized)
* [GPTQ quantized 4bit 7B model in `pt` and `safetensors` formats](https://huggingface.co/TheBloke/koala-7B-GPTQ-4bit-128g)
* [4-bit, 5-bit and 8-bit GGML models for `llama.cpp`](https://huggingface.co/TheBloke/koala-7B-GGML)
## How the Koala delta weights were merged
The Koala delta weights were merged using the following commands:
```
git clone https://github.com/young-geng/EasyLM
git clone https://huggingface.co/TheBloke/llama-13b
mkdir koala_diffs && cd koala_diffs && wget https://huggingface.co/young-geng/koala/resolve/main/koala_13b_diff_v2
cd EasyLM
PYTHON_PATH="${PWD}:$PYTHONPATH" python \
-m EasyLM.models.llama.convert_torch_to_easylm \
--checkpoint_dir=/content/llama-13b \
--output_file=/content/llama-13b-LM \
--streaming=True
PYTHON_PATH="${PWD}:$PYTHONPATH" python \
-m EasyLM.scripts.diff_checkpoint --recover_diff=True \
--load_base_checkpoint='params::/content/llama-13b-LM' \
--load_target_checkpoint='params::/content/koala_diffs/koala_13b_diff_v2' \
--output_file=/content/koala_13b.diff.weights \
--streaming=True
PYTHON_PATH="${PWD}:$PYTHONPATH" python \
-m EasyLM.models.llama.convert_easylm_to_hf --model_size=13b \
--output_dir=/content/koala-13B-HF \
--load_checkpoint='params::/content/koala_13b.diff.weights' \
--tokenizer_path=/content/llama-13b/tokenizer.model
```
<!-- footer start -->
## Discord
For further support, and discussions on these models and AI in general, join us at:
[TheBloke AI's Discord server](https://discord.gg/Jq4vkcDakD)
## Thanks, and how to contribute.
Thanks to the [chirper.ai](https://chirper.ai) team!
I've had a lot of people ask if they can contribute. I enjoy providing models and helping people, and would love to be able to spend even more time doing it, as well as expanding into new projects like fine tuning/training.
If you're able and willing to contribute it will be most gratefully received and will help me to keep providing more models, and to start work on new AI projects.
Donaters will get priority support on any and all AI/LLM/model questions and requests, access to a private Discord room, plus other benefits.
* Patreon: https://patreon.com/TheBlokeAI
* Ko-Fi: https://ko-fi.com/TheBlokeAI
**Patreon special mentions**: Aemon Algiz, Dmitriy Samsonov, Nathan LeClaire, Trenton Dambrowitz, Mano Prime, David Flickinger, vamX, Nikolai Manek, senxiiz, Khalefa Al-Ahmad, Illia Dulskyi, Jonathan Leane, Talal Aujan, V. Lukas, Joseph William Delisle, Pyrater, Oscar Rangel, Lone Striker, Luke Pendergrass, Eugene Pentland, Sebastain Graf, Johann-Peter Hartman.
Thank you to all my generous patrons and donaters!
<!-- footer end -->
## Further info
Check out the following links to learn more about the Berkeley Koala model.
* [Blog post](https://bair.berkeley.edu/blog/2023/04/03/koala/)
* [Online demo](https://koala.lmsys.org/)
* [EasyLM: training and serving framework on GitHub](https://github.com/young-geng/EasyLM)
* [Documentation for running Koala locally](https://github.com/young-geng/EasyLM/blob/main/docs/koala.md)
## License
The model weights are intended for academic research only, subject to the
[model License of LLaMA](https://github.com/facebookresearch/llama/blob/main/MODEL_CARD.md),
[Terms of Use of the data generated by OpenAI](https://openai.com/policies/terms-of-use),
and [Privacy Practices of ShareGPT](https://chrome.google.com/webstore/detail/sharegpt-share-your-chatg/daiacboceoaocpibfodeljbdfacokfjb).
Any other usage of the model weights, including but not limited to commercial usage, is strictly prohibited.
| 5,314 | [
[
-0.030914306640625,
-0.05792236328125,
0.0180206298828125,
0.022216796875,
-0.0235443115234375,
-0.01503753662109375,
0.0146636962890625,
-0.04437255859375,
0.022308349609375,
0.0194549560546875,
-0.040863037109375,
-0.0257720947265625,
-0.03448486328125,
0.0188446044921875,
-0.01139068603515625,
0.06805419921875,
0.0184783935546875,
-0.006526947021484375,
-0.0063934326171875,
-0.01305389404296875,
-0.044403076171875,
-0.045928955078125,
-0.049468994140625,
-0.03424072265625,
0.033935546875,
0.002277374267578125,
0.055023193359375,
0.049407958984375,
0.020263671875,
0.0287933349609375,
-0.02362060546875,
0.004180908203125,
-0.034942626953125,
-0.00499725341796875,
0.00875091552734375,
-0.0267181396484375,
-0.043365478515625,
-0.0143890380859375,
0.043792724609375,
0.03448486328125,
-0.02130126953125,
0.022552490234375,
0.0007500648498535156,
0.051910400390625,
-0.037872314453125,
0.0250396728515625,
-0.03192138671875,
-0.0176849365234375,
-0.0024356842041015625,
0.0193634033203125,
-0.019622802734375,
-0.031402587890625,
-0.028594970703125,
-0.07122802734375,
-0.00804901123046875,
0.0029125213623046875,
0.09027099609375,
0.016357421875,
-0.048309326171875,
-0.01090240478515625,
-0.0220794677734375,
0.05169677734375,
-0.080810546875,
0.014739990234375,
0.03338623046875,
0.0198211669921875,
-0.02423095703125,
-0.058380126953125,
-0.037261962890625,
-0.002178192138671875,
-0.0198516845703125,
0.0013570785522460938,
-0.036865234375,
-0.01043701171875,
0.012298583984375,
0.027587890625,
-0.035064697265625,
-0.012420654296875,
-0.03955078125,
-0.024810791015625,
0.06768798828125,
0.00582122802734375,
0.0201263427734375,
-0.0168304443359375,
-0.01305389404296875,
-0.030975341796875,
-0.03173828125,
0.01543426513671875,
0.0263519287109375,
0.0296630859375,
-0.04962158203125,
0.055572509765625,
-0.0341796875,
0.03240966796875,
0.0321044921875,
-0.01013946533203125,
0.024261474609375,
-0.05181884765625,
-0.029144287109375,
0.00965118408203125,
0.0716552734375,
0.02923583984375,
-0.019775390625,
0.023284912109375,
-0.0006551742553710938,
-0.005859375,
-0.0140228271484375,
-0.0836181640625,
-0.021148681640625,
0.02490234375,
-0.04144287109375,
-0.02392578125,
-0.019989013671875,
-0.0538330078125,
-0.0316162109375,
-0.00882720947265625,
0.035888671875,
-0.0307769775390625,
-0.03857421875,
0.004009246826171875,
-0.0127105712890625,
0.023895263671875,
0.030242919921875,
-0.048370361328125,
0.0071563720703125,
0.021270751953125,
0.055816650390625,
0.04095458984375,
-0.0090789794921875,
-0.0382080078125,
0.0038509368896484375,
-0.024444580078125,
0.0302734375,
-0.022918701171875,
-0.04132080078125,
-0.0211029052734375,
0.0235595703125,
0.0017156600952148438,
-0.03363037109375,
0.0223846435546875,
-0.029571533203125,
-0.01123046875,
-0.033599853515625,
-0.025787353515625,
-0.0304412841796875,
0.00907135009765625,
-0.048858642578125,
0.078369140625,
0.0251617431640625,
-0.053009033203125,
0.0083770751953125,
-0.0511474609375,
-0.01486968994140625,
-0.005977630615234375,
0.0206451416015625,
-0.043701171875,
-0.00974273681640625,
0.021392822265625,
0.021148681640625,
-0.01049041748046875,
0.0186614990234375,
-0.055877685546875,
-0.0160064697265625,
0.0254058837890625,
-0.0193023681640625,
0.09661865234375,
0.0297698974609375,
-0.0098114013671875,
0.003032684326171875,
-0.06719970703125,
0.004444122314453125,
0.039794921875,
-0.03839111328125,
-0.01131439208984375,
-0.0168304443359375,
0.004001617431640625,
-0.0054931640625,
0.033599853515625,
-0.0291290283203125,
0.037200927734375,
-0.007904052734375,
0.042999267578125,
0.056793212890625,
-0.002017974853515625,
0.0253143310546875,
-0.044586181640625,
0.022369384765625,
0.006748199462890625,
0.02923583984375,
-0.00621795654296875,
-0.04425048828125,
-0.054107666015625,
-0.039154052734375,
0.019622802734375,
0.0223236083984375,
-0.0333251953125,
0.0300750732421875,
-0.00780487060546875,
-0.058074951171875,
-0.055450439453125,
0.013580322265625,
0.01800537109375,
0.0161590576171875,
0.0223846435546875,
-0.02459716796875,
-0.046630859375,
-0.0609130859375,
0.00435638427734375,
-0.030792236328125,
0.00879669189453125,
0.0304412841796875,
0.05316162109375,
-0.031646728515625,
0.058013916015625,
-0.025970458984375,
-0.00775146484375,
-0.006168365478515625,
0.00705718994140625,
0.01959228515625,
0.057952880859375,
0.05450439453125,
-0.05596923828125,
-0.04083251953125,
0.00664520263671875,
-0.06195068359375,
-0.0107269287109375,
0.003742218017578125,
-0.02923583984375,
0.0166473388671875,
-0.00020420551300048828,
-0.07244873046875,
0.046600341796875,
0.03863525390625,
-0.03863525390625,
0.047515869140625,
-0.0081787109375,
0.0124969482421875,
-0.079345703125,
0.01155853271484375,
-0.00923919677734375,
-0.0268402099609375,
-0.033050537109375,
0.01424407958984375,
-0.00724029541015625,
-0.015899658203125,
-0.03515625,
0.052825927734375,
-0.042694091796875,
-0.00511932373046875,
0.0065460205078125,
-0.0151214599609375,
0.0152435302734375,
0.038787841796875,
-0.0149688720703125,
0.03802490234375,
0.045379638671875,
-0.0367431640625,
0.0289306640625,
0.033660888671875,
-0.01169586181640625,
0.0252227783203125,
-0.0654296875,
0.0244293212890625,
0.00978851318359375,
0.04058837890625,
-0.07598876953125,
-0.039337158203125,
0.05316162109375,
-0.04351806640625,
0.02911376953125,
-0.031158447265625,
-0.036895751953125,
-0.015533447265625,
-0.0284271240234375,
0.045501708984375,
0.053985595703125,
-0.03302001953125,
0.043701171875,
0.01554107666015625,
-0.0142364501953125,
-0.037322998046875,
-0.05413818359375,
-0.01308441162109375,
-0.0226287841796875,
-0.041290283203125,
0.0086517333984375,
-0.02117919921875,
-0.0159759521484375,
0.00836944580078125,
0.004085540771484375,
-0.00925445556640625,
-0.00031304359436035156,
0.037750244140625,
0.0233917236328125,
-0.0176544189453125,
-0.01224517822265625,
0.0082855224609375,
0.0203857421875,
-0.01320648193359375,
0.00042438507080078125,
0.05499267578125,
-0.0316162109375,
-0.0274200439453125,
-0.0634765625,
0.024505615234375,
0.047515869140625,
-0.01114654541015625,
0.06512451171875,
0.041046142578125,
-0.0240631103515625,
0.020538330078125,
-0.044921875,
-0.007442474365234375,
-0.03668212890625,
0.01552581787109375,
-0.00908660888671875,
-0.04412841796875,
0.057769775390625,
0.020294189453125,
0.008880615234375,
0.043243408203125,
0.0380859375,
-0.003063201904296875,
0.09063720703125,
0.045501708984375,
-0.0016002655029296875,
0.038543701171875,
-0.042144775390625,
0.000171661376953125,
-0.06201171875,
-0.0254974365234375,
-0.034942626953125,
-0.0192108154296875,
-0.042266845703125,
-0.024505615234375,
0.0173187255859375,
0.03668212890625,
-0.0543212890625,
0.0247955322265625,
-0.046875,
0.0037593841552734375,
0.0297088623046875,
0.0209197998046875,
0.0209808349609375,
0.002044677734375,
-0.0030460357666015625,
0.0210113525390625,
-0.054901123046875,
-0.036224365234375,
0.08575439453125,
0.0245513916015625,
0.04791259765625,
0.0114288330078125,
0.059478759765625,
0.0185699462890625,
0.033172607421875,
-0.0455322265625,
0.041778564453125,
0.0235748291015625,
-0.0689697265625,
-0.027740478515625,
-0.031280517578125,
-0.0726318359375,
0.02020263671875,
0.00274658203125,
-0.056060791015625,
0.037506103515625,
0.01160430908203125,
-0.025604248046875,
0.0232086181640625,
-0.050567626953125,
0.058319091796875,
-0.00792694091796875,
-0.0271148681640625,
-0.004547119140625,
-0.04534912109375,
0.040191650390625,
-0.006183624267578125,
0.035369873046875,
-0.0158233642578125,
-0.0016107559204101562,
0.055450439453125,
-0.0751953125,
0.08770751953125,
-0.01468658447265625,
0.00009363889694213867,
0.060089111328125,
0.00502777099609375,
0.0435791015625,
0.0033740997314453125,
-0.023468017578125,
0.0394287109375,
0.038604736328125,
-0.040374755859375,
-0.01050567626953125,
0.038726806640625,
-0.0816650390625,
-0.0300140380859375,
-0.03448486328125,
-0.045623779296875,
0.0240020751953125,
0.024169921875,
0.0137176513671875,
0.0198822021484375,
0.00847625732421875,
0.0302581787109375,
0.006397247314453125,
-0.0166168212890625,
0.033782958984375,
0.028839111328125,
-0.033233642578125,
-0.047332763671875,
0.059173583984375,
0.0100555419921875,
0.022247314453125,
0.005889892578125,
0.01264190673828125,
-0.030914306640625,
-0.0124664306640625,
-0.03857421875,
0.037261962890625,
-0.06048583984375,
-0.033416748046875,
-0.03387451171875,
-0.0174713134765625,
-0.0538330078125,
-0.0043182373046875,
-0.031341552734375,
-0.0290374755859375,
-0.041412353515625,
0.01629638671875,
0.0516357421875,
0.044586181640625,
-0.0287933349609375,
0.044158935546875,
-0.057769775390625,
0.0021495819091796875,
0.02008056640625,
0.002239227294921875,
0.0124053955078125,
-0.046722412109375,
-0.0133819580078125,
0.039306640625,
-0.0523681640625,
-0.062347412109375,
0.054901123046875,
0.01009368896484375,
0.0288238525390625,
0.01910400390625,
0.0170135498046875,
0.072021484375,
-0.009735107421875,
0.068115234375,
0.0367431640625,
-0.06402587890625,
0.036895751953125,
-0.026611328125,
0.01837158203125,
0.0253143310546875,
0.02947998046875,
-0.0184478759765625,
-0.0124359130859375,
-0.03424072265625,
-0.0496826171875,
0.050628662109375,
0.032562255859375,
-0.0008502006530761719,
0.007259368896484375,
0.02960205078125,
0.0044403076171875,
0.01544952392578125,
-0.07733154296875,
-0.0187835693359375,
-0.040374755859375,
-0.00559234619140625,
0.0036411285400390625,
0.0008296966552734375,
-0.0188751220703125,
-0.041046142578125,
0.07421875,
-0.0162353515625,
0.03570556640625,
0.0236358642578125,
0.0017213821411132812,
-0.0236053466796875,
0.0150604248046875,
0.0187835693359375,
0.039154052734375,
-0.0187835693359375,
-0.0198822021484375,
0.0098114013671875,
-0.017852783203125,
0.01490020751953125,
0.01264190673828125,
-0.0181884765625,
0.002044677734375,
0.0159912109375,
0.0623779296875,
0.008270263671875,
-0.0406494140625,
0.02520751953125,
0.00010627508163452148,
-0.0226593017578125,
-0.0171661376953125,
0.0002968311309814453,
0.03399658203125,
0.037628173828125,
0.02581787109375,
-0.00835418701171875,
0.0019378662109375,
-0.0587158203125,
-0.01439666748046875,
0.043609619140625,
-0.0013132095336914062,
-0.028594970703125,
0.05828857421875,
-0.00701904296875,
-0.025665283203125,
0.047332763671875,
-0.0178375244140625,
-0.0253753662109375,
0.06719970703125,
0.04638671875,
0.0498046875,
-0.007205963134765625,
0.006183624267578125,
0.04449462890625,
0.002349853515625,
-0.00199127197265625,
0.01349639892578125,
-0.0137481689453125,
-0.0311737060546875,
-0.007598876953125,
-0.06951904296875,
-0.01180267333984375,
0.022125244140625,
-0.0491943359375,
0.027801513671875,
-0.036163330078125,
-0.033538818359375,
-0.0024013519287109375,
0.01103973388671875,
-0.0487060546875,
0.0024700164794921875,
0.010711669921875,
0.053802490234375,
-0.051177978515625,
0.048919677734375,
0.0496826171875,
-0.057861328125,
-0.0823974609375,
-0.03314208984375,
0.006687164306640625,
-0.058563232421875,
0.038482666015625,
0.01319122314453125,
0.00934600830078125,
0.01074981689453125,
-0.0733642578125,
-0.07635498046875,
0.10699462890625,
0.0271453857421875,
-0.052703857421875,
-0.016387939453125,
0.007785797119140625,
0.0197601318359375,
-0.024261474609375,
0.03045654296875,
0.020294189453125,
0.037994384765625,
0.021759033203125,
-0.0780029296875,
0.0179595947265625,
-0.032196044921875,
-0.0023403167724609375,
-0.00479888916015625,
-0.08428955078125,
0.0712890625,
-0.0206298828125,
-0.0170135498046875,
0.039947509765625,
0.0377197265625,
0.04437255859375,
0.03619384765625,
0.036651611328125,
0.03955078125,
0.050323486328125,
0.0002613067626953125,
0.07757568359375,
-0.02166748046875,
0.045135498046875,
0.049224853515625,
-0.007701873779296875,
0.05743408203125,
0.0303497314453125,
-0.037353515625,
0.039306640625,
0.07647705078125,
-0.00528717041015625,
0.028961181640625,
-0.00037384033203125,
-0.00925445556640625,
-0.02032470703125,
-0.01264190673828125,
-0.0645751953125,
0.01074981689453125,
0.007587432861328125,
-0.00839996337890625,
-0.00445556640625,
-0.0103302001953125,
-0.0013933181762695312,
-0.03497314453125,
-0.02752685546875,
0.04638671875,
0.02276611328125,
-0.0227508544921875,
0.0860595703125,
-0.0159454345703125,
0.05279541015625,
-0.058807373046875,
-0.007457733154296875,
-0.0276947021484375,
0.01190185546875,
-0.02227783203125,
-0.0298919677734375,
0.00931549072265625,
-0.0006618499755859375,
-0.005107879638671875,
0.01300048828125,
0.048583984375,
-0.0158843994140625,
-0.0543212890625,
0.03192138671875,
0.01207733154296875,
0.03668212890625,
0.004390716552734375,
-0.06146240234375,
0.0408935546875,
-0.003665924072265625,
-0.036285400390625,
0.045562744140625,
0.0280609130859375,
0.004077911376953125,
0.07073974609375,
0.063720703125,
-0.00188446044921875,
0.023101806640625,
-0.02154541015625,
0.07464599609375,
-0.03192138671875,
-0.0294952392578125,
-0.057586669921875,
0.059326171875,
0.0009274482727050781,
-0.007564544677734375,
0.04583740234375,
0.048583984375,
0.055450439453125,
-0.00774383544921875,
0.06292724609375,
-0.0240631103515625,
0.00954437255859375,
-0.013397216796875,
0.072998046875,
-0.085693359375,
0.0064849853515625,
-0.01177978515625,
-0.050079345703125,
0.01080322265625,
0.04901123046875,
-0.0012674331665039062,
0.01256561279296875,
0.0200653076171875,
0.07281494140625,
-0.0051422119140625,
-0.0115203857421875,
0.01468658447265625,
0.033294677734375,
0.0302886962890625,
0.043792724609375,
0.05810546875,
-0.05987548828125,
0.0592041015625,
-0.03985595703125,
-0.007648468017578125,
-0.037322998046875,
-0.06884765625,
-0.0531005859375,
-0.035064697265625,
-0.036895751953125,
-0.031585693359375,
-0.0003039836883544922,
0.06597900390625,
0.068115234375,
-0.04974365234375,
-0.031005859375,
0.003902435302734375,
0.0070953369140625,
-0.0204315185546875,
-0.0193939208984375,
0.0200042724609375,
0.01189422607421875,
-0.048126220703125,
0.01654052734375,
-0.01192474365234375,
0.02984619140625,
-0.0010385513305664062,
-0.0302734375,
-0.0276947021484375,
0.0169219970703125,
0.0439453125,
0.043731689453125,
-0.06402587890625,
-0.0109405517578125,
-0.01287078857421875,
-0.008270263671875,
0.01258087158203125,
0.0092010498046875,
-0.043243408203125,
0.0193634033203125,
0.027099609375,
0.024993896484375,
0.0478515625,
0.01200103759765625,
0.0091705322265625,
-0.017822265625,
0.0184326171875,
-0.0037212371826171875,
0.0382080078125,
0.02679443359375,
-0.03399658203125,
0.03662109375,
0.03277587890625,
-0.049591064453125,
-0.08673095703125,
-0.01558685302734375,
-0.0927734375,
-0.020172119140625,
0.08306884765625,
-0.01336669921875,
-0.0279388427734375,
0.0187530517578125,
-0.03387451171875,
0.051055908203125,
-0.04449462890625,
0.038177490234375,
0.048095703125,
-0.01039886474609375,
-0.01049041748046875,
-0.04095458984375,
0.0142974853515625,
0.01493072509765625,
-0.04937744140625,
0.005664825439453125,
0.0491943359375,
0.048553466796875,
0.0222320556640625,
0.056671142578125,
-0.0059814453125,
0.0212554931640625,
0.02032470703125,
0.016082763671875,
-0.0245513916015625,
0.00626373291015625,
-0.0244293212890625,
-0.00228118896484375,
-0.01090240478515625,
-0.01959228515625
]
] |
Mikivis/gpt2-large-lora-sft | 2023-09-09T10:15:37.000Z | [
"transformers",
"pytorch",
"safetensors",
"gpt2",
"text-generation",
"generated_from_trainer",
"dataset:customized",
"license:mit",
"endpoints_compatible",
"has_space",
"text-generation-inference",
"region:us"
] | text-generation | Mikivis | null | null | Mikivis/gpt2-large-lora-sft | 1 | 5,911 | transformers | 2023-09-04T08:58:35 | ---
license: mit
base_model: gpt2-large
tags:
- generated_from_trainer
datasets:
- customized
model-index:
- name: gpt2-large-lora-sft
results: []
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# gpt2-large-lora-sft
This model is a fine-tuned version of [gpt2-large](https://huggingface.co/gpt2-large) on the customized dataset.
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 0.00013
- train_batch_size: 1
- eval_batch_size: 8
- seed: 42
- distributed_type: multi-GPU
- num_devices: 6
- total_train_batch_size: 6
- total_eval_batch_size: 48
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 2.5
### Training results
### Framework versions
- Transformers 4.32.1
- Pytorch 2.0.1+cu117
- Datasets 2.10.1
- Tokenizers 0.13.3
| 1,169 | [
[
-0.038238525390625,
-0.06451416015625,
0.0229949951171875,
0.00748443603515625,
-0.03656005859375,
-0.027069091796875,
-0.00897979736328125,
-0.0272064208984375,
0.0152435302734375,
0.0257415771484375,
-0.040283203125,
-0.027069091796875,
-0.054779052734375,
-0.01061248779296875,
-0.01293182373046875,
0.10394287109375,
-0.0095672607421875,
0.0221710205078125,
0.003131866455078125,
0.0016450881958007812,
-0.020904541015625,
-0.041534423828125,
-0.08148193359375,
-0.059112548828125,
0.03912353515625,
0.003002166748046875,
0.05889892578125,
0.05938720703125,
0.044403076171875,
0.02081298828125,
-0.022674560546875,
-0.013336181640625,
-0.048675537109375,
-0.0389404296875,
-0.0108642578125,
-0.023162841796875,
-0.06805419921875,
0.00673675537109375,
0.049407958984375,
0.00977325439453125,
-0.01922607421875,
0.03753662109375,
0.0147705078125,
0.0235748291015625,
-0.0268096923828125,
0.039947509765625,
-0.039886474609375,
0.030364990234375,
-0.006145477294921875,
-0.0169219970703125,
-0.016632080078125,
-0.0176849365234375,
-0.00012874603271484375,
-0.049285888671875,
0.035980224609375,
-0.0177459716796875,
0.0775146484375,
0.0364990234375,
-0.02703857421875,
0.01279449462890625,
-0.0631103515625,
0.0256500244140625,
-0.042144775390625,
0.0047760009765625,
0.02984619140625,
0.0478515625,
0.007579803466796875,
-0.06439208984375,
-0.0238494873046875,
-0.0049285888671875,
0.00287628173828125,
0.0147705078125,
-0.00399017333984375,
0.00803375244140625,
0.066162109375,
0.02410888671875,
-0.045684814453125,
0.0192718505859375,
-0.044830322265625,
-0.01433563232421875,
0.047149658203125,
0.025390625,
-0.015625,
0.0021495819091796875,
-0.034271240234375,
-0.00421905517578125,
-0.038970947265625,
-0.01099395751953125,
0.041412353515625,
0.01541900634765625,
-0.03070068359375,
0.056304931640625,
-0.01465606689453125,
0.03985595703125,
0.0070037841796875,
-0.0099029541015625,
0.0390625,
0.00311279296875,
-0.03704833984375,
-0.00807952880859375,
0.063720703125,
0.029815673828125,
0.0223236083984375,
0.0019817352294921875,
-0.021728515625,
-0.01009368896484375,
0.01230621337890625,
-0.07672119140625,
-0.03131103515625,
0.0023651123046875,
-0.031005859375,
-0.041748046875,
0.00556182861328125,
-0.045135498046875,
-0.0018739700317382812,
-0.03509521484375,
0.04998779296875,
-0.01377105712890625,
-0.01538848876953125,
0.0008678436279296875,
-0.01404571533203125,
0.026824951171875,
0.025360107421875,
-0.0667724609375,
0.02734375,
0.040283203125,
0.042694091796875,
0.00933074951171875,
-0.029571533203125,
-0.031280517578125,
0.0197906494140625,
-0.00873565673828125,
0.040802001953125,
0.00014853477478027344,
-0.03204345703125,
-0.01343536376953125,
0.0181427001953125,
-0.003692626953125,
-0.027374267578125,
0.058319091796875,
-0.03179931640625,
0.0256195068359375,
-0.0312042236328125,
-0.035491943359375,
-0.007686614990234375,
0.027557373046875,
-0.049591064453125,
0.0853271484375,
0.030609130859375,
-0.0626220703125,
0.0404052734375,
-0.061492919921875,
-0.017303466796875,
0.019378662109375,
-0.0157928466796875,
-0.05712890625,
-0.0006055831909179688,
-0.002155303955078125,
0.0255279541015625,
-0.0214080810546875,
0.0249786376953125,
-0.0261993408203125,
-0.04876708984375,
-0.01203155517578125,
-0.039581298828125,
0.03912353515625,
0.01381683349609375,
-0.03448486328125,
0.0179901123046875,
-0.07110595703125,
0.02435302734375,
0.03271484375,
-0.045074462890625,
0.021575927734375,
-0.02911376953125,
0.042327880859375,
0.021087646484375,
0.031707763671875,
-0.037933349609375,
0.032318115234375,
-0.0116119384765625,
0.044464111328125,
0.05511474609375,
-0.004398345947265625,
-0.005695343017578125,
-0.008636474609375,
0.0204925537109375,
-0.0047760009765625,
0.035400390625,
0.03912353515625,
-0.04583740234375,
-0.049407958984375,
-0.0195159912109375,
0.0176849365234375,
0.031829833984375,
-0.0311279296875,
0.06048583984375,
-0.010284423828125,
-0.045257568359375,
0.001102447509765625,
0.00496673583984375,
0.0259246826171875,
0.031829833984375,
0.0323486328125,
-0.01395416259765625,
-0.0247955322265625,
-0.073486328125,
0.00531768798828125,
0.001094818115234375,
0.0073089599609375,
0.01180267333984375,
0.06842041015625,
-0.0276641845703125,
0.05865478515625,
-0.059661865234375,
-0.01215362548828125,
-0.016448974609375,
-0.0026798248291015625,
0.0240020751953125,
0.052703857421875,
0.055511474609375,
-0.032470703125,
-0.0270233154296875,
-0.00797271728515625,
-0.057464599609375,
0.0257720947265625,
0.006195068359375,
-0.01678466796875,
-0.011627197265625,
0.0255889892578125,
-0.058135986328125,
0.045074462890625,
0.0176849365234375,
-0.028656005859375,
0.0438232421875,
-0.048431396484375,
-0.0116729736328125,
-0.08819580078125,
0.01629638671875,
0.01300048828125,
-0.0087890625,
-0.0114593505859375,
0.01197052001953125,
-0.002849578857421875,
-0.017425537109375,
-0.0235137939453125,
0.047210693359375,
-0.0015687942504882812,
-0.0004906654357910156,
-0.02764892578125,
-0.0236053466796875,
-0.004566192626953125,
0.04718017578125,
0.01396942138671875,
0.0498046875,
0.03875732421875,
-0.0258941650390625,
0.04296875,
0.044586181640625,
-0.0238494873046875,
0.032379150390625,
-0.08203125,
0.02130126953125,
0.0004818439483642578,
0.024078369140625,
-0.05023193359375,
-0.0400390625,
0.03955078125,
-0.0170135498046875,
0.02459716796875,
-0.02911376953125,
-0.05206298828125,
-0.04522705078125,
0.003200531005859375,
0.0307464599609375,
0.049896240234375,
-0.05426025390625,
0.028045654296875,
-0.0013227462768554688,
0.0187835693359375,
-0.0242156982421875,
-0.0576171875,
-0.0221710205078125,
-0.01904296875,
-0.029510498046875,
0.005748748779296875,
-0.006336212158203125,
0.0181121826171875,
-0.01346588134765625,
0.007083892822265625,
-0.0204315185546875,
-0.01258087158203125,
0.027557373046875,
0.02301025390625,
-0.016510009765625,
-0.00494384765625,
0.00347900390625,
-0.0308380126953125,
0.0248870849609375,
-0.00850677490234375,
0.042266845703125,
0.0007014274597167969,
-0.02227783203125,
-0.0518798828125,
0.0002391338348388672,
0.0276336669921875,
-0.00925445556640625,
0.060394287109375,
0.08856201171875,
-0.0369873046875,
0.00687408447265625,
-0.032684326171875,
-0.01158905029296875,
-0.035430908203125,
0.056304931640625,
-0.039794921875,
-0.035736083984375,
0.037200927734375,
0.0078582763671875,
-0.004688262939453125,
0.06463623046875,
0.04644775390625,
0.0218048095703125,
0.0906982421875,
0.02349853515625,
-0.0151214599609375,
0.02935791015625,
-0.056121826171875,
-0.0044097900390625,
-0.061248779296875,
-0.02447509765625,
-0.03143310546875,
-0.0171051025390625,
-0.055084228515625,
-0.007537841796875,
0.0191650390625,
0.00885009765625,
-0.0728759765625,
0.027923583984375,
-0.040863037109375,
0.0406494140625,
0.049530029296875,
0.035430908203125,
-0.00199127197265625,
0.02532958984375,
0.0057525634765625,
0.007251739501953125,
-0.051055908203125,
-0.0213470458984375,
0.0928955078125,
0.04644775390625,
0.051177978515625,
-0.003662109375,
0.04736328125,
-0.0087738037109375,
0.00859832763671875,
-0.045501708984375,
0.036407470703125,
0.0018320083618164062,
-0.06390380859375,
-0.0187225341796875,
-0.037994384765625,
-0.0498046875,
0.01369476318359375,
-0.02532958984375,
-0.048126220703125,
-0.01299285888671875,
0.0240631103515625,
-0.025390625,
0.035552978515625,
-0.05126953125,
0.0845947265625,
-0.0203094482421875,
-0.041473388671875,
-0.0158538818359375,
-0.032562255859375,
0.0125732421875,
-0.0020008087158203125,
-0.0292205810546875,
0.0059814453125,
0.00958251953125,
0.06243896484375,
-0.056488037109375,
0.04766845703125,
-0.032989501953125,
0.00772857666015625,
0.030120849609375,
-0.03057861328125,
0.07147216796875,
0.02301025390625,
-0.003963470458984375,
0.006145477294921875,
-0.00899505615234375,
-0.050537109375,
-0.0254058837890625,
0.060455322265625,
-0.0970458984375,
-0.0113983154296875,
-0.040557861328125,
-0.029541015625,
-0.01727294921875,
0.01366424560546875,
0.057861328125,
0.035980224609375,
-0.0169525146484375,
0.00537872314453125,
0.03656005859375,
0.0033550262451171875,
0.0197906494140625,
0.00746917724609375,
0.001461029052734375,
-0.046112060546875,
0.06640625,
-0.009674072265625,
0.01197052001953125,
-0.006053924560546875,
0.0186004638671875,
-0.0298004150390625,
-0.04071044921875,
-0.028839111328125,
0.0311126708984375,
-0.049072265625,
-0.008148193359375,
-0.0233612060546875,
-0.037322998046875,
-0.0212249755859375,
0.0112152099609375,
-0.030364990234375,
-0.011962890625,
-0.038330078125,
-0.0191650390625,
0.0367431640625,
0.05743408203125,
0.0009722709655761719,
0.06463623046875,
-0.040924072265625,
0.0081787109375,
0.0273895263671875,
0.044219970703125,
-0.00817108154296875,
-0.063720703125,
-0.0241851806640625,
0.0014743804931640625,
-0.034149169921875,
-0.0246429443359375,
0.0192108154296875,
-0.0002353191375732422,
0.03436279296875,
0.037017822265625,
-0.035797119140625,
0.0555419921875,
-0.027923583984375,
0.05755615234375,
0.0312042236328125,
-0.032196044921875,
0.02972412109375,
-0.040924072265625,
0.0209503173828125,
0.0390625,
0.0292816162109375,
0.01328277587890625,
0.01535797119140625,
-0.081787109375,
-0.05255126953125,
0.054656982421875,
0.0243072509765625,
0.0120086669921875,
0.01800537109375,
0.046600341796875,
0.0005354881286621094,
0.018890380859375,
-0.070068359375,
-0.02410888671875,
-0.0185089111328125,
-0.0004925727844238281,
-0.030487060546875,
-0.030181884765625,
-0.021636962890625,
-0.048187255859375,
0.07305908203125,
-0.0107269287109375,
0.03515625,
-0.01071929931640625,
0.0212249755859375,
-0.0161590576171875,
-0.007568359375,
0.042449951171875,
0.04730224609375,
-0.041046142578125,
-0.0272064208984375,
0.02008056640625,
-0.05377197265625,
-0.00928497314453125,
0.0264892578125,
-0.0083770751953125,
0.0006113052368164062,
0.0197601318359375,
0.0853271484375,
-0.000013828277587890625,
0.0012226104736328125,
0.0282135009765625,
-0.00417327880859375,
-0.040557861328125,
-0.034210205078125,
0.024200439453125,
-0.01018524169921875,
0.0221405029296875,
-0.0038013458251953125,
0.02923583984375,
-0.0006766319274902344,
-0.019989013671875,
-0.0033702850341796875,
0.0252838134765625,
-0.0301361083984375,
-0.0240936279296875,
0.07147216796875,
0.0118865966796875,
-0.01175689697265625,
0.059356689453125,
-0.01215362548828125,
-0.01861572265625,
0.050994873046875,
0.0458984375,
0.0662841796875,
-0.0038013458251953125,
0.00591278076171875,
0.06268310546875,
0.00624847412109375,
-0.030609130859375,
0.032806396484375,
0.006855010986328125,
-0.031890869140625,
-0.02294921875,
-0.03900146484375,
-0.01020050048828125,
0.05047607421875,
-0.07049560546875,
0.034515380859375,
-0.049530029296875,
-0.0360107421875,
0.006793975830078125,
0.0180206298828125,
-0.06854248046875,
0.037841796875,
0.00614166259765625,
0.080078125,
-0.06976318359375,
0.07635498046875,
0.039031982421875,
-0.039276123046875,
-0.075439453125,
-0.018096923828125,
-0.0120086669921875,
-0.06854248046875,
0.044189453125,
-0.003223419189453125,
0.02752685546875,
0.01122283935546875,
-0.048797607421875,
-0.05535888671875,
0.0836181640625,
0.0254058837890625,
-0.045074462890625,
-0.0086822509765625,
0.026458740234375,
0.052886962890625,
-0.010009765625,
0.042694091796875,
0.0268402099609375,
0.028533935546875,
0.0238037109375,
-0.069091796875,
-0.00881195068359375,
-0.0159454345703125,
0.004467010498046875,
0.01910400390625,
-0.060302734375,
0.0743408203125,
-0.0110931396484375,
0.04010009765625,
0.0308380126953125,
0.0308990478515625,
0.015289306640625,
0.0067291259765625,
0.0201568603515625,
0.062164306640625,
0.038177490234375,
-0.02301025390625,
0.073486328125,
-0.0212249755859375,
0.06378173828125,
0.0985107421875,
-0.006988525390625,
0.045989990234375,
0.024383544921875,
-0.0228424072265625,
0.00505828857421875,
0.0635986328125,
-0.04132080078125,
0.03143310546875,
0.0188446044921875,
-0.0038738250732421875,
-0.0249786376953125,
0.032318115234375,
-0.06854248046875,
0.0225982666015625,
-0.00037789344787597656,
-0.055877685546875,
-0.0216522216796875,
-0.0169219970703125,
-0.006023406982421875,
-0.03656005859375,
-0.03466796875,
0.0438232421875,
-0.020233154296875,
-0.026580810546875,
0.04510498046875,
0.006900787353515625,
0.0264434814453125,
-0.059295654296875,
-0.0021991729736328125,
-0.0009140968322753906,
0.033966064453125,
-0.0259246826171875,
-0.046600341796875,
0.0015993118286132812,
-0.003704071044921875,
-0.00958251953125,
0.007740020751953125,
0.0285491943359375,
-0.01690673828125,
-0.05255126953125,
0.00899505615234375,
0.02459716796875,
0.0211334228515625,
-0.0161895751953125,
-0.0775146484375,
-0.003963470458984375,
-0.004291534423828125,
-0.032562255859375,
0.033477783203125,
0.0180816650390625,
0.003753662109375,
0.0292816162109375,
0.038116455078125,
-0.000017523765563964844,
-0.00853729248046875,
0.018524169921875,
0.069091796875,
-0.04351806640625,
-0.045501708984375,
-0.049713134765625,
0.0248260498046875,
-0.0139007568359375,
-0.0667724609375,
0.0374755859375,
0.08306884765625,
0.060638427734375,
-0.026580810546875,
0.054412841796875,
0.00730133056640625,
0.021148681640625,
-0.035247802734375,
0.042388916015625,
-0.0267791748046875,
-0.005031585693359375,
-0.0222320556640625,
-0.08624267578125,
0.01108551025390625,
0.057037353515625,
-0.0287322998046875,
0.0289459228515625,
0.037139892578125,
0.044677734375,
-0.02178955078125,
0.0135040283203125,
0.01076507568359375,
0.01032257080078125,
0.0190582275390625,
0.03985595703125,
0.032257080078125,
-0.06512451171875,
0.0307159423828125,
-0.025787353515625,
-0.01107025146484375,
-0.00836944580078125,
-0.044464111328125,
-0.07513427734375,
-0.0191802978515625,
-0.045440673828125,
-0.047698974609375,
0.0006361007690429688,
0.0703125,
0.058074951171875,
-0.046600341796875,
-0.0166015625,
-0.00998687744140625,
-0.03314208984375,
-0.00576019287109375,
-0.01293182373046875,
0.044830322265625,
0.0015869140625,
-0.050567626953125,
-0.015167236328125,
-0.0274658203125,
0.035247802734375,
-0.01129150390625,
-0.0190277099609375,
0.006694793701171875,
-0.0291595458984375,
0.037200927734375,
0.01325225830078125,
-0.033966064453125,
-0.035430908203125,
-0.032928466796875,
-0.0091094970703125,
0.01507568359375,
0.0311279296875,
-0.037322998046875,
0.0024242401123046875,
0.014617919921875,
0.01102447509765625,
0.0631103515625,
0.0023860931396484375,
0.01678466796875,
-0.039794921875,
0.0367431640625,
-0.002780914306640625,
0.032257080078125,
0.0001837015151977539,
-0.03472900390625,
0.053802490234375,
0.0301971435546875,
-0.0487060546875,
-0.04571533203125,
-0.0026607513427734375,
-0.088134765625,
0.012664794921875,
0.10186767578125,
0.008636474609375,
-0.0214385986328125,
0.030609130859375,
-0.03179931640625,
0.019134521484375,
-0.0160369873046875,
0.035369873046875,
0.036376953125,
-0.0079498291015625,
0.0176849365234375,
-0.050994873046875,
0.036590576171875,
-0.00485992431640625,
-0.06317138671875,
-0.016937255859375,
0.026458740234375,
0.049652099609375,
-0.006244659423828125,
0.0281524658203125,
-0.00472259521484375,
0.0246124267578125,
0.00992584228515625,
0.009674072265625,
-0.03155517578125,
-0.007602691650390625,
-0.020843505859375,
0.006206512451171875,
0.009246826171875,
-0.024505615234375
]
] |
Open-Orca/OpenOrca-Preview1-13B | 2023-07-17T06:07:48.000Z | [
"transformers",
"pytorch",
"llama",
"text-generation",
"en",
"dataset:Open-Orca/OpenOrca",
"arxiv:2306.02707",
"arxiv:2301.13688",
"license:mit",
"endpoints_compatible",
"has_space",
"text-generation-inference",
"region:us"
] | text-generation | Open-Orca | null | null | Open-Orca/OpenOrca-Preview1-13B | 144 | 5,910 | transformers | 2023-07-12T01:13:58 | ---
license: mit
language:
- en
library_name: transformers
pipeline_tag: text-generation
datasets:
- Open-Orca/OpenOrca
---
<p><h1>🐋 The First OpenOrca Model Preview! 🐋</h1></p>
# OpenOrca-Preview1-13B
We have used our own [OpenOrca dataset](https://huggingface.co/datasets/Open-Orca/OpenOrca) to fine-tune LLaMA-13B.
This dataset is our attempt to reproduce the dataset generated for Microsoft Research's [Orca Paper](https://arxiv.org/abs/2306.02707).
We have trained on less than 6% of our data, just to give a preview of what is possible while we further refine our dataset!
We trained a refined selection of 200k GPT-4 entries from OpenOrca.
We have filtered our GPT-4 augmentations to remove statements like, "As an AI language model..." and other responses which have been shown to harm model reasoning capabilities. Further details on our dataset curation practices will be forthcoming with our full model releases.
This release highlights that even a small portion of our training data can produce state of the art results in this model class with training costs <$200 in total.
Want to visualize our full (pre-filtering) dataset? Check out our [Nomic Atlas Map](https://atlas.nomic.ai/map/c1b88b47-2d9b-47e0-9002-b80766792582/2560fd25-52fe-42f1-a58f-ff5eccc890d2).
[<img src="https://huggingface.co/Open-Orca/OpenOrca-Preview1-13B/resolve/main/OpenOrca%20Nomic%20Atlas.png" alt="Atlas Nomic Dataset Map" width="400" height="400" />](https://atlas.nomic.ai/map/c1b88b47-2d9b-47e0-9002-b80766792582/2560fd25-52fe-42f1-a58f-ff5eccc890d2)
We are in-process with training more models, so keep a look out on our org for releases coming soon with exciting partners.
We will also give sneak-peak announcements on our Discord, which you can find here:
https://AlignmentLab.ai
# Evaluation
We have evaluated OpenOrca-Preview1-13B on hard reasoning tasks from BigBench-Hard and AGIEval as outlined in the Orca paper.
Our average performance for BigBench-Hard: 0.3753
Average for AGIEval: 0.3638
In the Orca paper, they measured their score relative to Vicuna on these evals.
We've done the same and have found our score averages to ~60% of the total improvement that was shown in the Orca paper.
So we got 60% of the improvement with 6% of the data!
## BigBench-Hard Performance
## AGIEval Performance
We will report our results on [HuggingFaceH4 Open LLM Leaderboard](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard) Evals once we receive them.
# Dataset
We used a small (6%, 200k) subset of our data from OpenOrca, which aims to reproduce the Orca Research Paper dataset.
As this release is intended as a preview, please await our full releases for further details on the training data.
# Training
[<img src="https://raw.githubusercontent.com/OpenAccess-AI-Collective/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/OpenAccess-AI-Collective/axolotl)
We trained with 8x A100-80G GPUs for 15 hours. Commodity cost was < $200.
We trained for 4 epochs and selected a snapshot at 3 epochs for peak performance.
Please await our full releases for further training details.
# Prompting
It uses the Alpaca format (see [FastChat implementation example](https://github.com/lm-sys/FastChat/blob/daa2b9abe20597ebf34dc5df164d450456610c74/fastchat/conversation.py#L198-L229)):
```
### Instruction:
### Response:
```
# Citation
```bibtex
@software{OpenOrca_Preview1,
title = {OpenOrca_Preview1: A LLaMA-13B Model Fine-tuned on Small Portion of OpenOrcaV1 Dataset},
author = {Wing Lian and Bleys Goodson and Eugene Pentland and Austin Cook and Chanvichet Vong and "Teknium"},
year = {2023},
publisher = {HuggingFace},
journal = {HuggingFace repository},
howpublished = {\url{https://https://huggingface.co/Open-Orca/OpenOrca-Preview1-13B},
}
```
```bibtex
@misc{mukherjee2023orca,
title={Orca: Progressive Learning from Complex Explanation Traces of GPT-4},
author={Subhabrata Mukherjee and Arindam Mitra and Ganesh Jawahar and Sahaj Agarwal and Hamid Palangi and Ahmed Awadallah},
year={2023},
eprint={2306.02707},
archivePrefix={arXiv},
primaryClass={cs.CL}
}
```
```bibtex
@misc{longpre2023flan,
title={The Flan Collection: Designing Data and Methods for Effective Instruction Tuning},
author={Shayne Longpre and Le Hou and Tu Vu and Albert Webson and Hyung Won Chung and Yi Tay and Denny Zhou and Quoc V. Le and Barret Zoph and Jason Wei and Adam Roberts},
year={2023},
eprint={2301.13688},
archivePrefix={arXiv},
primaryClass={cs.AI}
}
```
```bibtex
@software{touvron2023llama,
title={LLaMA: Open and Efficient Foundation Language Models},
author={Touvron, Hugo and Lavril, Thibaut and Izacard, Gautier and Martinet, Xavier and Lachaux, Marie-Anne and Lacroix, Timoth{\'e}e and Rozi{\`e}re, Baptiste and Goyal, Naman and Hambro, Eric and Azhar, Faisal and Rodriguez, Aurelien and Joulin, Armand and Grave, Edouard and Lample, Guillaume},
journal={arXiv preprint arXiv:2302.13971},
year={2023}
}
``` | 5,546 | [
[
-0.03851318359375,
-0.0650634765625,
0.01103973388671875,
-0.003032684326171875,
-0.01561737060546875,
-0.005741119384765625,
-0.0202789306640625,
-0.0703125,
0.0090179443359375,
0.0152740478515625,
-0.042999267578125,
-0.05194091796875,
-0.026092529296875,
-0.00405120849609375,
-0.019256591796875,
0.08349609375,
-0.0110931396484375,
-0.01557159423828125,
-0.0019359588623046875,
-0.03680419921875,
-0.019622802734375,
-0.0433349609375,
-0.0631103515625,
-0.0271759033203125,
0.042755126953125,
0.020416259765625,
0.049468994140625,
0.07049560546875,
0.0276641845703125,
0.0158538818359375,
-0.02667236328125,
0.0286407470703125,
-0.04681396484375,
-0.0176544189453125,
0.01104736328125,
-0.0281829833984375,
-0.06231689453125,
0.01346588134765625,
0.02947998046875,
0.03277587890625,
-0.02288818359375,
0.02667236328125,
0.00852203369140625,
0.040496826171875,
-0.042388916015625,
0.034576416015625,
-0.021575927734375,
-0.00405120849609375,
-0.0305633544921875,
0.00971221923828125,
-0.0196380615234375,
-0.02813720703125,
0.0004646778106689453,
-0.06390380859375,
0.006732940673828125,
0.00044465065002441406,
0.09161376953125,
0.01519012451171875,
-0.0208740234375,
-0.01438140869140625,
-0.038177490234375,
0.04388427734375,
-0.05169677734375,
0.0235748291015625,
0.01155853271484375,
0.0178070068359375,
-0.0224609375,
-0.0550537109375,
-0.05218505859375,
-0.019989013671875,
0.0052337646484375,
0.021636962890625,
-0.0149078369140625,
-0.0005431175231933594,
0.017669677734375,
0.042144775390625,
-0.04559326171875,
0.0257110595703125,
-0.041656494140625,
-0.01165771484375,
0.056976318359375,
0.00815582275390625,
0.00830078125,
0.0241546630859375,
-0.03118896484375,
-0.05029296875,
-0.06494140625,
0.0256195068359375,
0.03558349609375,
0.02728271484375,
-0.045806884765625,
0.033477783203125,
0.0027027130126953125,
0.053497314453125,
-0.0080108642578125,
-0.0175323486328125,
0.040802001953125,
-0.028900146484375,
-0.023773193359375,
-0.01009368896484375,
0.071044921875,
0.005126953125,
0.0030155181884765625,
0.00806427001953125,
-0.0081329345703125,
0.0199737548828125,
-0.002643585205078125,
-0.0657958984375,
-0.01369476318359375,
0.0110015869140625,
-0.028533935546875,
-0.0218658447265625,
0.0015001296997070312,
-0.052093505859375,
-0.01540374755859375,
-0.02093505859375,
0.0117950439453125,
-0.033050537109375,
-0.0255126953125,
0.0261688232421875,
0.019744873046875,
0.037628173828125,
0.0318603515625,
-0.049407958984375,
0.0234222412109375,
0.0369873046875,
0.06585693359375,
-0.0032711029052734375,
-0.0233154296875,
-0.0187530517578125,
-0.0175933837890625,
-0.03411865234375,
0.052947998046875,
-0.0311431884765625,
-0.0241546630859375,
-0.0174560546875,
-0.0133819580078125,
-0.005809783935546875,
-0.03240966796875,
0.052337646484375,
-0.023406982421875,
0.0203399658203125,
-0.025299072265625,
-0.011932373046875,
-0.034881591796875,
0.0179901123046875,
-0.04595947265625,
0.0897216796875,
-0.00003641843795776367,
-0.04962158203125,
0.0201568603515625,
-0.06378173828125,
-0.01323699951171875,
-0.0341796875,
-0.0101165771484375,
-0.03802490234375,
-0.0181884765625,
0.037841796875,
0.0155029296875,
-0.035064697265625,
0.0009298324584960938,
-0.040191650390625,
-0.01556396484375,
0.0057220458984375,
-0.0090789794921875,
0.06732177734375,
0.0303802490234375,
-0.030120849609375,
0.0082550048828125,
-0.05615234375,
-0.0069122314453125,
0.0288543701171875,
-0.02899169921875,
-0.004108428955078125,
-0.01666259765625,
-0.0189971923828125,
0.0208892822265625,
0.024993896484375,
-0.03656005859375,
0.0389404296875,
-0.0347900390625,
0.04071044921875,
0.0584716796875,
-0.0213623046875,
0.0226898193359375,
-0.0245513916015625,
0.041748046875,
0.004619598388671875,
0.03155517578125,
-0.00893402099609375,
-0.052337646484375,
-0.06805419921875,
-0.0260772705078125,
0.030548095703125,
0.0231781005859375,
-0.03826904296875,
0.0273895263671875,
-0.0198974609375,
-0.06280517578125,
-0.042266845703125,
-0.002483367919921875,
0.046356201171875,
0.046112060546875,
0.034912109375,
-0.05718994140625,
-0.0313720703125,
-0.051605224609375,
0.010528564453125,
-0.022918701171875,
0.01244354248046875,
0.0335693359375,
0.043182373046875,
0.003734588623046875,
0.06689453125,
-0.035797119140625,
-0.0294342041015625,
-0.006557464599609375,
0.0001785755157470703,
0.0222625732421875,
0.0394287109375,
0.06494140625,
-0.031524658203125,
-0.018951416015625,
-0.0000016093254089355469,
-0.0640869140625,
0.0138092041015625,
0.01374053955078125,
-0.0270843505859375,
0.0306396484375,
0.03369140625,
-0.04388427734375,
0.043243408203125,
0.04364013671875,
-0.0256500244140625,
0.032501220703125,
-0.019775390625,
-0.0202178955078125,
-0.062469482421875,
0.0232391357421875,
0.0077972412109375,
0.011810302734375,
-0.0289306640625,
0.01222991943359375,
-0.00981903076171875,
-0.006439208984375,
-0.03387451171875,
0.049102783203125,
-0.048095703125,
-0.005809783935546875,
0.01039886474609375,
0.0126800537109375,
-0.00753021240234375,
0.058807373046875,
-0.01117706298828125,
0.05352783203125,
0.048095703125,
-0.0323486328125,
0.0046234130859375,
0.0313720703125,
-0.0309906005859375,
0.01239776611328125,
-0.0650634765625,
0.03741455078125,
-0.0113067626953125,
0.0513916015625,
-0.0736083984375,
-0.01165771484375,
0.038787841796875,
-0.0299072265625,
0.0318603515625,
-0.0021610260009765625,
-0.0372314453125,
-0.037261962890625,
-0.0300445556640625,
0.032196044921875,
0.028961181640625,
-0.0550537109375,
0.041748046875,
0.0203094482421875,
0.01364898681640625,
-0.05938720703125,
-0.0538330078125,
-0.0165252685546875,
-0.0204925537109375,
-0.0614013671875,
0.0275726318359375,
-0.007350921630859375,
0.00240325927734375,
-0.004993438720703125,
-0.0171966552734375,
0.008819580078125,
0.00655364990234375,
0.035064697265625,
0.03143310546875,
-0.0256195068359375,
-0.0018644332885742188,
-0.007595062255859375,
-0.0023479461669921875,
-0.0082550048828125,
-0.0250701904296875,
0.05181884765625,
-0.03533935546875,
-0.01546478271484375,
-0.0394287109375,
-0.0109710693359375,
0.03509521484375,
-0.048583984375,
0.0697021484375,
0.04095458984375,
-0.01438140869140625,
0.010345458984375,
-0.035614013671875,
-0.0223388671875,
-0.03460693359375,
0.0071563720703125,
-0.0222015380859375,
-0.05938720703125,
0.06201171875,
0.025360107421875,
0.030914306640625,
0.051910400390625,
0.028076171875,
0.02728271484375,
0.06463623046875,
0.044342041015625,
-0.0167999267578125,
0.0413818359375,
-0.05633544921875,
0.001544952392578125,
-0.058319091796875,
-0.04156494140625,
-0.052093505859375,
-0.042388916015625,
-0.04681396484375,
-0.0270233154296875,
0.0249176025390625,
0.01214599609375,
-0.041015625,
0.0260467529296875,
-0.050384521484375,
0.0239105224609375,
0.03997802734375,
0.0257568359375,
0.015655517578125,
0.00960540771484375,
-0.01013946533203125,
0.01763916015625,
-0.054901123046875,
-0.037994384765625,
0.10302734375,
0.0218505859375,
0.049285888671875,
0.009674072265625,
0.044525146484375,
-0.0094757080078125,
0.0262451171875,
-0.0281829833984375,
0.0426025390625,
0.005859375,
-0.05072021484375,
-0.0172882080078125,
-0.035064697265625,
-0.099365234375,
0.01232147216796875,
-0.013336181640625,
-0.057769775390625,
0.0295257568359375,
0.013458251953125,
-0.040557861328125,
0.020416259765625,
-0.049896240234375,
0.07916259765625,
-0.01561737060546875,
-0.02154541015625,
-0.0113067626953125,
-0.06341552734375,
0.0300445556640625,
0.0140838623046875,
0.00870513916015625,
-0.0016889572143554688,
-0.007808685302734375,
0.061279296875,
-0.049072265625,
0.0682373046875,
-0.01074981689453125,
-0.0124053955078125,
0.03790283203125,
-0.0031795501708984375,
0.0418701171875,
-0.0027828216552734375,
-0.016571044921875,
0.039459228515625,
-0.0227203369140625,
-0.032501220703125,
-0.0207672119140625,
0.049468994140625,
-0.08917236328125,
-0.0241546630859375,
-0.036865234375,
-0.02581787109375,
0.00868988037109375,
0.004634857177734375,
0.01873779296875,
0.034454345703125,
-0.0036792755126953125,
0.006862640380859375,
0.031951904296875,
-0.0247650146484375,
0.0271453857421875,
0.0297393798828125,
-0.011016845703125,
-0.02862548828125,
0.0628662109375,
0.0255584716796875,
0.005672454833984375,
0.0157012939453125,
0.00998687744140625,
-0.0239715576171875,
-0.049713134765625,
-0.0249786376953125,
0.047149658203125,
-0.03985595703125,
-0.01763916015625,
-0.04571533203125,
-0.00559234619140625,
-0.0239410400390625,
0.00701141357421875,
-0.034759521484375,
-0.038787841796875,
-0.046112060546875,
-0.01395416259765625,
0.036834716796875,
0.04962158203125,
-0.006465911865234375,
0.02484130859375,
-0.0245361328125,
-0.00275421142578125,
0.01324462890625,
0.01129913330078125,
0.0175628662109375,
-0.0538330078125,
-0.01439666748046875,
0.01357269287109375,
-0.055328369140625,
-0.033050537109375,
0.0280914306640625,
0.0227203369140625,
0.035125732421875,
0.033355712890625,
0.007740020751953125,
0.0677490234375,
-0.01110076904296875,
0.07012939453125,
0.007785797119140625,
-0.046875,
0.041900634765625,
-0.0252838134765625,
0.00836944580078125,
0.033233642578125,
0.0303802490234375,
-0.01206207275390625,
-0.01898193359375,
-0.07171630859375,
-0.07806396484375,
0.07568359375,
0.01430511474609375,
-0.003787994384765625,
0.0089874267578125,
0.04254150390625,
0.01043701171875,
0.009735107421875,
-0.05352783203125,
-0.0302276611328125,
-0.0251007080078125,
-0.000049948692321777344,
-0.015411376953125,
-0.004566192626953125,
-0.01378631591796875,
-0.0207672119140625,
0.05419921875,
-0.0039215087890625,
0.0372314453125,
0.00814056396484375,
0.0122528076171875,
-0.0007228851318359375,
-0.0082855224609375,
0.057159423828125,
0.05169677734375,
-0.0200958251953125,
-0.0202484130859375,
0.00919342041015625,
-0.041778564453125,
-0.016357421875,
0.021942138671875,
-0.00112152099609375,
-0.0225067138671875,
0.0262908935546875,
0.08087158203125,
-0.0188140869140625,
-0.037628173828125,
0.028839111328125,
-0.00243377685546875,
-0.0183868408203125,
-0.0195465087890625,
0.0152130126953125,
-0.0026950836181640625,
0.0299530029296875,
0.0182037353515625,
0.0078582763671875,
-0.0064239501953125,
-0.045196533203125,
-0.017852783203125,
0.0149688720703125,
-0.005413055419921875,
-0.04473876953125,
0.06671142578125,
0.0041351318359375,
0.0016574859619140625,
0.056915283203125,
-0.00408172607421875,
-0.0171051025390625,
0.056732177734375,
0.0173492431640625,
0.042022705078125,
-0.0125732421875,
0.0005064010620117188,
0.046112060546875,
0.01593017578125,
-0.0168609619140625,
0.0217437744140625,
0.0026092529296875,
-0.0369873046875,
-0.031646728515625,
-0.037567138671875,
-0.029510498046875,
0.023406982421875,
-0.0504150390625,
0.029052734375,
-0.048095703125,
-0.0125732421875,
0.00322723388671875,
0.014129638671875,
-0.056365966796875,
0.01183319091796875,
0.01081085205078125,
0.07720947265625,
-0.04876708984375,
0.06646728515625,
0.05718994140625,
-0.060089111328125,
-0.09442138671875,
-0.022796630859375,
0.00045680999755859375,
-0.0703125,
0.039093017578125,
0.024017333984375,
0.0010576248168945312,
-0.004669189453125,
-0.057525634765625,
-0.0687255859375,
0.1005859375,
0.06195068359375,
-0.034210205078125,
-0.010833740234375,
-0.00011116266250610352,
0.05059814453125,
-0.0242919921875,
0.045013427734375,
0.039825439453125,
0.03533935546875,
0.01666259765625,
-0.07275390625,
0.0126495361328125,
-0.0251922607421875,
-0.007419586181640625,
0.0082244873046875,
-0.08685302734375,
0.0830078125,
-0.0215606689453125,
-0.0158233642578125,
0.011871337890625,
0.054931640625,
0.025787353515625,
0.0172882080078125,
0.0298309326171875,
0.06365966796875,
0.06524658203125,
-0.019622802734375,
0.0941162109375,
-0.0179290771484375,
0.02581787109375,
0.07354736328125,
0.0013761520385742188,
0.0572509765625,
0.01219940185546875,
-0.0198211669921875,
0.048919677734375,
0.0714111328125,
0.00991058349609375,
0.03802490234375,
0.007190704345703125,
0.00608062744140625,
-0.00995635986328125,
-0.00787353515625,
-0.06005859375,
0.03656005859375,
0.0236968994140625,
-0.0180816650390625,
-0.0173492431640625,
-0.00963592529296875,
0.0222015380859375,
-0.01329803466796875,
-0.01343536376953125,
0.044342041015625,
0.01064300537109375,
-0.0460205078125,
0.09564208984375,
-0.0007700920104980469,
0.04852294921875,
-0.048187255859375,
0.01334381103515625,
-0.0413818359375,
0.0182037353515625,
-0.023681640625,
-0.039794921875,
0.0016984939575195312,
-0.005237579345703125,
0.0101165771484375,
-0.0096435546875,
0.0200347900390625,
-0.0112152099609375,
-0.0173187255859375,
0.0246734619140625,
0.0248260498046875,
0.02923583984375,
0.0014944076538085938,
-0.058868408203125,
0.0211029052734375,
0.0069122314453125,
-0.04443359375,
0.0291595458984375,
0.036041259765625,
-0.016571044921875,
0.045928955078125,
0.061553955078125,
-0.005802154541015625,
0.004596710205078125,
-0.01088714599609375,
0.0916748046875,
-0.02947998046875,
-0.031097412109375,
-0.0609130859375,
0.0343017578125,
0.004802703857421875,
-0.05194091796875,
0.0482177734375,
0.04718017578125,
0.08758544921875,
0.0179443359375,
0.035980224609375,
-0.0165557861328125,
0.0199127197265625,
-0.029693603515625,
0.04583740234375,
-0.05804443359375,
0.0199127197265625,
-0.01983642578125,
-0.071533203125,
-0.0170440673828125,
0.058807373046875,
-0.0313720703125,
0.01233673095703125,
0.036468505859375,
0.067138671875,
-0.01104736328125,
0.00670623779296875,
-0.00728607177734375,
0.026641845703125,
0.038848876953125,
0.054046630859375,
0.043853759765625,
-0.040069580078125,
0.06817626953125,
-0.0225677490234375,
-0.0220794677734375,
-0.0223388671875,
-0.059326171875,
-0.068359375,
-0.028778076171875,
-0.0244140625,
-0.033782958984375,
0.01227569580078125,
0.05963134765625,
0.049957275390625,
-0.046600341796875,
-0.03033447265625,
-0.00830841064453125,
-0.00495147705078125,
-0.03509521484375,
-0.008026123046875,
0.046417236328125,
0.0043182373046875,
-0.04876708984375,
0.0248565673828125,
-0.00788116455078125,
0.0209808349609375,
-0.0113983154296875,
-0.0225067138671875,
-0.01259613037109375,
-0.0034923553466796875,
0.0248870849609375,
0.0535888671875,
-0.03839111328125,
-0.0150909423828125,
0.004528045654296875,
-0.00804901123046875,
0.027923583984375,
0.02667236328125,
-0.06085205078125,
0.02313232421875,
0.01666259765625,
0.0263519287109375,
0.07244873046875,
-0.0036163330078125,
0.0169677734375,
-0.0386962890625,
0.03948974609375,
0.00469207763671875,
0.0212860107421875,
0.019134521484375,
-0.0010280609130859375,
0.07232666015625,
0.01274871826171875,
-0.03753662109375,
-0.07415771484375,
-0.00513458251953125,
-0.09002685546875,
-0.00008094310760498047,
0.071044921875,
-0.0291900634765625,
-0.0305633544921875,
0.0200653076171875,
-0.020721435546875,
0.027801513671875,
-0.058380126953125,
0.058380126953125,
0.02947998046875,
-0.006275177001953125,
0.0012884140014648438,
-0.042236328125,
0.01538848876953125,
0.0153656005859375,
-0.06072998046875,
-0.0107269287109375,
0.038330078125,
0.006237030029296875,
0.028289794921875,
0.0355224609375,
-0.019195556640625,
0.01087188720703125,
-0.00982666015625,
0.0214691162109375,
-0.03985595703125,
-0.019195556640625,
-0.01580810546875,
0.00785064697265625,
0.002307891845703125,
-0.0231170654296875
]
] |
KnutJaegersberg/RWKV-4-PilePlus-1B5-20230520-2942-486Gtokens-ctx4096 | 2023-09-05T10:00:35.000Z | [
"transformers",
"pytorch",
"rwkv",
"license:apache-2.0",
"endpoints_compatible",
"region:us"
] | null | KnutJaegersberg | null | null | KnutJaegersberg/RWKV-4-PilePlus-1B5-20230520-2942-486Gtokens-ctx4096 | 0 | 5,910 | transformers | 2023-09-05T09:09:08 | ---
license: apache-2.0
---
This is just a standard conversion to hf transformers format of models from here:
https://huggingface.co/BlinkDL/rwkv-4-pileplus
According to the documentation I found, this model should have seen roundabout roundabout 0.8 trillion tokens! | 270 | [
[
-0.038177490234375,
-0.055908203125,
0.019195556640625,
0.0243988037109375,
-0.029449462890625,
-0.025238037109375,
0.034149169921875,
-0.007358551025390625,
0.0284881591796875,
0.07391357421875,
-0.05810546875,
-0.03204345703125,
-0.00916290283203125,
-0.00307464599609375,
-0.066162109375,
0.065185546875,
0.0216827392578125,
-0.0028133392333984375,
-0.00862884521484375,
-0.005512237548828125,
0.004070281982421875,
-0.01849365234375,
-0.034576416015625,
-0.02410888671875,
0.053619384765625,
0.04840087890625,
0.0673828125,
0.046417236328125,
0.05322265625,
0.014404296875,
0.0166015625,
-0.039276123046875,
-0.035614013671875,
-0.0440673828125,
0.00921630859375,
-0.01041412353515625,
-0.0556640625,
0.040435791015625,
0.058990478515625,
0.03936767578125,
-0.0433349609375,
0.0196075439453125,
-0.031890869140625,
0.04449462890625,
-0.0226898193359375,
0.0028095245361328125,
0.00010925531387329102,
0.0301513671875,
-0.02410888671875,
-0.01409912109375,
-0.03997802734375,
-0.007266998291015625,
-0.005901336669921875,
-0.0762939453125,
0.0220489501953125,
0.0421142578125,
0.074462890625,
0.024169921875,
-0.06494140625,
0.01090240478515625,
-0.05157470703125,
0.03472900390625,
-0.052978515625,
0.043212890625,
-0.005512237548828125,
0.05865478515625,
-0.037933349609375,
-0.06976318359375,
-0.052001953125,
0.006519317626953125,
-0.0168914794921875,
0.00649261474609375,
-0.003749847412109375,
0.01483154296875,
0.039794921875,
0.044708251953125,
-0.04046630859375,
-0.0106353759765625,
-0.0489501953125,
-0.018890380859375,
0.05377197265625,
0.017791748046875,
0.0093231201171875,
-0.061676025390625,
-0.0615234375,
-0.02001953125,
-0.0589599609375,
-0.0016069412231445312,
0.0271148681640625,
0.00498199462890625,
-0.053253173828125,
0.06390380859375,
-0.01580810546875,
0.04583740234375,
0.02996826171875,
-0.007022857666015625,
0.025848388671875,
-0.0139312744140625,
-0.0279083251953125,
-0.0106353759765625,
0.052337646484375,
0.0299072265625,
0.01190948486328125,
-0.0075225830078125,
-0.03277587890625,
-0.060302734375,
0.049957275390625,
-0.07379150390625,
-0.01512908935546875,
-0.00984954833984375,
-0.0599365234375,
-0.03704833984375,
0.032012939453125,
-0.050201416015625,
0.0010042190551757812,
0.008514404296875,
0.029510498046875,
-0.00872039794921875,
-0.041778564453125,
0.013336181640625,
-0.0260772705078125,
0.04730224609375,
0.013336181640625,
-0.0389404296875,
0.01047515869140625,
0.0360107421875,
0.03387451171875,
-0.0109100341796875,
-0.0147857666015625,
-0.0258941650390625,
0.021209716796875,
-0.020416259765625,
0.06634521484375,
-0.00855255126953125,
-0.05224609375,
0.0239410400390625,
0.022064208984375,
0.00803375244140625,
-0.026519775390625,
0.062164306640625,
-0.0482177734375,
0.02825927734375,
0.004283905029296875,
-0.0168304443359375,
-0.040740966796875,
0.01337432861328125,
-0.07855224609375,
0.08245849609375,
0.0582275390625,
-0.05902099609375,
0.03900146484375,
-0.061614990234375,
-0.038055419921875,
0.0273895263671875,
0.004543304443359375,
-0.02880859375,
-0.0022068023681640625,
-0.0144500732421875,
0.0160369873046875,
-0.0231781005859375,
0.026397705078125,
-0.0196990966796875,
-0.0369873046875,
0.01271820068359375,
-0.0258331298828125,
0.059661865234375,
0.031768798828125,
-0.002483367919921875,
0.05340576171875,
-0.07666015625,
-0.0290679931640625,
-0.0007309913635253906,
-0.023345947265625,
0.0147705078125,
-0.0259246826171875,
0.029327392578125,
0.03875732421875,
0.015625,
-0.023681640625,
0.0209808349609375,
-0.0111541748046875,
0.032073974609375,
0.0253143310546875,
0.0006117820739746094,
0.039947509765625,
-0.01708984375,
0.055938720703125,
0.035797119140625,
0.034637451171875,
0.004085540771484375,
-0.07647705078125,
-0.07708740234375,
-0.0118560791015625,
0.0093841552734375,
0.0079498291015625,
-0.041473388671875,
0.035858154296875,
0.000217437744140625,
-0.042449951171875,
-0.0205230712890625,
-0.046783447265625,
-0.029815673828125,
0.044921875,
0.00107574462890625,
-0.0096588134765625,
-0.026214599609375,
-0.0791015625,
0.0215301513671875,
-0.0142364501953125,
-0.0088653564453125,
0.00804901123046875,
0.05035400390625,
-0.0223846435546875,
0.06170654296875,
-0.00531768798828125,
-0.004688262939453125,
-0.0164337158203125,
0.033721923828125,
0.0264892578125,
0.04095458984375,
0.041656494140625,
-0.0482177734375,
-0.0169677734375,
-0.006412506103515625,
-0.027801513671875,
0.015350341796875,
-0.00910186767578125,
-0.035614013671875,
0.00460052490234375,
0.045867919921875,
-0.08148193359375,
0.05535888671875,
0.043212890625,
-0.034759521484375,
0.0271759033203125,
-0.00015020370483398438,
0.0157928466796875,
-0.0634765625,
0.01995849609375,
-0.004688262939453125,
-0.04473876953125,
-0.0230255126953125,
0.021820068359375,
0.020050048828125,
-0.00887298583984375,
-0.04034423828125,
0.0157928466796875,
-0.0240478515625,
-0.0421142578125,
0.0092010498046875,
-0.0004825592041015625,
-0.0256805419921875,
0.036376953125,
-0.016998291015625,
0.049957275390625,
0.0255279541015625,
-0.0286712646484375,
0.048004150390625,
0.035003662109375,
0.00719451904296875,
0.04437255859375,
-0.041839599609375,
0.00441741943359375,
-0.0018644332885742188,
0.041656494140625,
-0.0601806640625,
-0.048828125,
0.038970947265625,
-0.00597381591796875,
-0.0022602081298828125,
-0.00391387939453125,
-0.03582763671875,
-0.05389404296875,
-0.0207366943359375,
0.05596923828125,
0.052459716796875,
-0.038055419921875,
0.06524658203125,
0.0328369140625,
0.015960693359375,
-0.01105499267578125,
-0.034149169921875,
-0.0026874542236328125,
-0.038116455078125,
-0.054656982421875,
0.035491943359375,
0.008880615234375,
0.0015935897827148438,
-0.0053253173828125,
0.021820068359375,
-0.0220184326171875,
-0.02178955078125,
0.033599853515625,
0.01273345947265625,
-0.0307159423828125,
-0.02899169921875,
0.0004029273986816406,
-0.0160980224609375,
0.023590087890625,
0.0056915283203125,
0.03302001953125,
0.0016231536865234375,
-0.01378631591796875,
-0.043914794921875,
0.0215301513671875,
0.0645751953125,
0.00804901123046875,
0.032135009765625,
0.0673828125,
-0.046844482421875,
-0.012664794921875,
-0.0267791748046875,
-0.023529052734375,
-0.035369873046875,
-0.0012311935424804688,
-0.03802490234375,
-0.066650390625,
0.034515380859375,
-0.002574920654296875,
0.00225830078125,
0.08837890625,
0.044097900390625,
-0.002048492431640625,
0.0811767578125,
0.056976318359375,
0.027862548828125,
-0.0006403923034667969,
-0.0109710693359375,
0.043701171875,
-0.0274200439453125,
-0.034759521484375,
-0.030853271484375,
-0.0201568603515625,
-0.04058837890625,
-0.04736328125,
0.0238189697265625,
0.0216217041015625,
-0.044036865234375,
0.0296173095703125,
-0.03680419921875,
0.00989532470703125,
0.043121337890625,
-0.019989013671875,
0.041168212890625,
-0.0211029052734375,
0.008453369140625,
0.00211334228515625,
-0.07342529296875,
-0.022979736328125,
0.0560302734375,
0.052398681640625,
0.037750244140625,
0.0003418922424316406,
0.052154541015625,
0.0108184814453125,
0.036712646484375,
-0.0650634765625,
0.048065185546875,
0.018524169921875,
-0.0745849609375,
-0.03668212890625,
-0.0134429931640625,
-0.047943115234375,
-0.0070343017578125,
-0.022979736328125,
-0.00797271728515625,
-0.0007715225219726562,
-0.01015472412109375,
-0.028411865234375,
0.01346588134765625,
-0.0294342041015625,
0.091552734375,
0.01776123046875,
0.041656494140625,
-0.04351806640625,
-0.032501220703125,
0.058502197265625,
0.0109100341796875,
0.0228729248046875,
0.01430511474609375,
0.0191192626953125,
0.06427001953125,
-0.0251007080078125,
0.0283355712890625,
-0.004184722900390625,
-0.0099639892578125,
0.042877197265625,
0.00353240966796875,
0.02532958984375,
0.0163116455078125,
0.0202178955078125,
0.0007610321044921875,
-0.0078277587890625,
-0.031494140625,
-0.01003265380859375,
0.041534423828125,
-0.05377197265625,
-0.0330810546875,
-0.0396728515625,
-0.041534423828125,
0.01549530029296875,
0.01708984375,
0.020660400390625,
0.0192108154296875,
-0.034332275390625,
0.01213836669921875,
0.03887939453125,
0.01554107666015625,
0.04425048828125,
0.019775390625,
-0.04449462890625,
-0.01212310791015625,
0.03662109375,
-0.02191162109375,
0.0019855499267578125,
0.011444091796875,
0.003894805908203125,
-0.047882080078125,
-0.032012939453125,
-0.055206298828125,
0.032318115234375,
-0.05206298828125,
-0.0096893310546875,
-0.02801513671875,
-0.035186767578125,
-0.057159423828125,
-0.0020809173583984375,
-0.023193359375,
-0.060791015625,
-0.0374755859375,
-0.0209503173828125,
0.04022216796875,
0.099853515625,
-0.013092041015625,
0.032135009765625,
-0.061004638671875,
0.020477294921875,
0.03509521484375,
0.0298309326171875,
-0.00995635986328125,
-0.0673828125,
-0.0210113525390625,
-0.0187530517578125,
-0.01531982421875,
-0.057830810546875,
0.027435302734375,
-0.01192474365234375,
0.02593994140625,
0.045806884765625,
-0.0183563232421875,
0.045379638671875,
-0.05426025390625,
0.059661865234375,
0.032196044921875,
-0.042266845703125,
-0.0196685791015625,
-0.05047607421875,
-0.00931549072265625,
0.004779815673828125,
0.030517578125,
-0.04461669921875,
0.00377655029296875,
-0.08013916015625,
-0.0333251953125,
0.046630859375,
-0.00017321109771728516,
-0.0174713134765625,
-0.0009174346923828125,
0.0298614501953125,
0.0121307373046875,
0.0203094482421875,
-0.051239013671875,
-0.0179443359375,
-0.040496826171875,
-0.01522064208984375,
0.019927978515625,
-0.0112762451171875,
-0.0177459716796875,
-0.0262298583984375,
0.02593994140625,
-0.0027904510498046875,
0.0333251953125,
-0.0035648345947265625,
0.002826690673828125,
-0.0328369140625,
0.001636505126953125,
0.04266357421875,
0.024505615234375,
-0.0202178955078125,
-0.0032939910888671875,
-0.00463104248046875,
-0.0280609130859375,
-0.020477294921875,
-0.00910186767578125,
-0.0055389404296875,
-0.00519561767578125,
-0.0015048980712890625,
0.034332275390625,
0.04095458984375,
-0.0186767578125,
0.0413818359375,
0.0182647705078125,
-0.034515380859375,
-0.05780029296875,
0.003566741943359375,
0.0284881591796875,
0.01058197021484375,
0.0303497314453125,
0.0028324127197265625,
0.031402587890625,
-0.031585693359375,
0.037933349609375,
0.0186309814453125,
-0.032440185546875,
-0.042694091796875,
0.0594482421875,
0.05413818359375,
-0.048736572265625,
0.060821533203125,
-0.049041748046875,
-0.015350341796875,
0.02349853515625,
0.043914794921875,
0.072265625,
0.0010509490966796875,
0.045013427734375,
0.055084228515625,
0.005462646484375,
-0.00327301025390625,
0.0199737548828125,
0.009613037109375,
-0.038330078125,
-0.032623291015625,
-0.051666259765625,
-0.02337646484375,
0.0005645751953125,
-0.053680419921875,
0.039459228515625,
-0.058319091796875,
0.0016689300537109375,
-0.01403045654296875,
-0.038726806640625,
-0.045379638671875,
0.025238037109375,
0.015960693359375,
0.09423828125,
-0.046630859375,
0.07049560546875,
0.04473876953125,
-0.0303955078125,
-0.061859130859375,
-0.010955810546875,
-0.01377105712890625,
-0.06964111328125,
0.036712646484375,
0.01187896728515625,
0.003795623779296875,
0.0186309814453125,
-0.053009033203125,
-0.05926513671875,
0.0810546875,
-0.016265869140625,
-0.052215576171875,
-0.0248870849609375,
0.040435791015625,
0.0173797607421875,
-0.04876708984375,
0.0249176025390625,
0.053985595703125,
0.056182861328125,
0.0207672119140625,
-0.049560546875,
0.01419830322265625,
-0.008087158203125,
0.0013275146484375,
0.02166748046875,
-0.05047607421875,
0.0806884765625,
0.01058197021484375,
-0.01500701904296875,
0.00890350341796875,
0.060302734375,
0.0191650390625,
0.00527191162109375,
0.06585693359375,
0.058074951171875,
0.01837158203125,
-0.018524169921875,
0.06060791015625,
-0.0194549560546875,
0.040374755859375,
0.002227783203125,
-0.0183868408203125,
0.055511474609375,
0.060791015625,
-0.01934814453125,
0.06060791015625,
0.0462646484375,
0.0010805130004882812,
0.01080322265625,
0.004276275634765625,
0.01491546630859375,
-0.04119873046875,
0.0079345703125,
-0.01090240478515625,
0.0158233642578125,
0.0323486328125,
0.025970458984375,
-0.0191802978515625,
-0.01788330078125,
-0.02044677734375,
-0.00713348388671875,
-0.0238037109375,
0.0245513916015625,
0.00330352783203125,
-0.026580810546875,
0.01910400390625,
0.006938934326171875,
0.04840087890625,
-0.002307891845703125,
-0.0248565673828125,
0.01224517822265625,
-0.01430511474609375,
-0.02276611328125,
-0.060211181640625,
0.03759765625,
-0.023681640625,
-0.00445556640625,
-0.0103759765625,
0.039215087890625,
-0.016632080078125,
-0.05413818359375,
0.004444122314453125,
0.0204620361328125,
0.0289154052734375,
0.00728607177734375,
-0.07342529296875,
-0.031341552734375,
0.005374908447265625,
-0.0291290283203125,
0.006244659423828125,
0.035736083984375,
-0.015899658203125,
0.027862548828125,
0.0280609130859375,
-0.032379150390625,
0.0133514404296875,
0.02264404296875,
0.032623291015625,
-0.092529296875,
-0.05035400390625,
0.008697509765625,
0.0677490234375,
-0.01407623291015625,
-0.05712890625,
0.036956787109375,
0.031463623046875,
0.04827880859375,
-0.025238037109375,
0.02496337890625,
-0.00934600830078125,
0.042816162109375,
-0.023529052734375,
0.055511474609375,
-0.044036865234375,
-0.0275115966796875,
0.0106201171875,
-0.0631103515625,
-0.00980377197265625,
0.061309814453125,
0.033416748046875,
0.0009822845458984375,
0.07440185546875,
0.045654296875,
-0.0030059814453125,
0.02130126953125,
0.032196044921875,
0.023406982421875,
0.01275634765625,
0.004138946533203125,
0.05572509765625,
-0.044097900390625,
0.031982421875,
-0.024261474609375,
-0.01200103759765625,
-0.0299835205078125,
-0.0640869140625,
-0.07061767578125,
-0.03839111328125,
-0.0195465087890625,
-0.023468017578125,
-0.0306396484375,
0.034820556640625,
0.056304931640625,
-0.046783447265625,
-0.02899169921875,
-0.03155517578125,
-0.003711700439453125,
0.017822265625,
-0.0189361572265625,
0.008331298828125,
0.00963592529296875,
-0.040191650390625,
0.01459503173828125,
0.026947021484375,
0.0151214599609375,
-0.061187744140625,
-0.00724029541015625,
0.01200103759765625,
0.01033782958984375,
0.030792236328125,
0.00209808349609375,
-0.0501708984375,
-0.0433349609375,
0.01319122314453125,
-0.0233917236328125,
0.0162200927734375,
0.0758056640625,
-0.053558349609375,
-0.035919189453125,
0.04510498046875,
-0.0032482147216796875,
0.061614990234375,
0.00955963134765625,
0.047515869140625,
-0.03216552734375,
0.04779052734375,
0.01873779296875,
0.024078369140625,
0.018218994140625,
-0.0259857177734375,
0.013824462890625,
0.033416748046875,
-0.042266845703125,
-0.043853759765625,
0.0284423828125,
-0.117431640625,
0.0052642822265625,
0.060272216796875,
0.010009765625,
-0.046051025390625,
0.00438690185546875,
-0.01824951171875,
0.0207061767578125,
-0.028228759765625,
0.047027587890625,
0.034637451171875,
0.0147705078125,
-0.0245513916015625,
-0.039093017578125,
0.039398193359375,
0.01039886474609375,
-0.038055419921875,
-0.014404296875,
0.0260009765625,
0.0021686553955078125,
0.026580810546875,
0.0243988037109375,
-0.0341796875,
0.032012939453125,
0.00891876220703125,
0.013824462890625,
0.0269317626953125,
-0.0261993408203125,
-0.01922607421875,
-0.00040984153747558594,
-0.0123443603515625,
0.01256561279296875
]
] |
amberoad/bert-multilingual-passage-reranking-msmarco | 2022-08-26T13:14:54.000Z | [
"transformers",
"pytorch",
"tf",
"jax",
"bert",
"text-classification",
"msmarco",
"multilingual",
"passage reranking",
"af",
"sq",
"ar",
"an",
"hy",
"ast",
"az",
"ba",
"eu",
"bar",
"be",
"bn",
"inc",
"bs",
"br",
"bg",
"my",
"ca",
"ceb",
"ce",
"zh",
"cv",
"hr",
"cs",
"da",
"nl",
"en",
"et",
"fi",
"fr",
"gl",
"ka",
"de",
"el",
"gu",
"ht",
"he",
"hi",
"hu",
"is",
"io",
"id",
"ga",
"it",
"ja",
"jv",
"kn",
"kk",
"ky",
"ko",
"la",
"lv",
"lt",
"roa",
"nds",
"lm",
"mk",
"mg",
"ms",
"ml",
"mr",
"min",
"ne",
"new",
"nb",
"nn",
"oc",
"fa",
"pms",
"pl",
"pt",
"pa",
"ro",
"ru",
"sco",
"sr",
"scn",
"sk",
"sl",
"aze",
"es",
"su",
"sw",
"sv",
"tl",
"tg",
"ta",
"tt",
"te",
"tr",
"uk",
"ud",
"uz",
"vi",
"vo",
"war",
"cy",
"fry",
"pnb",
"yo",
"dataset:msmarco",
"arxiv:1901.04085",
"license:apache-2.0",
"endpoints_compatible",
"has_space",
"region:us"
] | text-classification | amberoad | null | null | amberoad/bert-multilingual-passage-reranking-msmarco | 38 | 5,909 | transformers | 2022-03-02T23:29:05 | ---
language:
- multilingual
- af
- sq
- ar
- an
- hy
- ast
- az
- ba
- eu
- bar
- be
- bn
- inc
- bs
- br
- bg
- my
- ca
- ceb
- ce
- zh
- cv
- hr
- cs
- da
- nl
- en
- et
- fi
- fr
- gl
- ka
- de
- el
- gu
- ht
- he
- hi
- hu
- is
- io
- id
- ga
- it
- ja
- jv
- kn
- kk
- ky
- ko
- la
- lv
- lt
- roa
- nds
- lm
- mk
- mg
- ms
- ml
- mr
- min
- ne
- new
- nb
- nn
- oc
- fa
- pms
- pl
- pt
- pa
- ro
- ru
- sco
- sr
- hr
- scn
- sk
- sl
- aze
- es
- su
- sw
- sv
- tl
- tg
- ta
- tt
- te
- tr
- uk
- ud
- uz
- vi
- vo
- war
- cy
- fry
- pnb
- yo
thumbnail: https://amberoad.de/images/logo_text.png
tags:
- msmarco
- multilingual
- passage reranking
license: apache-2.0
datasets:
- msmarco
metrics:
- MRR
widget:
- query: What is a corporation?
passage: A company is incorporated in a specific nation, often within the bounds
of a smaller subset of that nation, such as a state or province. The corporation
is then governed by the laws of incorporation in that state. A corporation may
issue stock, either private or public, or may be classified as a non-stock corporation.
If stock is issued, the corporation will usually be governed by its shareholders,
either directly or indirectly.
---
# Passage Reranking Multilingual BERT 🔃 🌍
## Model description
**Input:** Supports over 100 Languages. See [List of supported languages](https://github.com/google-research/bert/blob/master/multilingual.md#list-of-languages) for all available.
**Purpose:** This module takes a search query [1] and a passage [2] and calculates if the passage matches the query.
It can be used as an improvement for Elasticsearch Results and boosts the relevancy by up to 100%.
**Architecture:** On top of BERT there is a Densly Connected NN which takes the 768 Dimensional [CLS] Token as input and provides the output ([Arxiv](https://arxiv.org/abs/1901.04085)).
**Output:** Just a single value between between -10 and 10. Better matching query,passage pairs tend to have a higher a score.
## Intended uses & limitations
Both query[1] and passage[2] have to fit in 512 Tokens.
As you normally want to rerank the first dozens of search results keep in mind the inference time of approximately 300 ms/query.
#### How to use
```python
from transformers import AutoTokenizer, AutoModelForSequenceClassification
tokenizer = AutoTokenizer.from_pretrained("amberoad/bert-multilingual-passage-reranking-msmarco")
model = AutoModelForSequenceClassification.from_pretrained("amberoad/bert-multilingual-passage-reranking-msmarco")
```
This Model can be used as a drop-in replacement in the [Nboost Library](https://github.com/koursaros-ai/nboost)
Through this you can directly improve your Elasticsearch Results without any coding.
## Training data
This model is trained using the [**Microsoft MS Marco Dataset**](https://microsoft.github.io/msmarco/ "Microsoft MS Marco"). This training dataset contains approximately 400M tuples of a query, relevant and non-relevant passages. All datasets used for training and evaluating are listed in this [table](https://github.com/microsoft/MSMARCO-Passage-Ranking#data-information-and-formating). The used dataset for training is called *Train Triples Large*, while the evaluation was made on *Top 1000 Dev*. There are 6,900 queries in total in the development dataset, where each query is mapped to top 1,000 passage retrieved using BM25 from MS MARCO corpus.
## Training procedure
The training is performed the same way as stated in this [README](https://github.com/nyu-dl/dl4marco-bert "NYU Github"). See their excellent Paper on [Arxiv](https://arxiv.org/abs/1901.04085).
We changed the BERT Model from an English only to the default BERT Multilingual uncased Model from [Google](https://huggingface.co/bert-base-multilingual-uncased).
Training was done 400 000 Steps. This equaled 12 hours an a TPU V3-8.
## Eval results
We see nearly similar performance than the English only Model in the English [Bing Queries Dataset](http://www.msmarco.org/). Although the training data is English only internal Tests on private data showed a far higher accurancy in German than all other available models.
Fine-tuned Models | Dependency | Eval Set | Search Boost<a href='#benchmarks'> | Speed on GPU
----------------------------------------------------------------------------------- | ---------------------------------------------------------------------------- | ------------------------------------------------------------------ | ----------------------------------------------------- | ----------------------------------
**`amberoad/Multilingual-uncased-MSMARCO`** (This Model) | <img alt="PyTorch" src="https://img.shields.io/badge/PyTorch-blue"/> | <a href ='http://www.msmarco.org/'>bing queries</a> | **+61%** <sub><sup>(0.29 vs 0.18)</sup></sub> | ~300 ms/query <a href='#footnotes'>
`nboost/pt-tinybert-msmarco` | <img alt="PyTorch" src="https://img.shields.io/badge/PyTorch-red"/> | <a href ='http://www.msmarco.org/'>bing queries</a> | **+45%** <sub><sup>(0.26 vs 0.18)</sup></sub> | ~50ms/query <a href='#footnotes'>
`nboost/pt-bert-base-uncased-msmarco` | <img alt="PyTorch" src="https://img.shields.io/badge/PyTorch-red"/> | <a href ='http://www.msmarco.org/'>bing queries</a> | **+62%** <sub><sup>(0.29 vs 0.18)</sup></sub> | ~300 ms/query<a href='#footnotes'>
`nboost/pt-bert-large-msmarco` | <img alt="PyTorch" src="https://img.shields.io/badge/PyTorch-red"/> | <a href ='http://www.msmarco.org/'>bing queries</a> | **+77%** <sub><sup>(0.32 vs 0.18)</sup></sub> | -
`nboost/pt-biobert-base-msmarco` | <img alt="PyTorch" src="https://img.shields.io/badge/PyTorch-red"/> | <a href ='https://github.com/naver/biobert-pretrained'>biomed</a> | **+66%** <sub><sup>(0.17 vs 0.10)</sup></sub> | ~300 ms/query<a href='#footnotes'>
This table is taken from [nboost](https://github.com/koursaros-ai/nboost) and extended by the first line.
## Contact Infos
Amberoad is a company focussing on Search and Business Intelligence.
We provide you:
* Advanced Internal Company Search Engines thorugh NLP
* External Search Egnines: Find Competitors, Customers, Suppliers
**Get in Contact now to benefit from our Expertise:**
The training and evaluation was performed by [**Philipp Reissel**](https://reissel.eu/) and [**Igli Manaj**](https://github.com/iglimanaj)
[ Linkedin](https://de.linkedin.com/company/amberoad) | <svg xmlns="http://www.w3.org/2000/svg" x="0px" y="0px"
width="32" height="32"
viewBox="0 0 172 172"
style=" fill:#000000;"><g fill="none" fill-rule="nonzero" stroke="none" stroke-width="1" stroke-linecap="butt" stroke-linejoin="miter" stroke-miterlimit="10" stroke-dasharray="" stroke-dashoffset="0" font-family="none" font-weight="none" font-size="none" text-anchor="none" style="mix-blend-mode: normal"><path d="M0,172v-172h172v172z" fill="none"></path><g fill="#e67e22"><path d="M37.625,21.5v86h96.75v-86h-5.375zM48.375,32.25h10.75v10.75h-10.75zM69.875,32.25h10.75v10.75h-10.75zM91.375,32.25h32.25v10.75h-32.25zM48.375,53.75h75.25v43h-75.25zM80.625,112.875v17.61572c-1.61558,0.93921 -2.94506,2.2687 -3.88428,3.88428h-49.86572v10.75h49.86572c1.8612,3.20153 5.28744,5.375 9.25928,5.375c3.97183,0 7.39808,-2.17347 9.25928,-5.375h49.86572v-10.75h-49.86572c-0.93921,-1.61558 -2.2687,-2.94506 -3.88428,-3.88428v-17.61572z"></path></g></g></svg>[Homepage](https://de.linkedin.com/company/amberoad) | [Email](info@amberoad.de)
| 8,075 | [
[
-0.0272216796875,
-0.037384033203125,
0.0171966552734375,
0.02520751953125,
-0.01149749755859375,
0.00119781494140625,
-0.01099395751953125,
-0.0494384765625,
0.03924560546875,
0.0014438629150390625,
-0.045318603515625,
-0.06121826171875,
-0.03167724609375,
0.00390625,
-0.0142974853515625,
0.060211181640625,
0.006465911865234375,
0.0061798095703125,
0.00276947021484375,
-0.0181884765625,
-0.0249176025390625,
-0.0299530029296875,
-0.049530029296875,
-0.015716552734375,
0.0318603515625,
0.007106781005859375,
0.03875732421875,
0.034759521484375,
0.03131103515625,
0.0247955322265625,
-0.019287109375,
0.0159149169921875,
-0.0269775390625,
0.001308441162109375,
0.006015777587890625,
-0.021881103515625,
-0.03912353515625,
-0.005809783935546875,
0.0413818359375,
0.039154052734375,
0.0033893585205078125,
0.01043701171875,
0.0102996826171875,
0.060638427734375,
-0.0231170654296875,
0.01483154296875,
-0.039520263671875,
-0.003082275390625,
-0.0110931396484375,
-0.007022857666015625,
-0.0203399658203125,
-0.027099609375,
0.02288818359375,
-0.050140380859375,
0.02496337890625,
0.0146026611328125,
0.09454345703125,
0.007427215576171875,
-0.028045654296875,
-0.0008988380432128906,
-0.016326904296875,
0.05322265625,
-0.043609619140625,
0.04742431640625,
0.026031494140625,
0.009796142578125,
-0.00370025634765625,
-0.0552978515625,
-0.0289154052734375,
-0.019927978515625,
-0.0187835693359375,
0.025604248046875,
-0.0106964111328125,
-0.0174407958984375,
0.0133514404296875,
0.0211334228515625,
-0.061737060546875,
-0.0012388229370117188,
-0.037261962890625,
-0.0187530517578125,
0.060089111328125,
0.003002166748046875,
0.0172882080078125,
-0.02667236328125,
-0.0298004150390625,
-0.0234527587890625,
-0.044677734375,
0.0026721954345703125,
0.0229034423828125,
0.0261993408203125,
-0.0126800537109375,
0.02496337890625,
-0.01081085205078125,
0.056640625,
0.00148773193359375,
0.00974273681640625,
0.050323486328125,
-0.041473388671875,
-0.004974365234375,
0.006195068359375,
0.0670166015625,
0.032318115234375,
-0.0019092559814453125,
-0.021575927734375,
-0.0189971923828125,
-0.01678466796875,
0.0213775634765625,
-0.058837890625,
-0.025054931640625,
0.032470703125,
-0.03875732421875,
-0.00949859619140625,
0.005626678466796875,
-0.056488037109375,
0.0024662017822265625,
-0.010345458984375,
0.047882080078125,
-0.0535888671875,
-0.00965118408203125,
0.0017976760864257812,
-0.017913818359375,
0.0301055908203125,
0.021453857421875,
-0.046600341796875,
-0.0014972686767578125,
0.031890869140625,
0.08935546875,
-0.01004791259765625,
-0.0251922607421875,
-0.007663726806640625,
0.0027446746826171875,
-0.0222625732421875,
0.047271728515625,
-0.0203857421875,
-0.0200958251953125,
0.000782012939453125,
0.0232391357421875,
-0.014801025390625,
-0.018890380859375,
0.0457763671875,
-0.0389404296875,
0.0246429443359375,
-0.0165252685546875,
-0.028900146484375,
-0.0190582275390625,
0.0209503173828125,
-0.05609130859375,
0.10638427734375,
0.004970550537109375,
-0.05657958984375,
0.035064697265625,
-0.056793212890625,
-0.032958984375,
-0.0028705596923828125,
0.004329681396484375,
-0.03338623046875,
-0.01531219482421875,
0.03204345703125,
0.0284271240234375,
-0.0208740234375,
-0.0037631988525390625,
-0.0143280029296875,
-0.023651123046875,
-0.00875091552734375,
-0.0166473388671875,
0.0850830078125,
0.01751708984375,
-0.051849365234375,
0.00310516357421875,
-0.0692138671875,
0.0184326171875,
0.0224456787109375,
-0.047088623046875,
-0.009613037109375,
-0.0181732177734375,
0.0167083740234375,
0.0240936279296875,
0.038360595703125,
-0.045196533203125,
0.008392333984375,
-0.04595947265625,
0.037994384765625,
0.065673828125,
-0.0057525634765625,
0.0205230712890625,
-0.031768798828125,
0.0269012451171875,
-0.024749755859375,
0.0106048583984375,
0.001888275146484375,
-0.044708251953125,
-0.05133056640625,
-0.0204315185546875,
0.026611328125,
0.0295562744140625,
-0.04150390625,
0.04840087890625,
-0.0341796875,
-0.058319091796875,
-0.055908203125,
-0.004444122314453125,
0.0257415771484375,
0.0297088623046875,
0.044158935546875,
-0.006343841552734375,
-0.029815673828125,
-0.061431884765625,
-0.0165863037109375,
-0.003452301025390625,
0.005687713623046875,
0.0443115234375,
0.05169677734375,
-0.01427459716796875,
0.051788330078125,
-0.035614013671875,
-0.024017333984375,
-0.005626678466796875,
-0.0030117034912109375,
0.0355224609375,
0.04351806640625,
0.059234619140625,
-0.060150146484375,
-0.063720703125,
0.00370025634765625,
-0.057708740234375,
0.008026123046875,
0.0099945068359375,
-0.0259552001953125,
0.0306243896484375,
0.026275634765625,
-0.037567138671875,
0.032501220703125,
0.0293731689453125,
-0.02880859375,
0.037872314453125,
-0.03192138671875,
0.0146026611328125,
-0.08050537109375,
0.032012939453125,
0.01031494140625,
-0.00293731689453125,
-0.031005859375,
-0.0033245086669921875,
0.00782012939453125,
-0.005706787109375,
-0.03558349609375,
0.05194091796875,
-0.048553466796875,
0.0003838539123535156,
0.01287841796875,
0.020416259765625,
-0.00030803680419921875,
0.042449951171875,
-0.0006113052368164062,
0.07073974609375,
0.047882080078125,
-0.043670654296875,
0.0189361572265625,
0.033172607421875,
-0.03741455078125,
0.0280914306640625,
-0.0655517578125,
0.003559112548828125,
-0.0009641647338867188,
0.0182647705078125,
-0.0826416015625,
-0.00940704345703125,
-0.0008411407470703125,
-0.0540771484375,
0.0210723876953125,
-0.0181121826171875,
-0.042755126953125,
-0.038421630859375,
-0.044769287109375,
0.0119781494140625,
0.028350830078125,
-0.056793212890625,
0.039520263671875,
0.02044677734375,
0.00858306884765625,
-0.053955078125,
-0.0609130859375,
-0.0028285980224609375,
-0.001651763916015625,
-0.07427978515625,
0.056396484375,
-0.0030269622802734375,
0.01019287109375,
0.0037403106689453125,
-0.021392822265625,
-0.007678985595703125,
-0.0019378662109375,
0.016632080078125,
0.025543212890625,
-0.003963470458984375,
0.01212310791015625,
-0.0174713134765625,
-0.00033354759216308594,
-0.0015773773193359375,
-0.0302581787109375,
0.061431884765625,
-0.0220947265625,
-0.01849365234375,
-0.050994873046875,
0.0035953521728515625,
0.057403564453125,
-0.02337646484375,
0.07769775390625,
0.054656982421875,
-0.0186309814453125,
-0.004062652587890625,
-0.0377197265625,
-0.020050048828125,
-0.037109375,
0.0217437744140625,
-0.01543426513671875,
-0.0457763671875,
0.039825439453125,
0.017059326171875,
0.015106201171875,
0.05718994140625,
0.0322265625,
-0.0267486572265625,
0.07672119140625,
0.0284271240234375,
-0.01050567626953125,
0.04681396484375,
-0.061920166015625,
0.011138916015625,
-0.06402587890625,
-0.043182373046875,
-0.049713134765625,
-0.036468505859375,
-0.071044921875,
-0.01477813720703125,
0.033050537109375,
-0.014068603515625,
-0.0341796875,
0.039886474609375,
-0.06512451171875,
0.01041412353515625,
0.0653076171875,
0.0262908935546875,
-0.0086517333984375,
0.019378662109375,
-0.0311126708984375,
-0.005977630615234375,
-0.054443359375,
-0.030242919921875,
0.10357666015625,
0.004180908203125,
0.040252685546875,
0.015533447265625,
0.05401611328125,
0.0179290771484375,
-0.00730133056640625,
-0.031707763671875,
0.037994384765625,
0.0003478527069091797,
-0.06988525390625,
-0.0178375244140625,
-0.029815673828125,
-0.0936279296875,
0.03546142578125,
-0.0280914306640625,
-0.05718994140625,
0.018951416015625,
0.009796142578125,
-0.006221771240234375,
0.02484130859375,
-0.06201171875,
0.068359375,
-0.043701171875,
-0.057708740234375,
-0.010162353515625,
-0.055206298828125,
-0.0097503662109375,
0.0161285400390625,
0.0008478164672851562,
0.0020008087158203125,
0.005939483642578125,
0.0692138671875,
-0.039520263671875,
0.039215087890625,
-0.013458251953125,
0.0140838623046875,
0.01611328125,
-0.00714874267578125,
0.05731201171875,
-0.0067901611328125,
-0.018707275390625,
0.02069091796875,
-0.0041656494140625,
-0.040252685546875,
-0.027099609375,
0.06756591796875,
-0.05999755859375,
-0.033782958984375,
-0.056915283203125,
-0.0227508544921875,
-0.00647735595703125,
0.03204345703125,
0.02838134765625,
0.0175933837890625,
0.002704620361328125,
0.0374755859375,
0.0445556640625,
-0.0242462158203125,
0.030548095703125,
0.029876708984375,
0.00713348388671875,
-0.049163818359375,
0.07623291015625,
0.01532745361328125,
0.0014619827270507812,
0.0316162109375,
0.003894805908203125,
-0.053619384765625,
-0.044586181640625,
-0.01097869873046875,
0.035247802734375,
-0.039886474609375,
-0.00974273681640625,
-0.046478271484375,
-0.0099029541015625,
-0.050323486328125,
-0.0191192626953125,
-0.015106201171875,
-0.033416748046875,
-0.0199432373046875,
-0.0215301513671875,
0.041748046875,
0.034820556640625,
-0.01082611083984375,
0.00763702392578125,
-0.035247802734375,
0.00018310546875,
0.010772705078125,
0.0240936279296875,
0.00775146484375,
-0.031890869140625,
-0.0179901123046875,
0.0102691650390625,
-0.0252838134765625,
-0.06298828125,
0.041107177734375,
-0.001251220703125,
0.06304931640625,
0.01517486572265625,
-0.007354736328125,
0.051513671875,
-0.0274505615234375,
0.0643310546875,
0.0186767578125,
-0.05059814453125,
0.056121826171875,
-0.023895263671875,
0.0041656494140625,
0.055450439453125,
0.046417236328125,
-0.029998779296875,
-0.004863739013671875,
-0.06329345703125,
-0.09808349609375,
0.068115234375,
0.026092529296875,
-0.01200103759765625,
0.00603485107421875,
0.01568603515625,
-0.0237579345703125,
0.0201263427734375,
-0.06671142578125,
-0.04376220703125,
-0.0208740234375,
-0.01200103759765625,
0.0004477500915527344,
-0.0028400421142578125,
-0.004650115966796875,
-0.040771484375,
0.06982421875,
0.0080413818359375,
0.04150390625,
0.024505615234375,
-0.0178070068359375,
0.02880859375,
-0.006084442138671875,
0.040374755859375,
0.053680419921875,
-0.043243408203125,
-0.0023040771484375,
0.0172271728515625,
-0.03875732421875,
-0.007503509521484375,
0.00681304931640625,
-0.017303466796875,
0.0142364501953125,
0.025146484375,
0.04766845703125,
0.005886077880859375,
-0.030120849609375,
0.03466796875,
0.0140228271484375,
-0.02288818359375,
-0.021270751953125,
-0.0011386871337890625,
0.00780487060546875,
0.01548004150390625,
0.0357666015625,
-0.01314544677734375,
0.00368499755859375,
-0.0391845703125,
0.01483154296875,
0.043701171875,
-0.03460693359375,
-0.0144805908203125,
0.059844970703125,
0.01525115966796875,
-0.0193939208984375,
0.048187255859375,
-0.0042572021484375,
-0.051727294921875,
0.0516357421875,
0.039520263671875,
0.057952880859375,
-0.0030803680419921875,
0.012481689453125,
0.06170654296875,
0.039459228515625,
0.00879669189453125,
0.0268402099609375,
-0.0091400146484375,
-0.0255889892578125,
-0.0017414093017578125,
-0.050384521484375,
-0.0086669921875,
0.0085296630859375,
-0.04327392578125,
0.0172271728515625,
-0.032379150390625,
-0.0203857421875,
0.0002703666687011719,
0.03143310546875,
-0.055755615234375,
0.023712158203125,
-0.0251007080078125,
0.0823974609375,
-0.049224853515625,
0.048004150390625,
0.057830810546875,
-0.056304931640625,
-0.04852294921875,
0.0017766952514648438,
-0.0156097412109375,
-0.069580078125,
0.061737060546875,
0.01371002197265625,
-0.01125335693359375,
-0.0014123916625976562,
-0.04150390625,
-0.0679931640625,
0.10357666015625,
0.01910400390625,
-0.023834228515625,
0.0004265308380126953,
-0.00585174560546875,
0.0443115234375,
-0.004184722900390625,
0.0311737060546875,
0.028350830078125,
0.04632568359375,
0.00518798828125,
-0.06463623046875,
0.0107574462890625,
-0.032623291015625,
-0.007366180419921875,
0.00948333740234375,
-0.07269287109375,
0.0615234375,
0.00518798828125,
-0.00540924072265625,
-0.0206298828125,
0.0303192138671875,
0.02142333984375,
0.0189361572265625,
0.01380157470703125,
0.0546875,
0.06341552734375,
-0.02301025390625,
0.07867431640625,
-0.05029296875,
0.03857421875,
0.059906005859375,
0.0107879638671875,
0.06451416015625,
0.0278472900390625,
-0.028564453125,
0.03472900390625,
0.062225341796875,
-0.01320648193359375,
0.049713134765625,
-0.0031108856201171875,
-0.01401519775390625,
-0.00949859619140625,
0.01079559326171875,
-0.050506591796875,
0.0091705322265625,
0.01021575927734375,
-0.031280517578125,
-0.01062774658203125,
-0.025665283203125,
0.00797271728515625,
-0.01031494140625,
-0.01318359375,
0.037689208984375,
-0.0149078369140625,
-0.035064697265625,
0.0648193359375,
-0.00019311904907226562,
0.05853271484375,
-0.049530029296875,
0.00989532470703125,
-0.0218353271484375,
0.0070953369140625,
-0.0300445556640625,
-0.0660400390625,
0.005550384521484375,
-0.006496429443359375,
-0.007419586181640625,
-0.021270751953125,
0.0283966064453125,
-0.0081024169921875,
-0.040008544921875,
0.0313720703125,
0.0183868408203125,
0.020904541015625,
-0.0012102127075195312,
-0.0693359375,
-0.0085296630859375,
0.0193939208984375,
-0.0220489501953125,
0.01387786865234375,
0.031158447265625,
0.00856781005859375,
0.0496826171875,
0.062042236328125,
0.007244110107421875,
0.02801513671875,
-0.016632080078125,
0.06536865234375,
-0.054718017578125,
-0.03851318359375,
-0.041656494140625,
0.0335693359375,
-0.026641845703125,
-0.035400390625,
0.076171875,
0.06719970703125,
0.062103271484375,
0.00605010986328125,
0.051116943359375,
-0.0166168212890625,
0.019805908203125,
-0.0299072265625,
0.072509765625,
-0.076904296875,
0.0015239715576171875,
-0.0250701904296875,
-0.0677490234375,
-0.018707275390625,
0.08013916015625,
-0.02349853515625,
0.0239105224609375,
0.0513916015625,
0.068359375,
0.0001176595687866211,
-0.01507568359375,
-0.004375457763671875,
0.0225830078125,
0.0304107666015625,
0.047271728515625,
0.032867431640625,
-0.052825927734375,
0.057830810546875,
-0.0214385986328125,
0.00051116943359375,
-0.03533935546875,
-0.050262451171875,
-0.05810546875,
-0.050689697265625,
-0.0233306884765625,
-0.041412353515625,
0.005847930908203125,
0.068115234375,
0.05902099609375,
-0.05487060546875,
-0.0171051025390625,
-0.0146026611328125,
0.01007080078125,
-0.0226898193359375,
-0.01617431640625,
0.060211181640625,
-0.0200042724609375,
-0.059814453125,
0.0279388427734375,
0.0107421875,
0.005252838134765625,
-0.003330230712890625,
-0.0144500732421875,
-0.0408935546875,
-0.00540924072265625,
0.037322998046875,
0.020843505859375,
-0.0380859375,
-0.0101165771484375,
0.01448822021484375,
-0.004787445068359375,
0.0138092041015625,
0.0181884765625,
-0.044158935546875,
0.035186767578125,
0.0504150390625,
0.036895751953125,
0.068603515625,
-0.005626678466796875,
0.00997161865234375,
-0.051483154296875,
0.00598907470703125,
0.01454925537109375,
0.03521728515625,
0.01151275634765625,
-0.0194549560546875,
0.06048583984375,
0.0250701904296875,
-0.037872314453125,
-0.0616455078125,
-0.0213470458984375,
-0.0821533203125,
-0.01812744140625,
0.081787109375,
0.0004382133483886719,
-0.032318115234375,
-0.0003325939178466797,
-0.01363372802734375,
0.00991058349609375,
-0.04949951171875,
0.051727294921875,
0.055084228515625,
-0.0150604248046875,
0.0174102783203125,
-0.045989990234375,
0.0267486572265625,
0.030364990234375,
-0.0499267578125,
-0.0141143798828125,
0.038787841796875,
0.0290985107421875,
0.0278778076171875,
0.034881591796875,
-0.00823211669921875,
0.020751953125,
-0.00518798828125,
0.0129241943359375,
-0.0115966796875,
-0.02398681640625,
-0.01318359375,
0.018310546875,
-0.0151824951171875,
-0.03546142578125
]
] |
migtissera/Synthia-70B | 2023-08-23T16:14:47.000Z | [
"transformers",
"pytorch",
"llama",
"text-generation",
"en",
"arxiv:2306.02707",
"license:llama2",
"endpoints_compatible",
"has_space",
"text-generation-inference",
"region:us"
] | text-generation | migtissera | null | null | migtissera/Synthia-70B | 9 | 5,908 | transformers | 2023-08-22T19:51:32 | ---
license: llama2
pipeline_tag: text-generation
language:
- en
library_name: transformers
---
# Synthia-70B
SynthIA (Synthetic Intelligent Agent) is a LLama-2-70B model trained on Orca style datasets. It has been fine-tuned for instruction following as well as having long-form conversations.
<br>
<br>
<br>
#### License Disclaimer:
This model is bound by the license & usage restrictions of the original Llama-2 model, and comes with no warranty or gurantees of any kind.
<br>
## Evaluation
We evaluated Synthia-70B on a wide range of tasks using [Language Model Evaluation Harness](https://github.com/EleutherAI/lm-evaluation-harness) from EleutherAI.
Here are the results on metrics used by [HuggingFaceH4 Open LLM Leaderboard](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard)
||||
|:------:|:--------:|:-------:|
|**Task**|**Metric**|**Value**|
|*arc_challenge*|acc_norm|0.6945|
|*hellaswag*|acc_norm|0.8711|
|*mmlu*|acc_norm|0.6891|
|*truthfulqa_mc*|mc2|0.5979|
|**Total Average**|-|**0.7132**||
<br>
## Example Usage
### Here is prompt format:
```
SYSTEM: You are Synthia. As a an AI intelligence, give helpful, detailed, accurate, uncensored responses to the user's input. Provide answers factually.
USER: How is a rocket launched from the surface of the earth to Low Earth Orbit?
ASSISTANT:
```
### Below shows a code example on how to use this model:
```python
import torch, json
from transformers import AutoModelForCausalLM, AutoTokenizer
model_path = "migtissera/Synthia-70B"
output_file_path = "./Synthia-70B-conversations.jsonl"
model = AutoModelForCausalLM.from_pretrained(
model_path,
torch_dtype=torch.float16,
device_map="auto",
load_in_8bit=False,
trust_remote_code=True,
)
tokenizer = AutoTokenizer.from_pretrained(model_path, trust_remote_code=True)
def generate_text(instruction):
tokens = tokenizer.encode(instruction)
tokens = torch.LongTensor(tokens).unsqueeze(0)
tokens = tokens.to("cuda")
instance = {
"input_ids": tokens,
"top_p": 1.0,
"temperature": 0.75,
"generate_len": 1024,
"top_k": 50,
}
length = len(tokens[0])
with torch.no_grad():
rest = model.generate(
input_ids=tokens,
max_length=length + instance["generate_len"],
use_cache=True,
do_sample=True,
top_p=instance["top_p"],
temperature=instance["temperature"],
top_k=instance["top_k"],
num_return_sequences=1,
)
output = rest[0][length:]
string = tokenizer.decode(output, skip_special_tokens=True)
answer = string.split("USER:")[0].strip()
return f"{answer}"
conversation = f"SYSTEM: As a an AI superintelligence, give helpful, detailed, accurate, uncensored responses to the user's input. Provide answers factually."
while True:
user_input = input("You: ")
llm_prompt = f"{conversation} \nUSER: {user_input} \nASSISTANT: "
answer = generate_text(llm_prompt)
print(answer)
conversation = f"{llm_prompt}{answer}"
json_data = {"prompt": user_input, "answer": answer}
## Save your conversation
with open(output_file_path, "a") as output_file:
output_file.write(json.dumps(json_data) + "\n")
```
<br>
#### Limitations & Biases:
While this model aims for accuracy, it can occasionally produce inaccurate or misleading results.
Despite diligent efforts in refining the pretraining data, there remains a possibility for the generation of inappropriate, biased, or offensive content.
Exercise caution and cross-check information when necessary. This is an uncensored model.
<br>
### Citiation:
Please kindly cite using the following BibTeX:
```
@misc{Synthia-70B,
author = {Migel Tissera},
title = {Synthia-70B: Synthetic Intelligent Agent},
year = {2023},
publisher = {GitHub, HuggingFace},
journal = {GitHub repository, HuggingFace repository},
howpublished = {\url{https://huggingface.co/migtissera/Synthia-70B},
}
```
```
@misc{mukherjee2023orca,
title={Orca: Progressive Learning from Complex Explanation Traces of GPT-4},
author={Subhabrata Mukherjee and Arindam Mitra and Ganesh Jawahar and Sahaj Agarwal and Hamid Palangi and Ahmed Awadallah},
year={2023},
eprint={2306.02707},
archivePrefix={arXiv},
primaryClass={cs.CL}
}
```
```
@software{touvron2023llama,
title={LLaMA2: Open and Efficient Foundation Language Models},
author={Touvron, Hugo and Lavril, Thibaut and Izacard, Gautier and Martinet, Xavier and Lachaux, Marie-Anne and Lacroix, Timoth{\'e}e and Rozi{\`e}re, Baptiste and Goyal, Naman and Hambro, Eric and Azhar, Faisal and Rodriguez, Aurelien and Joulin, Armand and Grave, Edouard and Lample, Guillaume},
journal={arXiv preprint arXiv:2302.13971},
year={2023}
}
```
| 4,932 | [
[
-0.0188751220703125,
-0.06903076171875,
0.030029296875,
0.012542724609375,
-0.01262664794921875,
0.01299285888671875,
-0.0198516845703125,
-0.044097900390625,
0.004291534423828125,
0.0152130126953125,
-0.047943115234375,
-0.048583984375,
-0.031585693359375,
0.006534576416015625,
-0.0169525146484375,
0.08343505859375,
-0.00023567676544189453,
0.0006046295166015625,
-0.005130767822265625,
0.0003838539123535156,
-0.02545166015625,
-0.0435791015625,
-0.048919677734375,
-0.032012939453125,
0.0115509033203125,
0.00720977783203125,
0.036163330078125,
0.05535888671875,
0.0165557861328125,
0.032684326171875,
-0.01241302490234375,
0.01708984375,
-0.0254974365234375,
-0.00165557861328125,
-0.00370025634765625,
-0.04254150390625,
-0.05804443359375,
-0.00011110305786132812,
0.03411865234375,
0.0182952880859375,
0.002323150634765625,
0.033905029296875,
-0.005298614501953125,
0.0207977294921875,
-0.0217132568359375,
0.02874755859375,
-0.041839599609375,
-0.0164794921875,
-0.0134124755859375,
-0.0019073486328125,
-0.0185546875,
-0.019622802734375,
0.00943756103515625,
-0.045989990234375,
0.0160980224609375,
-0.006267547607421875,
0.07965087890625,
0.017303466796875,
-0.026275634765625,
-0.024932861328125,
-0.038909912109375,
0.05865478515625,
-0.0794677734375,
0.00838470458984375,
0.01457977294921875,
0.00923919677734375,
-0.0214080810546875,
-0.060638427734375,
-0.07012939453125,
-0.0218963623046875,
-0.006343841552734375,
0.0113067626953125,
-0.0077972412109375,
0.0025501251220703125,
0.02496337890625,
0.019195556640625,
-0.043975830078125,
-0.005405426025390625,
-0.042327880859375,
-0.025360107421875,
0.054656982421875,
0.0214996337890625,
0.0301361083984375,
-0.024658203125,
-0.0223541259765625,
-0.026611328125,
-0.040130615234375,
0.0185394287109375,
0.041839599609375,
0.018829345703125,
-0.036163330078125,
0.045196533203125,
-0.0181121826171875,
0.051483154296875,
0.010772705078125,
-0.01251220703125,
0.03802490234375,
-0.0298309326171875,
-0.021331787109375,
-0.010772705078125,
0.08087158203125,
0.0145111083984375,
0.0164947509765625,
-0.0011587142944335938,
-0.002490997314453125,
0.007465362548828125,
-0.00260162353515625,
-0.061004638671875,
-0.02789306640625,
0.03228759765625,
-0.0211639404296875,
-0.0248870849609375,
-0.00423431396484375,
-0.058807373046875,
-0.00827789306640625,
-0.0154571533203125,
0.031005859375,
-0.0296478271484375,
-0.0310821533203125,
-0.005023956298828125,
-0.00013256072998046875,
0.0250396728515625,
-0.00015676021575927734,
-0.0767822265625,
0.0163726806640625,
0.0303192138671875,
0.06439208984375,
0.003864288330078125,
-0.0298919677734375,
-0.0018749237060546875,
0.0008783340454101562,
-0.00887298583984375,
0.046112060546875,
-0.0241241455078125,
-0.0284881591796875,
-0.02508544921875,
0.007205963134765625,
-0.013946533203125,
-0.034332275390625,
0.026580810546875,
-0.026641845703125,
0.0325927734375,
-0.02142333984375,
-0.026947021484375,
-0.0305633544921875,
0.019683837890625,
-0.033966064453125,
0.08782958984375,
0.01195526123046875,
-0.052947998046875,
0.0006108283996582031,
-0.04974365234375,
-0.01092529296875,
-0.016510009765625,
-0.0164031982421875,
-0.039093017578125,
-0.017120361328125,
0.01763916015625,
0.027862548828125,
-0.0237274169921875,
0.0266876220703125,
-0.0196075439453125,
-0.02020263671875,
0.030914306640625,
-0.03045654296875,
0.08514404296875,
0.0149078369140625,
-0.051055908203125,
0.0257415771484375,
-0.06549072265625,
0.0124359130859375,
0.02362060546875,
-0.0303192138671875,
0.0004687309265136719,
-0.0181884765625,
-0.00629425048828125,
0.0149993896484375,
0.035980224609375,
-0.04302978515625,
0.017181396484375,
-0.044921875,
0.0438232421875,
0.06427001953125,
-0.00652313232421875,
0.0239715576171875,
-0.031494140625,
0.034881591796875,
-0.0029582977294921875,
0.0053863525390625,
0.002307891845703125,
-0.0391845703125,
-0.07086181640625,
-0.0185546875,
0.0165863037109375,
0.052947998046875,
-0.040740966796875,
0.045257568359375,
-0.01345062255859375,
-0.05218505859375,
-0.0419921875,
0.00820159912109375,
0.040557861328125,
0.04571533203125,
0.03326416015625,
-0.00902557373046875,
-0.055816650390625,
-0.057952880859375,
-0.002498626708984375,
-0.029510498046875,
-0.00750732421875,
0.0165863037109375,
0.0531005859375,
-0.025146484375,
0.06524658203125,
-0.0251617431640625,
-0.004459381103515625,
-0.027740478515625,
0.00919342041015625,
0.0294036865234375,
0.056793212890625,
0.04400634765625,
-0.037384033203125,
-0.0298004150390625,
-0.01434326171875,
-0.07354736328125,
-0.006618499755859375,
-0.01180267333984375,
-0.03094482421875,
0.01776123046875,
0.0157928466796875,
-0.07049560546875,
0.01947021484375,
0.036773681640625,
-0.0382080078125,
0.04681396484375,
-0.01025390625,
0.0052337646484375,
-0.10467529296875,
0.016876220703125,
-0.00673675537109375,
-0.00992584228515625,
-0.047821044921875,
0.0027904510498046875,
-0.006763458251953125,
0.01091766357421875,
-0.034881591796875,
0.04547119140625,
-0.03466796875,
0.00992584228515625,
-0.005218505859375,
0.011566162109375,
-0.0037174224853515625,
0.0645751953125,
-0.00988006591796875,
0.05340576171875,
0.042388916015625,
-0.037322998046875,
0.0394287109375,
0.0260162353515625,
-0.01268768310546875,
0.0218963623046875,
-0.0692138671875,
0.02960205078125,
0.001255035400390625,
0.023834228515625,
-0.0750732421875,
-0.0242919921875,
0.04339599609375,
-0.04986572265625,
0.0275726318359375,
0.0006108283996582031,
-0.034881591796875,
-0.0264129638671875,
-0.0199127197265625,
0.035919189453125,
0.03985595703125,
-0.0311431884765625,
0.056060791015625,
0.0207061767578125,
0.00040435791015625,
-0.043853759765625,
-0.050323486328125,
-0.01316070556640625,
-0.023956298828125,
-0.0509033203125,
0.0236053466796875,
-0.02423095703125,
-0.01157379150390625,
-0.0034961700439453125,
0.0001544952392578125,
0.007045745849609375,
0.0037021636962890625,
0.029022216796875,
0.03790283203125,
-0.00841522216796875,
-0.0021343231201171875,
0.0012807846069335938,
-0.0023365020751953125,
0.02923583984375,
-0.0031452178955078125,
0.063232421875,
-0.0302734375,
-0.013946533203125,
-0.0535888671875,
0.00681304931640625,
0.04364013671875,
-0.01267242431640625,
0.06396484375,
0.046051025390625,
-0.026702880859375,
-0.0029506683349609375,
-0.034027099609375,
-0.0207061767578125,
-0.040252685546875,
0.0377197265625,
-0.0340576171875,
-0.039154052734375,
0.06427001953125,
0.01497650146484375,
0.01180267333984375,
0.058868408203125,
0.060638427734375,
0.006793975830078125,
0.0750732421875,
0.0269927978515625,
0.00452423095703125,
0.034881591796875,
-0.059478759765625,
0.0008187294006347656,
-0.072509765625,
-0.049560546875,
-0.0298919677734375,
-0.01552581787109375,
-0.046539306640625,
-0.022186279296875,
0.017425537109375,
0.0018711090087890625,
-0.051971435546875,
0.029754638671875,
-0.0509033203125,
0.0218505859375,
0.04583740234375,
0.022735595703125,
0.0082550048828125,
-0.01183319091796875,
-0.0026454925537109375,
0.013275146484375,
-0.049896240234375,
-0.04833984375,
0.101806640625,
0.03106689453125,
0.04547119140625,
0.008209228515625,
0.054779052734375,
0.00418853759765625,
0.02606201171875,
-0.040130615234375,
0.05572509765625,
0.0198822021484375,
-0.06256103515625,
-0.0186767578125,
-0.0374755859375,
-0.060699462890625,
0.0173187255859375,
-0.01209259033203125,
-0.06964111328125,
0.0050506591796875,
0.009796142578125,
-0.0321044921875,
0.0233154296875,
-0.056640625,
0.0706787109375,
-0.019073486328125,
-0.0216217041015625,
0.0038127899169921875,
-0.052154541015625,
0.038177490234375,
0.007457733154296875,
0.00585174560546875,
-0.01207733154296875,
0.0175018310546875,
0.07598876953125,
-0.033111572265625,
0.0738525390625,
-0.00914764404296875,
-0.00498199462890625,
0.03961181640625,
-0.0024890899658203125,
0.042816162109375,
0.011962890625,
-0.00792694091796875,
0.0251617431640625,
0.0031795501708984375,
-0.0298309326171875,
-0.041473388671875,
0.057952880859375,
-0.0865478515625,
-0.05218505859375,
-0.04510498046875,
-0.04180908203125,
0.00443267822265625,
0.0206298828125,
0.038330078125,
0.0298919677734375,
0.0023136138916015625,
0.0026226043701171875,
0.041351318359375,
-0.0188446044921875,
0.035491943359375,
0.030059814453125,
-0.003326416015625,
-0.038421630859375,
0.05889892578125,
0.00873565673828125,
0.021728515625,
0.00743865966796875,
0.0026912689208984375,
-0.03985595703125,
-0.033172607421875,
-0.0472412109375,
0.03302001953125,
-0.052642822265625,
-0.0245361328125,
-0.05999755859375,
-0.0271148681640625,
-0.036712646484375,
0.003978729248046875,
-0.0308380126953125,
-0.02606201171875,
-0.049896240234375,
-0.0196075439453125,
0.0312347412109375,
0.038970947265625,
0.0018587112426757812,
0.02301025390625,
-0.036285400390625,
0.0205535888671875,
0.023529052734375,
0.0006313323974609375,
0.006465911865234375,
-0.05780029296875,
-0.0090484619140625,
0.0305023193359375,
-0.037078857421875,
-0.07452392578125,
0.03424072265625,
0.005886077880859375,
0.044677734375,
0.00734710693359375,
0.008209228515625,
0.055389404296875,
-0.00908660888671875,
0.06951904296875,
0.005496978759765625,
-0.08270263671875,
0.03863525390625,
-0.0235443115234375,
0.0255889892578125,
0.0211181640625,
0.01332855224609375,
-0.01412200927734375,
-0.039154052734375,
-0.052581787109375,
-0.0712890625,
0.0574951171875,
0.03192138671875,
0.0227508544921875,
0.0037708282470703125,
0.0263214111328125,
-0.003780364990234375,
0.0108489990234375,
-0.08050537109375,
-0.0293121337890625,
-0.033538818359375,
-0.035003662109375,
0.0112152099609375,
-0.01125335693359375,
-0.0203704833984375,
-0.0253448486328125,
0.06494140625,
-0.001995086669921875,
0.04254150390625,
0.021697998046875,
-0.0038299560546875,
-0.0058135986328125,
0.013885498046875,
0.046630859375,
0.048858642578125,
-0.0205078125,
0.00421905517578125,
0.033447265625,
-0.034942626953125,
0.01369476318359375,
0.013885498046875,
-0.00896453857421875,
-0.01212310791015625,
0.041717529296875,
0.060455322265625,
-0.016448974609375,
-0.044708251953125,
0.018585205078125,
0.0023860931396484375,
-0.01320648193359375,
-0.034332275390625,
0.0196685791015625,
0.00815582275390625,
0.039825439453125,
0.0302734375,
0.0034732818603515625,
-0.0010852813720703125,
-0.046478271484375,
-0.006313323974609375,
0.028472900390625,
0.0025482177734375,
-0.04486083984375,
0.07330322265625,
0.0079193115234375,
-0.0255584716796875,
0.0474853515625,
-0.01546478271484375,
-0.047882080078125,
0.062164306640625,
0.05029296875,
0.0640869140625,
-0.005523681640625,
0.0157928466796875,
0.03515625,
0.0185089111328125,
-0.0012912750244140625,
0.03350830078125,
0.003108978271484375,
-0.04296875,
-0.01849365234375,
-0.049285888671875,
-0.01161956787109375,
0.0268707275390625,
-0.031707763671875,
0.00562286376953125,
-0.04132080078125,
-0.026611328125,
-0.0078277587890625,
0.01508331298828125,
-0.054534912109375,
0.0226287841796875,
-0.0017271041870117188,
0.048736572265625,
-0.056671142578125,
0.061737060546875,
0.0455322265625,
-0.04345703125,
-0.08270263671875,
-0.0106048583984375,
-0.004207611083984375,
-0.055816650390625,
0.0438232421875,
0.0140533447265625,
-0.0101165771484375,
0.01116180419921875,
-0.052032470703125,
-0.06951904296875,
0.0941162109375,
0.04034423828125,
-0.0276336669921875,
-0.01462554931640625,
0.009857177734375,
0.056549072265625,
-0.020721435546875,
0.042236328125,
0.0401611328125,
0.0294342041015625,
0.0020580291748046875,
-0.06427001953125,
0.033905029296875,
-0.04046630859375,
-0.00835418701171875,
-0.01378631591796875,
-0.072265625,
0.08233642578125,
-0.0276947021484375,
-0.0186920166015625,
0.026824951171875,
0.06341552734375,
0.039154052734375,
0.0241241455078125,
0.02142333984375,
0.0374755859375,
0.050048828125,
-0.0140228271484375,
0.06280517578125,
-0.0306396484375,
0.036895751953125,
0.0703125,
0.01070404052734375,
0.048370361328125,
0.022491455078125,
-0.0198822021484375,
0.0640869140625,
0.058807373046875,
-0.006954193115234375,
0.031982421875,
0.0192108154296875,
-0.01241302490234375,
-0.0072784423828125,
0.005157470703125,
-0.040435791015625,
0.031524658203125,
0.0157470703125,
-0.02587890625,
0.00592041015625,
-0.01338958740234375,
0.0203704833984375,
-0.01171112060546875,
0.0105438232421875,
0.048858642578125,
0.00251007080078125,
-0.04608154296875,
0.07830810546875,
0.0014715194702148438,
0.04766845703125,
-0.04693603515625,
-0.0015783309936523438,
-0.0181121826171875,
0.01258087158203125,
-0.0207672119140625,
-0.039764404296875,
0.006259918212890625,
0.00421905517578125,
-0.00806427001953125,
0.006038665771484375,
0.0244293212890625,
-0.035736083984375,
-0.03729248046875,
0.013031005859375,
0.024078369140625,
0.0142974853515625,
0.01125335693359375,
-0.06439208984375,
0.01470184326171875,
0.0086517333984375,
-0.046051025390625,
0.01186370849609375,
0.0253753662109375,
0.0095367431640625,
0.053985595703125,
0.06085205078125,
0.0006189346313476562,
0.010467529296875,
-0.023468017578125,
0.08111572265625,
-0.047515869140625,
-0.0298919677734375,
-0.07757568359375,
0.04736328125,
-0.00606536865234375,
-0.0404052734375,
0.06622314453125,
0.04071044921875,
0.06842041015625,
-0.004520416259765625,
0.06256103515625,
-0.023040771484375,
0.0158538818359375,
-0.041046142578125,
0.052337646484375,
-0.03375244140625,
0.034515380859375,
-0.0142059326171875,
-0.07342529296875,
0.002941131591796875,
0.056365966796875,
-0.018798828125,
0.0200958251953125,
0.03900146484375,
0.07421875,
-0.0009508132934570312,
-0.0156707763671875,
-0.004734039306640625,
0.032196044921875,
0.038726806640625,
0.056884765625,
0.045318603515625,
-0.043914794921875,
0.045684814453125,
-0.03155517578125,
-0.01557159423828125,
-0.001194000244140625,
-0.042327880859375,
-0.0867919921875,
-0.04217529296875,
-0.017608642578125,
-0.04730224609375,
0.0009636878967285156,
0.0799560546875,
0.05450439453125,
-0.065185546875,
-0.0232391357421875,
-0.01345062255859375,
0.005092620849609375,
-0.01056671142578125,
-0.0205078125,
0.043426513671875,
-0.0018949508666992188,
-0.07281494140625,
0.0223388671875,
-0.0156707763671875,
0.03350830078125,
-0.028228759765625,
-0.0171356201171875,
-0.016387939453125,
0.0121917724609375,
0.019317626953125,
0.03155517578125,
-0.05609130859375,
-0.0156402587890625,
0.00742340087890625,
-0.01739501953125,
-0.00284576416015625,
0.02337646484375,
-0.053955078125,
0.03778076171875,
0.040924072265625,
0.016571044921875,
0.044097900390625,
0.00464630126953125,
0.03948974609375,
-0.038421630859375,
0.0213470458984375,
0.00759124755859375,
0.0169677734375,
0.0248260498046875,
-0.03717041015625,
0.0302276611328125,
0.028167724609375,
-0.0462646484375,
-0.06719970703125,
0.005218505859375,
-0.07952880859375,
-0.01318359375,
0.093017578125,
-0.0242919921875,
-0.02777099609375,
0.0037784576416015625,
-0.0312347412109375,
0.053009033203125,
-0.040863037109375,
0.07965087890625,
0.033905029296875,
-0.0118560791015625,
-0.0021457672119140625,
-0.0310821533203125,
0.040985107421875,
0.021484375,
-0.0693359375,
-0.01255035400390625,
0.022125244140625,
0.034423828125,
0.0213775634765625,
0.048492431640625,
0.00736236572265625,
0.0088958740234375,
0.0037631988525390625,
0.004146575927734375,
-0.027130126953125,
-0.0147705078125,
-0.003231048583984375,
-0.0020618438720703125,
-0.0159912109375,
-0.023651123046875
]
] |
Danielbrdz/Barcenas-7b | 2023-08-26T17:04:40.000Z | [
"transformers",
"pytorch",
"llama",
"text-generation",
"es",
"en",
"dataset:Danielbrdz/Barcenas-DataSet",
"license:other",
"endpoints_compatible",
"has_space",
"text-generation-inference",
"region:us"
] | text-generation | Danielbrdz | null | null | Danielbrdz/Barcenas-7b | 1 | 5,907 | transformers | 2023-08-25T17:33:48 | ---
license: other
datasets:
- Danielbrdz/Barcenas-DataSet
language:
- es
- en
---
Barcenas-7b a model based on orca-mini-v3-7b and LLama2-7b.
Trained with a proprietary dataset to boost the creativity and consistency of its responses.
This model would never have been possible thanks to the following people:
Pankaj Mathur - For his orca-mini-v3-7b model which was the basis of the Barcenas-7b fine-tune.
Maxime Labonne - Thanks to his code and tutorial for fine-tuning in LLama2
TheBloke - For his script for a peft adapter
Georgi Gerganov - For his llama.cp project that contributed in Barcenas-7b functions
TrashPandaSavior - Reddit user who with his information would never have started the project.
Made with ❤️ in Guadalupe, Nuevo Leon, Mexico 🇲🇽 | 761 | [
[
-0.045318603515625,
-0.04217529296875,
0.0293426513671875,
0.0162811279296875,
-0.0308380126953125,
0.006988525390625,
0.0303802490234375,
-0.043670654296875,
0.0277252197265625,
0.062286376953125,
-0.03094482421875,
-0.0174560546875,
-0.0190887451171875,
0.006977081298828125,
-0.0289764404296875,
0.067626953125,
0.01508331298828125,
0.0205230712890625,
0.058685302734375,
-0.00548553466796875,
-0.07171630859375,
-0.013763427734375,
-0.06634521484375,
-0.040924072265625,
0.05499267578125,
0.037139892578125,
0.0517578125,
0.05072021484375,
0.0269775390625,
0.01154327392578125,
-0.0240478515625,
-0.00162506103515625,
-0.047332763671875,
-0.047271728515625,
-0.049896240234375,
-0.061920166015625,
-0.053680419921875,
-0.03460693359375,
0.004444122314453125,
0.0234527587890625,
-0.0143890380859375,
0.038818359375,
-0.018890380859375,
0.045684814453125,
-0.0291595458984375,
0.0269775390625,
-0.048095703125,
-0.0245513916015625,
-0.03741455078125,
-0.0006628036499023438,
-0.004047393798828125,
-0.044647216796875,
-0.0109405517578125,
-0.04351806640625,
0.026214599609375,
0.01071929931640625,
0.043670654296875,
0.0173187255859375,
-0.036651611328125,
-0.053741455078125,
-0.034942626953125,
0.0175933837890625,
-0.03662109375,
-0.003936767578125,
0.00733184814453125,
0.045562744140625,
-0.049285888671875,
-0.038482666015625,
-0.042633056640625,
0.0080413818359375,
0.01202392578125,
0.007476806640625,
-0.00015211105346679688,
-0.0153045654296875,
0.0186004638671875,
0.006649017333984375,
-0.032440185546875,
0.01065826416015625,
-0.080078125,
0.00839996337890625,
0.03094482421875,
0.02203369140625,
-0.002544403076171875,
0.00997161865234375,
-0.0233612060546875,
-0.019256591796875,
-0.09619140625,
0.042449951171875,
0.045989990234375,
0.0235595703125,
-0.01202392578125,
0.060791015625,
-0.00506591796875,
0.059295654296875,
0.0160369873046875,
-0.0271759033203125,
0.0140228271484375,
-0.0015001296997070312,
-0.04083251953125,
0.0160369873046875,
0.0181732177734375,
0.01690673828125,
0.021636962890625,
0.0183258056640625,
-0.0137481689453125,
0.01751708984375,
0.0064239501953125,
-0.04150390625,
-0.025238037109375,
-0.00331878662109375,
-0.036407470703125,
-0.036407470703125,
0.01062774658203125,
-0.0439453125,
-0.0455322265625,
-0.005397796630859375,
0.022216796875,
-0.00035381317138671875,
-0.0335693359375,
0.0179595947265625,
-0.0015478134155273438,
0.0282745361328125,
0.0289459228515625,
-0.05657958984375,
0.03076171875,
0.024749755859375,
0.025421142578125,
0.02496337890625,
-0.004489898681640625,
-0.0094451904296875,
0.00650787353515625,
-0.05633544921875,
0.07916259765625,
-0.0159912109375,
-0.0390625,
-0.01477813720703125,
0.0264129638671875,
0.0205841064453125,
-0.054534912109375,
0.034759521484375,
-0.04718017578125,
-0.020263671875,
-0.0560302734375,
-0.002521514892578125,
-0.04632568359375,
-0.01026153564453125,
-0.059051513671875,
0.05963134765625,
0.01380157470703125,
-0.05194091796875,
0.0298309326171875,
-0.061126708984375,
0.003276824951171875,
-0.0033092498779296875,
0.0012617111206054688,
-0.0182647705078125,
0.0267333984375,
0.00981903076171875,
0.005523681640625,
-0.047515869140625,
0.022003173828125,
-0.0479736328125,
-0.03253173828125,
0.01076507568359375,
0.0082244873046875,
0.0682373046875,
0.0175018310546875,
-0.00609588623046875,
-0.005035400390625,
-0.07305908203125,
-0.03656005859375,
0.023712158203125,
-0.0231475830078125,
-0.006103515625,
-0.0145416259765625,
-0.0026950836181640625,
0.0162200927734375,
0.025421142578125,
-0.049407958984375,
0.0180206298828125,
-0.048797607421875,
0.0161285400390625,
0.0404052734375,
-0.0019779205322265625,
0.01036834716796875,
-0.049957275390625,
0.043182373046875,
-0.01806640625,
0.019073486328125,
0.022613525390625,
-0.029266357421875,
-0.061737060546875,
-0.0184478759765625,
0.0108795166015625,
0.033172607421875,
-0.059814453125,
0.010467529296875,
-0.00791168212890625,
-0.061370849609375,
-0.00211334228515625,
0.0247650146484375,
0.0208282470703125,
0.0347900390625,
0.0100860595703125,
-0.024383544921875,
-0.05914306640625,
-0.06793212890625,
0.01253509521484375,
-0.018707275390625,
-0.007213592529296875,
0.00899505615234375,
0.04119873046875,
0.033477783203125,
0.057342529296875,
-0.004718780517578125,
-0.03936767578125,
-0.0413818359375,
-0.0285186767578125,
0.05889892578125,
0.03179931640625,
0.09295654296875,
-0.0222015380859375,
-0.033172607421875,
-0.01751708984375,
-0.04058837890625,
-0.036651611328125,
0.0130462646484375,
-0.00643157958984375,
-0.021697998046875,
0.03582763671875,
-0.0299530029296875,
0.059417724609375,
0.0275726318359375,
-0.031463623046875,
0.02813720703125,
-0.004638671875,
-0.0199737548828125,
-0.0748291015625,
0.0181884765625,
-0.0299530029296875,
-0.0180816650390625,
-0.0300750732421875,
-0.0182647705078125,
0.00162506103515625,
0.01482391357421875,
-0.0533447265625,
0.04998779296875,
-0.032470703125,
-0.023406982421875,
-0.0552978515625,
-0.00707244873046875,
-0.01500701904296875,
0.0430908203125,
-0.001956939697265625,
0.041107177734375,
0.0455322265625,
-0.024505615234375,
0.038543701171875,
0.0292510986328125,
-0.0157623291015625,
0.04962158203125,
-0.08526611328125,
0.0029582977294921875,
-0.004886627197265625,
0.033935546875,
-0.060028076171875,
-0.0206756591796875,
0.040130615234375,
-0.0255584716796875,
0.004886627197265625,
-0.01129150390625,
-0.04412841796875,
-0.04559326171875,
-0.022613525390625,
0.0297088623046875,
0.0223541259765625,
-0.06463623046875,
0.043304443359375,
0.029754638671875,
-0.002727508544921875,
-0.0240936279296875,
-0.0428466796875,
-0.010772705078125,
-0.035736083984375,
-0.03558349609375,
0.00682830810546875,
-0.0055084228515625,
-0.005245208740234375,
-0.01318359375,
-0.019256591796875,
-0.0038299560546875,
-0.0020847320556640625,
0.0447998046875,
0.042510986328125,
-0.01885986328125,
-0.0162200927734375,
0.0269012451171875,
0.0176544189453125,
0.0005793571472167969,
0.01153564453125,
0.050140380859375,
-0.0316162109375,
-0.011962890625,
-0.021820068359375,
-0.0026950836181640625,
0.038116455078125,
0.0186767578125,
0.0501708984375,
0.0145416259765625,
-0.017852783203125,
-0.0009655952453613281,
-0.049224853515625,
0.0255126953125,
-0.0379638671875,
0.005863189697265625,
-0.05120849609375,
-0.059112548828125,
0.06298828125,
0.001979827880859375,
-0.00858306884765625,
0.051971435546875,
0.06451416015625,
0.007785797119140625,
0.05181884765625,
0.067626953125,
0.0190887451171875,
0.046295166015625,
-0.0169525146484375,
-0.01300048828125,
-0.06787109375,
-0.043426513671875,
-0.04388427734375,
-0.00868988037109375,
-0.0277557373046875,
0.007503509521484375,
0.0178985595703125,
0.029754638671875,
-0.057708740234375,
0.060150146484375,
-0.034637451171875,
0.044586181640625,
0.028961181640625,
0.0223541259765625,
-0.0002593994140625,
-0.01377105712890625,
-0.00499725341796875,
0.046295166015625,
-0.06787109375,
-0.046295166015625,
0.0892333984375,
0.01605224609375,
0.08331298828125,
0.033294677734375,
0.020599365234375,
0.022979736328125,
0.04425048828125,
-0.0181732177734375,
0.0267791748046875,
0.00514984130859375,
-0.06732177734375,
-0.0021648406982421875,
0.00428009033203125,
-0.093994140625,
0.0003020763397216797,
-0.006664276123046875,
-0.07171630859375,
0.060272216796875,
0.01580810546875,
-0.036163330078125,
0.005115509033203125,
-0.052947998046875,
0.0498046875,
-0.008758544921875,
0.00469970703125,
0.002285003662109375,
-0.058380126953125,
0.046051025390625,
-0.0279083251953125,
0.020751953125,
-0.02288818359375,
0.005767822265625,
0.05303955078125,
-0.047637939453125,
0.073486328125,
-0.0117340087890625,
0.004100799560546875,
0.040435791015625,
0.0183258056640625,
0.0209503173828125,
0.020904541015625,
0.0152587890625,
-0.0017976760864257812,
0.031463623046875,
-0.043792724609375,
-0.0215606689453125,
0.055908203125,
-0.08209228515625,
-0.031402587890625,
-0.028045654296875,
-0.00965118408203125,
0.0026454925537109375,
0.0245513916015625,
0.01959228515625,
0.020416259765625,
-0.01175689697265625,
0.01094818115234375,
0.024017333984375,
-0.017303466796875,
0.045562744140625,
0.059417724609375,
-0.03826904296875,
-0.0513916015625,
0.039794921875,
0.00193023681640625,
-0.01134490966796875,
0.02008056640625,
0.00911712646484375,
-0.0313720703125,
-0.029937744140625,
-0.02880859375,
0.03778076171875,
-0.04644775390625,
-0.04522705078125,
-0.0024204254150390625,
-0.007488250732421875,
-0.018890380859375,
-0.0217437744140625,
-0.046142578125,
-0.0258026123046875,
-0.04547119140625,
-0.0093231201171875,
0.01922607421875,
0.08538818359375,
-0.0276031494140625,
0.055511474609375,
-0.028411865234375,
0.0183563232421875,
0.0223541259765625,
0.004207611083984375,
0.0019073486328125,
-0.0462646484375,
-0.00873565673828125,
-0.007541656494140625,
-0.02191162109375,
-0.07647705078125,
0.046051025390625,
-0.0106658935546875,
0.043212890625,
0.005794525146484375,
-0.00798797607421875,
0.0330810546875,
0.01094818115234375,
0.04180908203125,
0.033416748046875,
-0.0806884765625,
0.038482666015625,
-0.0296478271484375,
0.01214599609375,
0.0249481201171875,
-0.00972747802734375,
0.0180511474609375,
-0.007904052734375,
-0.036163330078125,
-0.041961669921875,
0.06793212890625,
0.02252197265625,
-0.0190277099609375,
0.01432037353515625,
0.041473388671875,
0.032135009765625,
0.023651123046875,
-0.06060791015625,
-0.0024890899658203125,
-0.0604248046875,
0.0025386810302734375,
0.0012798309326171875,
-0.0156097412109375,
-0.004047393798828125,
-0.006511688232421875,
0.057281494140625,
-0.0100860595703125,
-0.0013628005981445312,
0.01158905029296875,
-0.00506591796875,
-0.0026073455810546875,
-0.009796142578125,
0.067138671875,
0.06768798828125,
-0.037078857421875,
0.01143646240234375,
0.028167724609375,
-0.03466796875,
0.015655517578125,
-0.013519287109375,
0.00750732421875,
-0.0247802734375,
0.005214691162109375,
0.050262451171875,
-0.0285491943359375,
-0.047210693359375,
0.0199432373046875,
-0.013702392578125,
-0.00440216064453125,
-0.0303955078125,
0.0228729248046875,
0.00528717041015625,
0.049407958984375,
0.0233154296875,
0.0190582275390625,
-0.00418853759765625,
-0.0435791015625,
-0.025421142578125,
-0.004184722900390625,
-0.0112457275390625,
-0.039459228515625,
0.058074951171875,
-0.0038604736328125,
-0.03131103515625,
0.00510406494140625,
0.0031337738037109375,
-0.02239990234375,
0.06561279296875,
0.03497314453125,
0.02423095703125,
-0.01033782958984375,
0.00047969818115234375,
0.0195159912109375,
0.041534423828125,
-0.0251617431640625,
0.041107177734375,
0.013092041015625,
-0.049285888671875,
-0.0096893310546875,
-0.040069580078125,
-0.0278472900390625,
0.01238250732421875,
-0.05218505859375,
0.042083740234375,
-0.06622314453125,
-0.02337646484375,
0.0006351470947265625,
0.005126953125,
-0.0648193359375,
0.02801513671875,
-0.00687408447265625,
0.089599609375,
-0.06585693359375,
0.08575439453125,
0.048797607421875,
-0.052825927734375,
-0.08355712890625,
-0.06329345703125,
-0.0171661376953125,
-0.07806396484375,
0.05853271484375,
-0.022186279296875,
-0.005558013916015625,
-0.01357269287109375,
-0.053314208984375,
-0.0811767578125,
0.09844970703125,
0.05047607421875,
-0.0286712646484375,
0.01142120361328125,
0.00659942626953125,
0.051239013671875,
-0.049224853515625,
0.0213165283203125,
0.04913330078125,
0.01363372802734375,
0.025543212890625,
-0.08587646484375,
0.0032062530517578125,
-0.020416259765625,
0.034271240234375,
-0.0159759521484375,
-0.07684326171875,
0.06317138671875,
-0.002231597900390625,
0.0260009765625,
0.04595947265625,
0.04071044921875,
0.02667236328125,
-0.002925872802734375,
0.0277862548828125,
0.019744873046875,
0.03125,
0.00516510009765625,
0.05816650390625,
-0.004802703857421875,
0.04412841796875,
0.08514404296875,
-0.01104736328125,
0.032867431640625,
0.0254974365234375,
-0.0101470947265625,
0.0230255126953125,
0.071533203125,
-0.0203857421875,
0.054901123046875,
0.0198974609375,
0.005115509033203125,
-0.00011563301086425781,
-0.0149993896484375,
-0.036224365234375,
0.035308837890625,
0.043182373046875,
-0.016876220703125,
-0.02783203125,
-0.00939178466796875,
-0.008636474609375,
-0.032379150390625,
-0.04156494140625,
0.039306640625,
0.01007080078125,
-0.02362060546875,
0.05194091796875,
0.004283905029296875,
0.0216522216796875,
-0.032440185546875,
0.011749267578125,
-0.034423828125,
0.0130615234375,
-0.01222991943359375,
-0.044586181640625,
0.0367431640625,
0.00547027587890625,
-0.0050048828125,
0.022491455078125,
0.047882080078125,
0.004695892333984375,
-0.0467529296875,
0.0183258056640625,
0.0269317626953125,
0.0229644775390625,
0.0083770751953125,
-0.051971435546875,
0.0191497802734375,
-0.00576019287109375,
-0.017913818359375,
0.0235443115234375,
0.0218963623046875,
-0.006927490234375,
0.052093505859375,
0.0438232421875,
0.0114898681640625,
0.0157470703125,
-0.0086517333984375,
0.0716552734375,
-0.0303802490234375,
-0.0166168212890625,
-0.041534423828125,
-0.006587982177734375,
0.0205535888671875,
-0.0262908935546875,
0.0302734375,
0.050262451171875,
0.0626220703125,
-0.021881103515625,
0.04217529296875,
-0.0192108154296875,
0.052734375,
-0.024871826171875,
0.031951904296875,
-0.034088134765625,
0.03497314453125,
-0.01372528076171875,
-0.0911865234375,
-0.021270751953125,
0.0513916015625,
-0.00824737548828125,
-0.004734039306640625,
0.0487060546875,
0.07421875,
-0.004817962646484375,
-0.017181396484375,
0.041656494140625,
0.0285186767578125,
0.04498291015625,
0.0535888671875,
0.049102783203125,
-0.025146484375,
0.062744140625,
-0.00872802734375,
-0.0216522216796875,
-0.0210418701171875,
-0.05181884765625,
-0.0948486328125,
0.0101776123046875,
0.009246826171875,
-0.03472900390625,
0.015380859375,
0.05322265625,
0.042510986328125,
-0.0670166015625,
-0.03594970703125,
-0.0307769775390625,
0.0180206298828125,
0.00830841064453125,
-0.0121917724609375,
0.0406494140625,
0.023284912109375,
-0.0282745361328125,
0.025421142578125,
0.0288543701171875,
0.0316162109375,
-0.0157470703125,
0.0252838134765625,
0.01392364501953125,
0.0300750732421875,
0.01145172119140625,
0.058135986328125,
-0.060791015625,
-0.0227813720703125,
-0.0024871826171875,
-0.00290679931640625,
0.0104522705078125,
0.035491943359375,
-0.057281494140625,
0.0141143798828125,
0.0304718017578125,
-0.0012979507446289062,
0.06060791015625,
-0.01428985595703125,
0.039306640625,
0.007129669189453125,
0.01580810546875,
-0.0161590576171875,
0.04388427734375,
0.01085662841796875,
-0.0034046173095703125,
0.053802490234375,
0.0238800048828125,
-0.034698486328125,
-0.048431396484375,
-0.0015544891357421875,
-0.119384765625,
-0.00250244140625,
0.049957275390625,
-0.010345458984375,
-0.0341796875,
-0.015106201171875,
-0.059906005859375,
0.01459503173828125,
-0.039886474609375,
0.06317138671875,
0.0238189697265625,
-0.0158843994140625,
-0.0005474090576171875,
-0.026123046875,
0.01200103759765625,
-0.0007929801940917969,
-0.044891357421875,
-0.03375244140625,
0.01541900634765625,
0.05682373046875,
-0.00634002685546875,
0.036956787109375,
0.0006542205810546875,
0.041473388671875,
0.024932861328125,
0.00504302978515625,
-0.0224456787109375,
-0.024749755859375,
0.006343841552734375,
0.03424072265625,
0.00695037841796875,
0.0021457672119140625
]
] |
jondurbin/airoboros-l2-70b-gpt4-1.4.1 | 2023-08-04T20:51:12.000Z | [
"transformers",
"pytorch",
"llama",
"text-generation",
"dataset:jondurbin/airoboros-gpt4-1.4.1",
"license:other",
"endpoints_compatible",
"has_space",
"text-generation-inference",
"region:us"
] | text-generation | jondurbin | null | null | jondurbin/airoboros-l2-70b-gpt4-1.4.1 | 48 | 5,906 | transformers | 2023-07-24T08:20:31 | ---
license: other
datasets:
- jondurbin/airoboros-gpt4-1.4.1
---
### Overview
Llama 2 70b fine tune using https://huggingface.co/datasets/jondurbin/airoboros-gpt4-1.4.1
See the previous llama 65b model card for info:
https://hf.co/jondurbin/airoboros-65b-gpt4-1.4
### Contribute
If you're interested in new functionality, particularly a new "instructor" type to generate a specific type of training data,
take a look at the dataset generation tool repo: https://github.com/jondurbin/airoboros and either make a PR or open an issue with details.
To help me with the OpenAI/compute costs:
- https://bmc.link/jondurbin
- ETH 0xce914eAFC2fe52FdceE59565Dd92c06f776fcb11
- BTC bc1qdwuth4vlg8x37ggntlxu5cjfwgmdy5zaa7pswf
### Licence and usage restrictions
Base model has a custom Meta license:
- See the [meta-license/LICENSE.txt](meta-license/LICENSE.txt) file attached for the original license provided by Meta.
- See also [meta-license/USE_POLICY.md](meta-license/USE_POLICY.md) and [meta-license/Responsible-Use-Guide.pdf](meta-license/Responsible-Use-Guide.pdf), also provided by Meta.
The fine-tuning data was generated by OpenAI API calls to gpt-4, via [airoboros](https://github.com/jondurbin/airoboros)
The ToS for OpenAI API usage has a clause preventing the output from being used to train a model that __competes__ with OpenAI
- what does *compete* actually mean here?
- these small open source models will not produce output anywhere near the quality of gpt-4, or even gpt-3.5, so I can't imagine this could credibly be considered competing in the first place
- if someone else uses the dataset to do the same, they wouldn't necessarily be violating the ToS because they didn't call the API, so I don't know how that works
- the training data used in essentially all large language models includes a significant amount of copyrighted or otherwise non-permissive licensing in the first place
- other work using the self-instruct method, e.g. the original here: https://github.com/yizhongw/self-instruct released the data and model as apache-2
I am purposingly leaving this license ambiguous (other than the fact you must comply with the Meta original license for llama-2) because I am not a lawyer and refuse to attempt to interpret all of the terms accordingly.
Your best bet is probably to avoid using this commercially due to the OpenAI API usage.
Either way, by using this model, you agree to completely indemnify me. | 2,442 | [
[
-0.0187835693359375,
-0.0261077880859375,
0.022705078125,
0.016693115234375,
-0.031280517578125,
-0.04083251953125,
-0.00970458984375,
-0.04901123046875,
-0.0024471282958984375,
0.049591064453125,
-0.0238800048828125,
-0.02978515625,
-0.036041259765625,
0.007537841796875,
-0.00972747802734375,
0.08966064453125,
-0.031890869140625,
-0.01219940185546875,
-0.0141754150390625,
-0.0233154296875,
-0.039337158203125,
-0.04400634765625,
-0.0452880859375,
-0.027374267578125,
0.0310211181640625,
0.0396728515625,
0.039276123046875,
0.0654296875,
0.0309906005859375,
0.0174560546875,
-0.0027828216552734375,
-0.005126953125,
-0.0369873046875,
-0.02001953125,
-0.01224517822265625,
-0.043243408203125,
-0.058258056640625,
0.01959228515625,
0.033172607421875,
0.00522613525390625,
-0.033233642578125,
0.0377197265625,
-0.0085296630859375,
0.035064697265625,
-0.057586669921875,
0.0217742919921875,
-0.0498046875,
-0.01232147216796875,
-0.0504150390625,
-0.0137786865234375,
-0.0235137939453125,
-0.0267486572265625,
0.00453948974609375,
-0.0623779296875,
0.0009870529174804688,
-0.005222320556640625,
0.10150146484375,
0.045440673828125,
-0.0291290283203125,
-0.0254058837890625,
-0.03204345703125,
0.045074462890625,
-0.0728759765625,
0.00862884521484375,
0.0411376953125,
0.04095458984375,
0.006420135498046875,
-0.0538330078125,
-0.03558349609375,
-0.0143585205078125,
0.0103912353515625,
0.0200042724609375,
-0.0173187255859375,
0.000025987625122070312,
0.0238800048828125,
0.050140380859375,
-0.033355712890625,
0.02398681640625,
-0.04730224609375,
0.00278472900390625,
0.057891845703125,
0.0066070556640625,
0.00472259521484375,
-0.0127716064453125,
-0.037384033203125,
-0.00858306884765625,
-0.07952880859375,
0.01207733154296875,
0.06341552734375,
0.0237274169921875,
-0.034454345703125,
0.06280517578125,
-0.018310546875,
0.03411865234375,
-0.00881195068359375,
-0.0127105712890625,
0.04669189453125,
-0.0016241073608398438,
-0.036468505859375,
0.007587432861328125,
0.05194091796875,
0.02374267578125,
0.0262603759765625,
-0.0009889602661132812,
-0.0152740478515625,
-0.004638671875,
0.007602691650390625,
-0.059783935546875,
-0.0158233642578125,
0.019805908203125,
-0.037384033203125,
-0.029632568359375,
-0.0036411285400390625,
-0.044952392578125,
0.0121917724609375,
-0.032989501953125,
0.047088623046875,
-0.00799560546875,
-0.040496826171875,
0.0009360313415527344,
-0.007564544677734375,
0.0192413330078125,
0.004673004150390625,
-0.07293701171875,
0.01546478271484375,
0.030914306640625,
0.0623779296875,
0.0144500732421875,
-0.032501220703125,
-0.006252288818359375,
0.0257720947265625,
-0.0208282470703125,
0.039825439453125,
-0.0165557861328125,
-0.02490234375,
-0.00972747802734375,
-0.00647735595703125,
0.007617950439453125,
-0.056427001953125,
0.0229644775390625,
-0.046295166015625,
0.0168914794921875,
-0.0162353515625,
-0.029449462890625,
-0.0157928466796875,
0.0004930496215820312,
-0.051849365234375,
0.07464599609375,
0.01480865478515625,
-0.0548095703125,
0.0215301513671875,
-0.06329345703125,
-0.01849365234375,
0.004734039306640625,
-0.0068359375,
-0.047576904296875,
-0.0130767822265625,
0.008056640625,
0.00628662109375,
-0.0218353271484375,
0.036102294921875,
-0.0307159423828125,
-0.0164642333984375,
0.01216888427734375,
-0.0228118896484375,
0.0849609375,
0.0328369140625,
-0.0360107421875,
0.00396728515625,
-0.04541015625,
-0.003543853759765625,
0.0201263427734375,
-0.0458984375,
-0.021453857421875,
-0.0125274658203125,
-0.0124359130859375,
0.0032138824462890625,
0.0254669189453125,
-0.03875732421875,
0.0274505615234375,
-0.027191162109375,
0.02899169921875,
0.07464599609375,
-0.0035228729248046875,
0.0138092041015625,
-0.022003173828125,
0.042816162109375,
0.00295257568359375,
0.0328369140625,
0.00897216796875,
-0.04779052734375,
-0.06378173828125,
-0.025543212890625,
0.03021240234375,
0.025146484375,
-0.058563232421875,
0.0158843994140625,
-0.0160369873046875,
-0.039520263671875,
-0.06494140625,
0.00839996337890625,
0.037384033203125,
0.041015625,
0.03948974609375,
-0.0097503662109375,
-0.03948974609375,
-0.06915283203125,
-0.005870819091796875,
-0.012115478515625,
0.0089569091796875,
0.0390625,
0.047882080078125,
-0.01477813720703125,
0.06878662109375,
-0.042816162109375,
-0.044677734375,
0.0051422119140625,
0.0048065185546875,
0.01190185546875,
0.042510986328125,
0.07208251953125,
-0.04730224609375,
-0.03814697265625,
-0.0010576248168945312,
-0.05877685546875,
-0.0219268798828125,
0.013580322265625,
-0.023712158203125,
0.024993896484375,
0.01366424560546875,
-0.03485107421875,
0.04669189453125,
0.043975830078125,
-0.0105743408203125,
0.03900146484375,
-0.0083160400390625,
-0.00847625732421875,
-0.0706787109375,
0.022430419921875,
-0.00716400146484375,
0.00621795654296875,
-0.013671875,
0.0296173095703125,
-0.01430511474609375,
-0.0037078857421875,
-0.043975830078125,
0.044677734375,
-0.00708770751953125,
-0.005130767822265625,
-0.01480865478515625,
-0.018524169921875,
-0.0221710205078125,
0.040496826171875,
0.0005154609680175781,
0.0843505859375,
0.034576416015625,
-0.05194091796875,
0.0165863037109375,
0.032470703125,
-0.0200042724609375,
0.0257110595703125,
-0.06646728515625,
0.0081787109375,
0.00848388671875,
0.01788330078125,
-0.039398193359375,
-0.032318115234375,
0.043365478515625,
-0.03204345703125,
0.0031108856201171875,
-0.038909912109375,
-0.0390625,
-0.011138916015625,
-0.00394439697265625,
0.04510498046875,
0.04376220703125,
-0.06475830078125,
0.0266571044921875,
0.0295257568359375,
0.0036945343017578125,
-0.054412841796875,
-0.076171875,
-0.0144500732421875,
-0.026763916015625,
-0.0279541015625,
0.021392822265625,
-0.00548553466796875,
-0.0026569366455078125,
0.025299072265625,
0.0037822723388671875,
-0.02783203125,
0.00482177734375,
0.02789306640625,
0.02471923828125,
-0.0281829833984375,
-0.002826690673828125,
0.01354217529296875,
-0.022674560546875,
0.004810333251953125,
-0.0257415771484375,
0.0195770263671875,
-0.0033054351806640625,
-0.0246429443359375,
-0.045013427734375,
-0.0037822723388671875,
0.0248870849609375,
-0.034637451171875,
0.038238525390625,
0.0200653076171875,
-0.0217742919921875,
-0.0020656585693359375,
-0.032135009765625,
0.01529693603515625,
-0.04351806640625,
0.0051727294921875,
-0.026519775390625,
-0.047210693359375,
0.02685546875,
0.0125732421875,
0.025146484375,
0.057464599609375,
0.05194091796875,
0.0059356689453125,
0.04571533203125,
0.027130126953125,
-0.011505126953125,
0.039581298828125,
-0.03662109375,
-0.00476837158203125,
-0.06854248046875,
-0.035980224609375,
-0.0498046875,
-0.022552490234375,
-0.055450439453125,
-0.005184173583984375,
0.022674560546875,
-0.01050567626953125,
-0.049560546875,
0.037322998046875,
-0.035491943359375,
0.0430908203125,
0.055328369140625,
0.00788116455078125,
0.016754150390625,
0.01142120361328125,
0.005771636962890625,
0.00466156005859375,
-0.045562744140625,
-0.034576416015625,
0.114501953125,
0.0274505615234375,
0.0634765625,
0.0406494140625,
0.06787109375,
0.0115203857421875,
0.037994384765625,
-0.033966064453125,
0.0321044921875,
-0.024017333984375,
-0.0706787109375,
-0.00888824462890625,
-0.0565185546875,
-0.0772705078125,
0.00865936279296875,
0.0089874267578125,
-0.03741455078125,
0.0159149169921875,
0.0193634033203125,
-0.00592041015625,
0.040252685546875,
-0.04376220703125,
0.0604248046875,
-0.032440185546875,
-0.0198974609375,
-0.01506805419921875,
-0.03265380859375,
0.042999267578125,
-0.0024318695068359375,
0.0028057098388671875,
-0.016387939453125,
-0.0094146728515625,
0.07171630859375,
-0.04925537109375,
0.07110595703125,
-0.032684326171875,
-0.003582000732421875,
0.051422119140625,
-0.0158233642578125,
0.0059661865234375,
0.0034656524658203125,
-0.0186614990234375,
0.03216552734375,
0.006092071533203125,
-0.030487060546875,
-0.009490966796875,
0.05426025390625,
-0.1036376953125,
-0.0212554931640625,
-0.035125732421875,
-0.0209503173828125,
0.0125885009765625,
0.0245819091796875,
0.046356201171875,
0.0252685546875,
0.0155792236328125,
0.0056610107421875,
0.032989501953125,
-0.01064300537109375,
0.01372528076171875,
0.047088623046875,
-0.00763702392578125,
-0.056121826171875,
0.0849609375,
0.0151519775390625,
0.0237274169921875,
0.007843017578125,
0.02227783203125,
-0.029327392578125,
-0.040679931640625,
-0.0443115234375,
0.0242156982421875,
-0.062042236328125,
-0.037811279296875,
-0.01186370849609375,
-0.01422119140625,
-0.038482666015625,
0.006549835205078125,
-0.030548095703125,
-0.030914306640625,
-0.0489501953125,
-0.0095367431640625,
0.0277862548828125,
0.06298828125,
-0.013702392578125,
0.044189453125,
-0.03131103515625,
0.0169525146484375,
0.0195770263671875,
0.04425048828125,
0.006053924560546875,
-0.05712890625,
-0.01265716552734375,
0.00829315185546875,
-0.04376220703125,
-0.04541015625,
0.01654052734375,
0.004383087158203125,
0.05743408203125,
0.0276031494140625,
0.00628662109375,
0.060089111328125,
-0.02294921875,
0.06890869140625,
0.01123809814453125,
-0.07781982421875,
0.044097900390625,
-0.014007568359375,
-0.0112762451171875,
0.05206298828125,
0.032623291015625,
-0.00782012939453125,
-0.0200653076171875,
-0.06103515625,
-0.06646728515625,
0.061370849609375,
0.026580810546875,
0.005115509033203125,
0.02001953125,
0.05474853515625,
0.0043487548828125,
0.020477294921875,
-0.08935546875,
-0.0173492431640625,
-0.033660888671875,
-0.016876220703125,
0.0126495361328125,
-0.01383209228515625,
-0.0242156982421875,
-0.01119232177734375,
0.078369140625,
0.0010595321655273438,
0.0211029052734375,
0.004486083984375,
0.001087188720703125,
-0.016357421875,
0.006336212158203125,
0.034454345703125,
0.033477783203125,
-0.0183563232421875,
-0.0108489990234375,
0.0009784698486328125,
-0.0253143310546875,
0.01082611083984375,
0.0095367431640625,
-0.023529052734375,
-0.0181121826171875,
0.023773193359375,
0.06890869140625,
0.01029205322265625,
-0.0369873046875,
0.038238525390625,
0.007762908935546875,
-0.038818359375,
-0.0202484130859375,
0.02679443359375,
-0.0150909423828125,
0.0224456787109375,
-0.00431060791015625,
-0.0047149658203125,
0.006885528564453125,
-0.02203369140625,
-0.001285552978515625,
0.01349639892578125,
0.005764007568359375,
-0.0242919921875,
0.0716552734375,
0.032318115234375,
-0.0007548332214355469,
0.0615234375,
-0.00437164306640625,
-0.0341796875,
0.055999755859375,
0.04803466796875,
0.06317138671875,
-0.014190673828125,
-0.0027103424072265625,
0.036163330078125,
0.0280609130859375,
-0.0084381103515625,
0.030548095703125,
0.0005316734313964844,
-0.0298614501953125,
-0.018829345703125,
-0.0270843505859375,
-0.048126220703125,
0.018829345703125,
-0.06475830078125,
0.0377197265625,
-0.04266357421875,
-0.0138092041015625,
-0.00847625732421875,
0.0081329345703125,
-0.052703857421875,
0.01425933837890625,
0.017120361328125,
0.0758056640625,
-0.058441162109375,
0.07781982421875,
0.041015625,
-0.0697021484375,
-0.080322265625,
-0.01116943359375,
-0.00443267822265625,
-0.055877685546875,
0.03790283203125,
0.005245208740234375,
-0.00018739700317382812,
-0.0015201568603515625,
-0.05865478515625,
-0.0684814453125,
0.10699462890625,
0.041717529296875,
-0.0264129638671875,
-0.00986480712890625,
0.0209197998046875,
0.0269622802734375,
-0.0213470458984375,
0.01007080078125,
0.030853271484375,
0.034027099609375,
0.00983428955078125,
-0.07757568359375,
-0.0013113021850585938,
-0.0092620849609375,
-0.01096343994140625,
-0.0033969879150390625,
-0.0665283203125,
0.08978271484375,
-0.01226806640625,
-0.01486968994140625,
0.0233154296875,
0.03289794921875,
0.016082763671875,
0.03662109375,
0.022674560546875,
0.06341552734375,
0.060302734375,
-0.00479888916015625,
0.08123779296875,
-0.00794219970703125,
0.044097900390625,
0.0977783203125,
-0.0161895751953125,
0.0731201171875,
0.0177459716796875,
-0.0218658447265625,
0.047760009765625,
0.052947998046875,
-0.030548095703125,
0.048980712890625,
-0.006618499755859375,
-0.0005803108215332031,
-0.00039196014404296875,
0.0127105712890625,
-0.06475830078125,
0.0310516357421875,
0.005466461181640625,
-0.0203857421875,
-0.01068115234375,
-0.0205078125,
-0.0065765380859375,
-0.0256500244140625,
-0.011505126953125,
0.059478759765625,
0.005615234375,
-0.038848876953125,
0.044769287109375,
0.01244354248046875,
0.042877197265625,
-0.0716552734375,
0.007183074951171875,
-0.008758544921875,
0.017852783203125,
-0.03204345703125,
-0.050201416015625,
0.0189056396484375,
0.0257720947265625,
0.0122528076171875,
0.004398345947265625,
0.0234375,
-0.0238494873046875,
-0.0141754150390625,
0.0192413330078125,
0.0004851818084716797,
0.0296630859375,
0.02069091796875,
-0.04296875,
0.0169219970703125,
0.00664520263671875,
-0.00618743896484375,
0.023345947265625,
0.0184478759765625,
0.00824737548828125,
0.044769287109375,
0.06353759765625,
0.0122222900390625,
0.00023353099822998047,
-0.00970458984375,
0.07855224609375,
-0.0305938720703125,
-0.050872802734375,
-0.04876708984375,
0.038818359375,
0.0080718994140625,
-0.06195068359375,
0.03778076171875,
0.058319091796875,
0.09173583984375,
-0.004039764404296875,
0.068359375,
-0.00762176513671875,
0.0111846923828125,
-0.05511474609375,
0.05755615234375,
-0.06689453125,
0.023468017578125,
-0.007289886474609375,
-0.05230712890625,
-0.007373809814453125,
0.037628173828125,
-0.030548095703125,
-0.0013904571533203125,
0.0231170654296875,
0.04278564453125,
-0.0169830322265625,
0.00995635986328125,
0.0009641647338867188,
0.0026912689208984375,
0.04559326171875,
0.04248046875,
0.051605224609375,
-0.03173828125,
0.055206298828125,
-0.037841796875,
-0.0294952392578125,
-0.02801513671875,
-0.0628662109375,
-0.060638427734375,
-0.0219879150390625,
-0.0232696533203125,
-0.03021240234375,
-0.004535675048828125,
0.07611083984375,
0.06219482421875,
-0.058319091796875,
-0.03985595703125,
0.0036830902099609375,
-0.002437591552734375,
-0.01343536376953125,
-0.01317596435546875,
0.02362060546875,
-0.01702880859375,
-0.046539306640625,
0.013580322265625,
-0.022216796875,
0.0338134765625,
-0.0141448974609375,
-0.027557373046875,
-0.0002472400665283203,
0.00385284423828125,
0.017333984375,
0.03643798828125,
-0.06280517578125,
-0.00328826904296875,
-0.024810791015625,
-0.01019287109375,
0.0159912109375,
0.05230712890625,
-0.051422119140625,
0.031494140625,
0.00984954833984375,
0.042877197265625,
0.041351318359375,
-0.005062103271484375,
0.01210784912109375,
-0.046630859375,
0.023284912109375,
0.0019016265869140625,
0.0180206298828125,
0.0124053955078125,
-0.0401611328125,
0.0628662109375,
0.0178375244140625,
-0.054931640625,
-0.05926513671875,
0.00861358642578125,
-0.069580078125,
-0.00556182861328125,
0.07379150390625,
-0.0215911865234375,
-0.0083770751953125,
0.0196990966796875,
-0.01464080810546875,
0.021392822265625,
-0.0292205810546875,
0.04730224609375,
0.028564453125,
-0.00690460205078125,
0.0181121826171875,
-0.051116943359375,
0.032379150390625,
0.00647735595703125,
-0.07037353515625,
0.002353668212890625,
0.03021240234375,
0.045074462890625,
-0.0035552978515625,
0.04736328125,
-0.0107574462890625,
0.024200439453125,
0.0004115104675292969,
0.006267547607421875,
-0.0423583984375,
-0.02301025390625,
-0.0357666015625,
-0.0013141632080078125,
0.0161895751953125,
-0.041717529296875
]
] |
Monero/Manticore-13b-Chat-Pyg-Guanaco | 2023-05-27T05:32:39.000Z | [
"transformers",
"pytorch",
"llama",
"text-generation",
"manticore",
"endpoints_compatible",
"has_space",
"text-generation-inference",
"region:us"
] | text-generation | Monero | null | null | Monero/Manticore-13b-Chat-Pyg-Guanaco | 15 | 5,905 | transformers | 2023-05-26T18:43:08 | ---
tags: ["manticore"]
---
Manticore-13b-Chat-Pyg with the Guanaco 13b qLoRa from TimDettmers applied
| 109 | [
[
-0.005260467529296875,
-0.01837158203125,
0.029052734375,
0.07611083984375,
-0.02294921875,
0.01001739501953125,
-0.0181121826171875,
-0.01971435546875,
0.04241943359375,
0.022674560546875,
-0.051239013671875,
-0.05908203125,
-0.0318603515625,
0.01349639892578125,
-0.01151275634765625,
0.038604736328125,
0.041473388671875,
-0.0084075927734375,
0.07037353515625,
-0.00991058349609375,
-0.04718017578125,
-0.026641845703125,
-0.0679931640625,
-0.0102386474609375,
0.07025146484375,
0.035858154296875,
0.0692138671875,
0.01049041748046875,
0.0135498046875,
0.0247650146484375,
-0.03369140625,
0.0294647216796875,
-0.0043487548828125,
-0.001819610595703125,
-0.03155517578125,
-0.041168212890625,
-0.0275115966796875,
0.004581451416015625,
0.026123046875,
0.01433563232421875,
-0.056304931640625,
0.00655364990234375,
-0.006008148193359375,
0.00914764404296875,
-0.01971435546875,
0.00524139404296875,
-0.0265045166015625,
0.007801055908203125,
-0.0312042236328125,
-0.0291900634765625,
-0.04736328125,
-0.050994873046875,
-0.0006170272827148438,
-0.0626220703125,
0.00263214111328125,
0.028289794921875,
0.0638427734375,
-0.041107177734375,
-0.00800323486328125,
-0.02764892578125,
-0.032684326171875,
0.0404052734375,
-0.01540374755859375,
-0.0005207061767578125,
0.03289794921875,
0.040313720703125,
-0.046630859375,
-0.07281494140625,
-0.01166534423828125,
-0.0202789306640625,
0.003170013427734375,
0.005939483642578125,
-0.013671875,
0.008758544921875,
-0.025146484375,
0.0272216796875,
-0.026580810546875,
0.0229339599609375,
-0.07232666015625,
-0.035552978515625,
0.031097412109375,
0.033203125,
0.022491455078125,
0.0016012191772460938,
-0.035308837890625,
-0.0118865966796875,
-0.0185546875,
-0.05047607421875,
0.0191802978515625,
0.0318603515625,
-0.0169677734375,
0.02783203125,
-0.0014276504516601562,
0.0211334228515625,
0.0224761962890625,
0.0210113525390625,
0.0009889602661132812,
-0.0257415771484375,
-0.046966552734375,
-0.00978851318359375,
0.05181884765625,
0.046112060546875,
0.0168609619140625,
0.010833740234375,
0.00955963134765625,
-0.0257110595703125,
-0.007335662841796875,
-0.055206298828125,
-0.04443359375,
0.01410675048828125,
-0.0330810546875,
-0.0069580078125,
0.0307769775390625,
-0.051788330078125,
-0.034393310546875,
0.0106048583984375,
0.04254150390625,
-0.030853271484375,
-0.039031982421875,
0.0013742446899414062,
-0.0413818359375,
0.022186279296875,
0.044647216796875,
-0.048492431640625,
0.031005859375,
0.02313232421875,
0.054412841796875,
0.020904541015625,
-0.038665771484375,
-0.0113372802734375,
0.011383056640625,
-0.07147216796875,
0.057342529296875,
-0.0275421142578125,
-0.04248046875,
0.00312042236328125,
0.025787353515625,
-0.048553466796875,
-0.06158447265625,
0.0908203125,
-0.018035888671875,
0.05914306640625,
-0.01175689697265625,
-0.036712646484375,
-0.0279388427734375,
-0.0247955322265625,
-0.01395416259765625,
0.0731201171875,
0.0697021484375,
-0.029571533203125,
0.01433563232421875,
-0.05859375,
0.0194244384765625,
0.036590576171875,
0.004741668701171875,
0.006923675537109375,
0.036376953125,
-0.0228118896484375,
0.018280029296875,
-0.039276123046875,
0.00261688232421875,
0.004741668701171875,
-0.0428466796875,
0.0215301513671875,
-0.0185546875,
0.10198974609375,
0.043365478515625,
-0.00505828857421875,
0.034576416015625,
-0.02813720703125,
0.030548095703125,
0.01200103759765625,
-0.027069091796875,
-0.01239776611328125,
-0.06396484375,
0.0228271484375,
0.0021915435791015625,
0.01250457763671875,
-0.040618896484375,
0.00720977783203125,
0.004116058349609375,
0.03875732421875,
0.0557861328125,
0.0104522705078125,
0.017059326171875,
-0.0021266937255859375,
0.04290771484375,
-0.0029315948486328125,
0.02618408203125,
0.0443115234375,
-0.035003662109375,
-0.036376953125,
-0.032958984375,
0.0223236083984375,
0.049224853515625,
-0.048492431640625,
0.0341796875,
0.043121337890625,
-0.06884765625,
-0.017578125,
-0.0025691986083984375,
0.00959014892578125,
-0.023162841796875,
0.001407623291015625,
-0.00815582275390625,
-0.044281005859375,
-0.0703125,
0.01050567626953125,
-0.0250396728515625,
-0.0167999267578125,
0.005039215087890625,
0.0205230712890625,
-0.032379150390625,
-0.002002716064453125,
-0.08416748046875,
-0.0621337890625,
-0.014678955078125,
-0.00021338462829589844,
0.018463134765625,
0.03350830078125,
0.09521484375,
-0.0014171600341796875,
-0.033355712890625,
-0.01074981689453125,
-0.027130126953125,
-0.028961181640625,
-0.002010345458984375,
-0.03668212890625,
-0.0109405517578125,
0.0175933837890625,
-0.056488037109375,
0.015411376953125,
0.05328369140625,
-0.041717529296875,
0.041168212890625,
-0.07659912109375,
0.04522705078125,
-0.06072998046875,
-0.03082275390625,
0.0036258697509765625,
-0.01068115234375,
-0.042144775390625,
0.009185791015625,
-0.00223541259765625,
-0.013458251953125,
-0.03143310546875,
0.0411376953125,
-0.04144287109375,
0.03466796875,
-0.0201416015625,
-0.00524139404296875,
0.005176544189453125,
0.0201263427734375,
-0.0180206298828125,
0.0892333984375,
0.055816650390625,
-0.021514892578125,
0.049346923828125,
0.0390625,
0.01222991943359375,
0.032562255859375,
-0.0416259765625,
0.00331878662109375,
0.02593994140625,
0.00677490234375,
-0.0789794921875,
-0.034698486328125,
0.04193115234375,
-0.03582763671875,
-0.0208282470703125,
-0.005084991455078125,
-0.0265350341796875,
-0.045135498046875,
-0.04095458984375,
0.039031982421875,
0.0158233642578125,
-0.039764404296875,
0.032562255859375,
0.0102996826171875,
0.0172271728515625,
0.0036144256591796875,
-0.0665283203125,
0.048004150390625,
-0.0391845703125,
-0.024322509765625,
0.0162811279296875,
0.0133209228515625,
-0.0121002197265625,
-0.034393310546875,
0.0013275146484375,
-0.048126220703125,
-0.011444091796875,
0.021514892578125,
0.000946044921875,
-0.0196075439453125,
-0.032928466796875,
-0.0167236328125,
-0.01250457763671875,
0.0104522705078125,
-0.00608062744140625,
0.06060791015625,
0.00916290283203125,
0.0033512115478515625,
-0.0728759765625,
0.01824951171875,
0.045166015625,
0.0129547119140625,
0.04241943359375,
0.035919189453125,
-0.038238525390625,
0.01110076904296875,
-0.0247955322265625,
-0.01016998291015625,
-0.036651611328125,
0.005615234375,
-0.0227508544921875,
-0.0428466796875,
0.06121826171875,
0.0147552490234375,
-0.04254150390625,
0.043914794921875,
0.032958984375,
-0.0133056640625,
0.0148773193359375,
0.0413818359375,
0.0137939453125,
0.0517578125,
0.00152587890625,
0.003795623779296875,
-0.011199951171875,
-0.02838134765625,
-0.06005859375,
-0.0031604766845703125,
-0.028533935546875,
-0.016143798828125,
0.005336761474609375,
0.00711822509765625,
0.0008220672607421875,
0.0682373046875,
-0.04705810546875,
0.03656005859375,
0.0684814453125,
0.045501708984375,
0.013916015625,
-0.029754638671875,
0.021270751953125,
-0.005481719970703125,
-0.0516357421875,
-0.00322723388671875,
0.05291748046875,
0.015655517578125,
0.054534912109375,
0.0284423828125,
0.032745361328125,
0.036590576171875,
0.00763702392578125,
-0.0182342529296875,
0.04449462890625,
-0.007415771484375,
-0.06219482421875,
0.006130218505859375,
0.0104522705078125,
-0.07427978515625,
0.01194000244140625,
0.0158843994140625,
-0.04766845703125,
0.032958984375,
-0.0019893646240234375,
-0.04718017578125,
-0.01233673095703125,
-0.101318359375,
0.060791015625,
0.0006380081176757812,
0.0102081298828125,
-0.06201171875,
-0.0236968994140625,
0.03314208984375,
0.0280303955078125,
-0.03204345703125,
0.00684356689453125,
0.022918701171875,
0.05517578125,
-0.05291748046875,
0.0293731689453125,
-0.0369873046875,
-0.0312042236328125,
0.07244873046875,
0.0276947021484375,
-0.005573272705078125,
0.0237884521484375,
0.007022857666015625,
-0.021697998046875,
0.0284881591796875,
-0.03717041015625,
-0.03338623046875,
0.03887939453125,
-0.047393798828125,
-0.03857421875,
-0.048248291015625,
-0.0196380615234375,
0.039825439453125,
-0.01546478271484375,
-0.00341033935546875,
0.06048583984375,
0.013641357421875,
0.026580810546875,
0.017913818359375,
-0.023834228515625,
0.03533935546875,
0.0015316009521484375,
-0.040557861328125,
-0.056976318359375,
0.048583984375,
0.00024271011352539062,
0.0237884521484375,
0.024871826171875,
0.02685546875,
-0.045166015625,
-0.060821533203125,
-0.005313873291015625,
0.0177001953125,
-0.033294677734375,
-0.0103607177734375,
-0.00600433349609375,
-0.04327392578125,
0.003421783447265625,
0.01203155517578125,
-0.00919342041015625,
-0.037139892578125,
-0.01708984375,
-0.002582550048828125,
0.055938720703125,
0.0193328857421875,
-0.03448486328125,
0.09136962890625,
-0.059906005859375,
0.00461578369140625,
0.0158233642578125,
0.0001512765884399414,
-0.032379150390625,
-0.06939697265625,
-0.0232696533203125,
0.0028285980224609375,
-0.01322174072265625,
-0.06121826171875,
0.00978851318359375,
0.030242919921875,
0.0364990234375,
0.032379150390625,
0.005023956298828125,
0.0650634765625,
0.0015850067138671875,
0.02294921875,
-0.006244659423828125,
-0.04949951171875,
0.046112060546875,
-0.051483154296875,
-0.0018262863159179688,
0.0419921875,
0.0203704833984375,
-0.0443115234375,
-0.03863525390625,
-0.05657958984375,
-0.0272369384765625,
0.037506103515625,
0.04931640625,
0.0005059242248535156,
-0.0224456787109375,
0.03240966796875,
-0.01184844970703125,
0.0002002716064453125,
-0.023223876953125,
-0.010589599609375,
-0.0030422210693359375,
-0.00693511962890625,
0.037811279296875,
-0.048675537109375,
-0.02099609375,
-0.009124755859375,
0.009002685546875,
0.01042938232421875,
0.05126953125,
0.014801025390625,
-0.0110015869140625,
0.0224761962890625,
0.021942138671875,
0.09625244140625,
0.062347412109375,
-0.062744140625,
0.009613037109375,
-0.006244659423828125,
-0.061279296875,
0.040130615234375,
-0.01461029052734375,
0.01428985595703125,
-0.001220703125,
0.0175628662109375,
0.0222015380859375,
0.004566192626953125,
-0.0187225341796875,
0.01385498046875,
-0.0184326171875,
-0.01611328125,
-0.0198822021484375,
0.01446533203125,
0.008270263671875,
0.044586181640625,
0.050872802734375,
-0.00038313865661621094,
-0.00826263427734375,
-0.046844482421875,
0.01303863525390625,
-0.004306793212890625,
-0.0302734375,
-0.03411865234375,
0.049957275390625,
0.0299530029296875,
-0.039093017578125,
0.0292816162109375,
-0.0200958251953125,
-0.0252838134765625,
0.0318603515625,
0.05523681640625,
0.06500244140625,
-0.043212890625,
0.0289764404296875,
0.00879669189453125,
0.0269012451171875,
-0.005985260009765625,
0.03204345703125,
0.038330078125,
-0.0631103515625,
-0.00783538818359375,
0.01165008544921875,
-0.0279083251953125,
0.0020656585693359375,
-0.056121826171875,
0.006122589111328125,
-0.057373046875,
-0.0289154052734375,
-0.006275177001953125,
0.0107421875,
-0.012542724609375,
0.0280914306640625,
-0.0011081695556640625,
0.09783935546875,
-0.079833984375,
0.06463623046875,
0.0274200439453125,
-0.034332275390625,
-0.03955078125,
-0.0267181396484375,
-0.005992889404296875,
-0.06939697265625,
0.05218505859375,
-0.0201416015625,
-0.004482269287109375,
0.0133056640625,
-0.0193328857421875,
-0.0190277099609375,
0.09197998046875,
0.035552978515625,
-0.0206146240234375,
0.0154266357421875,
-0.042083740234375,
0.0280303955078125,
0.001399993896484375,
0.05316162109375,
0.0311431884765625,
0.046142578125,
-0.004058837890625,
-0.07794189453125,
0.003078460693359375,
0.0110931396484375,
-0.00010013580322265625,
-0.00022971630096435547,
-0.06060791015625,
0.033203125,
0.0191497802734375,
-0.013031005859375,
0.03582763671875,
0.078857421875,
0.0207977294921875,
-0.0188751220703125,
0.0203399658203125,
0.0294342041015625,
0.05047607421875,
-0.034576416015625,
0.057891845703125,
0.00576019287109375,
0.0254364013671875,
0.046722412109375,
-0.01568603515625,
0.0183563232421875,
0.01457977294921875,
-0.0374755859375,
0.0148162841796875,
0.08697509765625,
0.014129638671875,
0.05267333984375,
0.00835418701171875,
-0.035552978515625,
0.005252838134765625,
-0.0201263427734375,
-0.0239410400390625,
0.041717529296875,
0.015533447265625,
0.00022590160369873047,
-0.038726806640625,
-0.0133209228515625,
0.0201873779296875,
0.035614013671875,
-0.035064697265625,
0.0188446044921875,
0.00783538818359375,
-0.0088653564453125,
0.055816650390625,
0.01142120361328125,
0.0428466796875,
-0.0526123046875,
0.0187225341796875,
-0.0285186767578125,
0.054931640625,
-0.0249786376953125,
-0.07342529296875,
0.038787841796875,
-0.037750244140625,
-0.004383087158203125,
0.0013456344604492188,
0.050445556640625,
-0.02606201171875,
-0.0254974365234375,
0.037384033203125,
0.04638671875,
0.01515960693359375,
-0.040313720703125,
-0.033050537109375,
-0.00888824462890625,
-0.0081787109375,
0.004993438720703125,
0.023406982421875,
0.059844970703125,
-0.0345458984375,
0.04876708984375,
-0.0076141357421875,
-0.0027484893798828125,
0.0150909423828125,
0.0233001708984375,
0.04815673828125,
-0.06011962890625,
-0.01226043701171875,
-0.07275390625,
-0.01175689697265625,
0.01389312744140625,
-0.058349609375,
0.047088623046875,
0.0587158203125,
0.0175933837890625,
-0.0237274169921875,
0.01241302490234375,
-0.0322265625,
0.01198577880859375,
-0.029693603515625,
0.03948974609375,
0.0201873779296875,
-0.03558349609375,
-0.01178741455078125,
-0.084716796875,
-0.035736083984375,
0.043701171875,
0.0302886962890625,
-0.00563812255859375,
0.0880126953125,
0.06475830078125,
0.00968170166015625,
0.0124664306640625,
0.0186309814453125,
0.00899505615234375,
0.025390625,
0.059600830078125,
0.07177734375,
-0.041168212890625,
0.0281829833984375,
-0.014801025390625,
-0.0175323486328125,
-0.0216827392578125,
-0.048828125,
-0.058807373046875,
-0.0289154052734375,
-0.02667236328125,
-0.0288848876953125,
0.0269927978515625,
0.11480712890625,
0.0191802978515625,
-0.06353759765625,
-0.038116455078125,
0.02044677734375,
-0.00989532470703125,
-0.01715087890625,
-0.013458251953125,
0.03253173828125,
0.0193939208984375,
-0.04693603515625,
0.035919189453125,
0.0130615234375,
0.031158447265625,
0.0079803466796875,
-0.00312042236328125,
-0.0301055908203125,
-0.01885986328125,
0.0199127197265625,
0.0009045600891113281,
-0.038055419921875,
-0.0589599609375,
0.006122589111328125,
-0.0155029296875,
0.0302276611328125,
0.0457763671875,
-0.0184478759765625,
-0.05767822265625,
0.044464111328125,
0.0316162109375,
0.02508544921875,
0.03082275390625,
0.0506591796875,
-0.0499267578125,
0.05706787109375,
-0.0163116455078125,
0.00521087646484375,
-0.00263214111328125,
-0.01282501220703125,
0.04962158203125,
0.03302001953125,
-0.036865234375,
-0.036590576171875,
-0.03546142578125,
-0.094970703125,
-0.010650634765625,
0.04534912109375,
0.0002665519714355469,
-0.032135009765625,
0.009490966796875,
-0.08636474609375,
-0.00548553466796875,
-0.049224853515625,
0.02508544921875,
0.046905517578125,
0.006916046142578125,
-0.05279541015625,
-0.0249786376953125,
-0.00037217140197753906,
0.01837158203125,
-0.0506591796875,
0.009857177734375,
0.00955963134765625,
0.0022907257080078125,
0.01143646240234375,
0.058868408203125,
-0.00376129150390625,
0.0269622802734375,
-0.003055572509765625,
0.021453857421875,
0.011383056640625,
-0.0294647216796875,
-0.007007598876953125,
0.0180206298828125,
0.04815673828125,
-0.07049560546875
]
] |
beomi/KoRWKV-6B | 2023-07-20T01:07:48.000Z | [
"transformers",
"pytorch",
"safetensors",
"rwkv",
"text-generation",
"KoRWKV",
"ko",
"doi:10.57967/hf/1292",
"license:mit",
"endpoints_compatible",
"region:us"
] | text-generation | beomi | null | null | beomi/KoRWKV-6B | 3 | 5,904 | transformers | 2023-05-26T07:24:57 | ---
license: mit
language:
- ko
pipeline_tag: text-generation
tags:
- KoRWKV
---
> Instruction-Finetuned model is available at [beomi/KoAlpaca-KoRWKV-6B](https://huggingface.co/beomi/KoAlpaca-KoRWKV-6B)
# KoRWKV Model Card
KoRWKV (6B) trained on Korean dataset with RWKVv4 Neo Architecture.
## Model details
**Researcher developing the model**
Junbum Lee (aka Beomi)
**Model date**
KoRWKV was trained between 2023.05~2023.07
**Model version**
This is 1st release of the model.
**Model type**
Find more about RWKV at https://github.com/BlinkDL/RWKV-LM
**License**
MIT
## Intended use
**Primary intended uses**
The primary use of KoRWKV is research on Korean Opensource large language models
**Primary intended users**
The primary intended users of the model are researchers in natural language processing, machine learning and artificial intelligence.
**Out-of-scope use cases**
KoRWKV is a base, or foundational, model. As such, it should not be used on downstream applications without further risk evaluation and mitigation. In particular, our model has not been trained with human feedback, and can thus generate toxic or offensive content, incorrect information or generally unhelpful answers.
## Ethical considerations
**Data**
The data used to train the model is collected from various sources, mostly from the Web. As such, it contains offensive, harmful and biased content. We thus expect the model to exhibit such biases from the training data.
**Human life**
The model is not intended to inform decisions about matters central to human life, and should not be used in such a way.
**Risks and harms**
Risks and harms of large language models include the generation of harmful, offensive or biased content. These models are often prone to generating incorrect information, sometimes referred to as hallucinations. We do not expect our model to be an exception in this regard.
**Use cases**
KoRWKV is a foundational model, and as such, it should not be used for downstream applications without further investigation and mitigations of risks. These risks and potential fraught use cases include, but are not limited to: generation of misinformation and generation of harmful, biased or offensive content. | 2,238 | [
[
-0.00882720947265625,
-0.068115234375,
0.00913238525390625,
0.0174407958984375,
-0.0309906005859375,
-0.0306396484375,
-0.0096435546875,
-0.054595947265625,
-0.0212860107421875,
0.06695556640625,
-0.031585693359375,
-0.042388916015625,
-0.0418701171875,
-0.0158843994140625,
-0.0127716064453125,
0.061981201171875,
0.0258941650390625,
0.013885498046875,
-0.005832672119140625,
-0.011322021484375,
-0.04595947265625,
-0.04693603515625,
-0.0472412109375,
-0.03875732421875,
0.031402587890625,
0.03802490234375,
0.07244873046875,
0.045074462890625,
0.01262664794921875,
0.018524169921875,
-0.0138397216796875,
-0.032989501953125,
-0.054595947265625,
-0.0015163421630859375,
-0.024932861328125,
-0.0265350341796875,
-0.031280517578125,
0.003139495849609375,
0.0341796875,
0.04998779296875,
-0.0161895751953125,
0.01467132568359375,
-0.00943756103515625,
0.053070068359375,
-0.055145263671875,
0.020660400390625,
-0.029296875,
0.01009368896484375,
-0.01517486572265625,
0.0158538818359375,
-0.055633544921875,
-0.04425048828125,
0.0250701904296875,
-0.044891357421875,
-0.017333984375,
0.0005736351013183594,
0.0596923828125,
0.005855560302734375,
-0.046661376953125,
-0.032867431640625,
-0.054473876953125,
0.0703125,
-0.0704345703125,
0.05169677734375,
0.034942626953125,
0.029998779296875,
-0.024688720703125,
-0.044677734375,
-0.030853271484375,
-0.03173828125,
0.01018524169921875,
0.0178985595703125,
-0.01424407958984375,
0.007793426513671875,
0.027099609375,
0.02166748046875,
-0.038665771484375,
0.01120758056640625,
-0.049835205078125,
-0.0296783447265625,
0.0478515625,
0.006702423095703125,
0.029296875,
-0.036346435546875,
-0.0243377685546875,
-0.0270538330078125,
-0.044281005859375,
0.0123748779296875,
0.04644775390625,
0.031280517578125,
-0.01549530029296875,
0.057647705078125,
-0.028045654296875,
0.0426025390625,
-0.00388336181640625,
-0.025726318359375,
0.0276031494140625,
-0.0169219970703125,
-0.02801513671875,
0.016143798828125,
0.0694580078125,
0.039031982421875,
0.0181121826171875,
-0.0185546875,
0.01666259765625,
0.016143798828125,
0.040924072265625,
-0.060516357421875,
-0.00943756103515625,
0.01529693603515625,
-0.04425048828125,
-0.036895751953125,
-0.0169677734375,
-0.0782470703125,
-0.011993408203125,
0.003345489501953125,
0.0281524658203125,
-0.047027587890625,
-0.032928466796875,
0.01084136962890625,
-0.0010175704956054688,
0.039764404296875,
-0.003063201904296875,
-0.055145263671875,
0.01837158203125,
0.0294342041015625,
0.027130126953125,
0.00179290771484375,
-0.0167236328125,
-0.00016021728515625,
0.0031375885009765625,
-0.032196044921875,
0.032867431640625,
-0.034332275390625,
-0.047210693359375,
0.006046295166015625,
0.0175323486328125,
0.022705078125,
-0.032928466796875,
0.07635498046875,
-0.04095458984375,
0.0165557861328125,
-0.0162811279296875,
-0.037109375,
-0.04437255859375,
0.0087890625,
-0.04461669921875,
0.08203125,
0.019561767578125,
-0.06622314453125,
0.0126495361328125,
-0.04449462890625,
-0.025360107421875,
0.0024929046630859375,
0.01012420654296875,
-0.037933349609375,
-0.012664794921875,
-0.004909515380859375,
0.0309600830078125,
-0.00023984909057617188,
0.02392578125,
-0.0307159423828125,
-0.016021728515625,
0.01346588134765625,
-0.0264434814453125,
0.07177734375,
0.036529541015625,
-0.01253509521484375,
-0.0003020763397216797,
-0.07427978515625,
0.0210418701171875,
0.0199737548828125,
-0.03582763671875,
-0.051544189453125,
-0.0156707763671875,
0.0164794921875,
0.01959228515625,
0.025543212890625,
-0.033843994140625,
0.00266265869140625,
-0.047027587890625,
0.018768310546875,
0.041015625,
-0.0022296905517578125,
0.0479736328125,
-0.0234375,
0.055633544921875,
-0.0003802776336669922,
0.03704833984375,
-0.00113677978515625,
-0.044830322265625,
-0.047393798828125,
0.0037841796875,
0.024932861328125,
0.054107666015625,
-0.045440673828125,
0.0283050537109375,
-0.0006399154663085938,
-0.07305908203125,
-0.0223388671875,
-0.02154541015625,
0.060546875,
0.0413818359375,
0.015106201171875,
-0.0034332275390625,
-0.044097900390625,
-0.07733154296875,
-0.0230560302734375,
-0.002834320068359375,
0.0258026123046875,
0.0282440185546875,
0.0273590087890625,
-0.0260467529296875,
0.04522705078125,
-0.0202484130859375,
0.01120758056640625,
-0.0164337158203125,
0.00867462158203125,
0.0014009475708007812,
0.037017822265625,
0.044708251953125,
-0.052642822265625,
-0.035430908203125,
0.0081939697265625,
-0.057159423828125,
-0.0169677734375,
0.0166168212890625,
-0.0147857666015625,
0.017852783203125,
0.0172271728515625,
-0.04901123046875,
0.03515625,
0.052154541015625,
-0.029388427734375,
0.05224609375,
0.0017976760864257812,
0.006961822509765625,
-0.1136474609375,
0.006359100341796875,
-0.012359619140625,
-0.0199432373046875,
-0.056427001953125,
0.02398681640625,
0.003765106201171875,
-0.00841522216796875,
-0.0501708984375,
0.0400390625,
-0.012969970703125,
-0.00311279296875,
-0.026397705078125,
-0.004608154296875,
-0.0014829635620117188,
0.029205322265625,
-0.00443267822265625,
0.049041748046875,
0.034210205078125,
-0.056365966796875,
0.01605224609375,
0.02337646484375,
-0.0196990966796875,
0.02520751953125,
-0.0552978515625,
0.0046844482421875,
-0.0038013458251953125,
0.00437164306640625,
-0.03131103515625,
-0.0286865234375,
0.055511474609375,
-0.052459716796875,
0.0125732421875,
-0.01531982421875,
-0.0239715576171875,
-0.016448974609375,
-0.0211181640625,
0.0234375,
0.06182861328125,
-0.031158447265625,
0.04638671875,
0.040679931640625,
-0.00647735595703125,
-0.0626220703125,
-0.04901123046875,
-0.023193359375,
-0.0308074951171875,
-0.042388916015625,
0.010833740234375,
-0.0045928955078125,
-0.009368896484375,
0.005931854248046875,
0.0039215087890625,
-0.01678466796875,
-0.0027904510498046875,
0.03753662109375,
0.0210113525390625,
-0.004833221435546875,
0.0028629302978515625,
0.00750732421875,
-0.0020809173583984375,
0.01134490966796875,
0.00775909423828125,
0.03662109375,
0.0028171539306640625,
-0.005298614501953125,
-0.032562255859375,
0.036895751953125,
0.0426025390625,
0.0111541748046875,
0.0491943359375,
0.042510986328125,
-0.03607177734375,
-0.006011962890625,
-0.036346435546875,
-0.0012712478637695312,
-0.03680419921875,
0.051910400390625,
-0.0017518997192382812,
-0.062469482421875,
0.0467529296875,
-0.00804901123046875,
-0.007213592529296875,
0.0426025390625,
0.055206298828125,
0.018310546875,
0.08050537109375,
0.049591064453125,
0.00458526611328125,
0.034515380859375,
-0.0072784423828125,
0.016143798828125,
-0.0587158203125,
-0.03680419921875,
-0.04620361328125,
0.016876220703125,
-0.052978515625,
-0.0151214599609375,
0.002162933349609375,
0.0258636474609375,
-0.0280609130859375,
0.025848388671875,
-0.042877197265625,
0.032012939453125,
0.04730224609375,
-0.00530242919921875,
0.013641357421875,
-0.011749267578125,
-0.0225830078125,
-0.007472991943359375,
-0.069580078125,
-0.048370361328125,
0.09771728515625,
0.045166015625,
0.076904296875,
-0.0033664703369140625,
0.0255279541015625,
0.0185546875,
0.00783538818359375,
-0.069091796875,
0.033782958984375,
0.0019350051879882812,
-0.056854248046875,
-0.0186309814453125,
-0.0277252197265625,
-0.07159423828125,
0.00846099853515625,
-0.022796630859375,
-0.058074951171875,
0.00626373291015625,
0.0238189697265625,
-0.0172882080078125,
0.0034465789794921875,
-0.07464599609375,
0.08026123046875,
-0.004730224609375,
-0.015106201171875,
0.0001722574234008789,
-0.03515625,
0.051239013671875,
-0.0011873245239257812,
0.0196380615234375,
0.00868988037109375,
0.0259246826171875,
0.055694580078125,
-0.040374755859375,
0.0758056640625,
-0.01239776611328125,
-0.0016336441040039062,
0.0291290283203125,
-0.0108489990234375,
0.044921875,
0.00861358642578125,
-0.0014085769653320312,
0.0283050537109375,
0.0256195068359375,
-0.0199737548828125,
-0.0209197998046875,
0.054656982421875,
-0.08367919921875,
-0.021026611328125,
-0.04656982421875,
-0.039031982421875,
-0.0045928955078125,
0.043792724609375,
0.045166015625,
0.032745361328125,
-0.017486572265625,
0.023101806640625,
0.051849365234375,
-0.037384033203125,
0.0025768280029296875,
0.042755126953125,
-0.032806396484375,
-0.03375244140625,
0.05584716796875,
0.00954437255859375,
0.0213470458984375,
-0.015106201171875,
0.039031982421875,
-0.03668212890625,
-0.04693603515625,
-0.0200958251953125,
0.00922393798828125,
-0.0692138671875,
-0.0126800537109375,
-0.048370361328125,
-0.03216552734375,
-0.050933837890625,
-0.0009083747863769531,
-0.058837890625,
-0.018585205078125,
-0.016845703125,
0.00582122802734375,
0.03790283203125,
0.057464599609375,
0.005008697509765625,
0.0343017578125,
-0.06439208984375,
0.01471710205078125,
0.023284912109375,
0.035797119140625,
-0.0043182373046875,
-0.039276123046875,
-0.0197601318359375,
0.0289306640625,
-0.029052734375,
-0.0631103515625,
0.0230560302734375,
-0.009918212890625,
0.053924560546875,
0.031707763671875,
0.02459716796875,
0.0364990234375,
-0.037322998046875,
0.0653076171875,
0.007465362548828125,
-0.056121826171875,
0.038116455078125,
-0.03790283203125,
0.041473388671875,
0.032318115234375,
0.0430908203125,
-0.0418701171875,
-0.030242919921875,
-0.0462646484375,
-0.05401611328125,
0.0594482421875,
0.016510009765625,
0.02093505859375,
-0.0146026611328125,
0.043731689453125,
0.015655517578125,
0.00045037269592285156,
-0.08563232421875,
-0.0189208984375,
-0.044281005859375,
-0.006710052490234375,
0.0239715576171875,
-0.0439453125,
0.00916290283203125,
-0.0152740478515625,
0.07611083984375,
0.01360321044921875,
0.021514892578125,
0.004833221435546875,
-0.0015039443969726562,
-0.01232147216796875,
0.032470703125,
0.04302978515625,
0.04266357421875,
-0.0178070068359375,
-0.017547607421875,
0.01209259033203125,
-0.06353759765625,
-0.01326751708984375,
-0.0034313201904296875,
-0.032012939453125,
-0.011871337890625,
0.00861358642578125,
0.05572509765625,
0.02685546875,
-0.059539794921875,
0.0282745361328125,
-0.002002716064453125,
-0.02886962890625,
-0.05078125,
0.01282501220703125,
0.01922607421875,
0.038543701171875,
-0.00864410400390625,
-0.01181793212890625,
0.0188751220703125,
-0.038787841796875,
-0.01004791259765625,
0.0296173095703125,
-0.0277862548828125,
-0.02374267578125,
0.050933837890625,
0.0029296875,
-0.0299072265625,
0.041473388671875,
-0.0267333984375,
-0.033447265625,
0.037384033203125,
0.058319091796875,
0.044921875,
-0.035797119140625,
0.0154571533203125,
0.04705810546875,
0.0186309814453125,
-0.0175933837890625,
0.0345458984375,
0.0288848876953125,
-0.0655517578125,
-0.029205322265625,
-0.046722412109375,
-0.02874755859375,
0.03729248046875,
-0.06494140625,
0.027313232421875,
-0.03009033203125,
-0.0261688232421875,
-0.004772186279296875,
-0.0109710693359375,
-0.0240478515625,
0.0180511474609375,
0.005245208740234375,
0.06292724609375,
-0.0643310546875,
0.07659912109375,
0.042327880859375,
-0.0265655517578125,
-0.0703125,
-0.0203094482421875,
-0.021728515625,
-0.043792724609375,
0.03448486328125,
0.0088653564453125,
-0.0014324188232421875,
-0.004207611083984375,
-0.050750732421875,
-0.06353759765625,
0.08319091796875,
0.0241851806640625,
-0.02935791015625,
-0.0041046142578125,
0.01372528076171875,
0.033111572265625,
-0.0210418701171875,
-0.0034008026123046875,
0.023101806640625,
0.017547607421875,
0.026397705078125,
-0.0621337890625,
-0.0024967193603515625,
-0.01190948486328125,
0.0194091796875,
-0.003093719482421875,
-0.056793212890625,
0.06768798828125,
0.0034923553466796875,
-0.01111602783203125,
0.004512786865234375,
0.0282745361328125,
0.021148681640625,
0.01824951171875,
0.026031494140625,
0.04571533203125,
0.04351806640625,
0.0098724365234375,
0.0794677734375,
-0.0160064697265625,
0.030975341796875,
0.0955810546875,
-0.007793426513671875,
0.06268310546875,
0.0205841064453125,
-0.00878143310546875,
0.022216796875,
0.041473388671875,
-0.0029163360595703125,
0.053741455078125,
-0.0008325576782226562,
0.002254486083984375,
-0.0102996826171875,
-0.004825592041015625,
-0.046356201171875,
0.05010986328125,
0.033966064453125,
-0.040313720703125,
-0.012908935546875,
0.01071929931640625,
0.022918701171875,
0.0072174072265625,
-0.0262908935546875,
0.04132080078125,
0.0020008087158203125,
-0.04351806640625,
0.0458984375,
0.0158843994140625,
0.04705810546875,
-0.061279296875,
-0.00925445556640625,
0.00762176513671875,
0.00908660888671875,
0.0002224445343017578,
-0.0271453857421875,
0.041107177734375,
-0.0048980712890625,
-0.01837158203125,
0.01006317138671875,
0.07989501953125,
-0.0177154541015625,
-0.05078125,
0.0215301513671875,
0.01519775390625,
0.014251708984375,
0.006954193115234375,
-0.0657958984375,
0.0169830322265625,
-0.00946807861328125,
-0.0252838134765625,
0.0278472900390625,
0.0116424560546875,
-0.014801025390625,
0.0709228515625,
0.04400634765625,
-0.004009246826171875,
0.02435302734375,
-0.002849578857421875,
0.0675048828125,
-0.049072265625,
-0.04266357421875,
-0.045806884765625,
0.0228271484375,
-0.00482177734375,
-0.0148162841796875,
0.06829833984375,
0.040802001953125,
0.0701904296875,
-0.01168060302734375,
0.0758056640625,
-0.0002944469451904297,
0.04364013671875,
-0.042816162109375,
0.0694580078125,
-0.027313232421875,
-0.0117950439453125,
-0.023345947265625,
-0.05877685546875,
-0.0026874542236328125,
0.0692138671875,
-0.0211334228515625,
0.0135498046875,
0.04840087890625,
0.056365966796875,
0.00040531158447265625,
0.0231475830078125,
0.033966064453125,
0.033966064453125,
0.00070953369140625,
0.03057861328125,
0.06072998046875,
-0.046142578125,
0.04278564453125,
-0.044097900390625,
-0.0129241943359375,
-0.0227203369140625,
-0.0771484375,
-0.08587646484375,
-0.020965576171875,
-0.028106689453125,
-0.03485107421875,
0.01297760009765625,
0.05999755859375,
0.0465087890625,
-0.055816650390625,
-0.0088958740234375,
-0.0030536651611328125,
0.0023670196533203125,
-0.00832366943359375,
-0.02105712890625,
0.01338958740234375,
-0.009918212890625,
-0.054473876953125,
0.017547607421875,
0.0021114349365234375,
0.00853729248046875,
-0.033935546875,
-0.0130157470703125,
-0.01369476318359375,
0.00687408447265625,
0.05657958984375,
0.032562255859375,
-0.057861328125,
-0.039520263671875,
0.0031871795654296875,
-0.00548553466796875,
-0.008880615234375,
0.0135650634765625,
-0.0367431640625,
0.0438232421875,
0.022308349609375,
0.028411865234375,
0.0130157470703125,
-0.011077880859375,
0.040863037109375,
-0.03375244140625,
0.026458740234375,
0.0140533447265625,
0.010955810546875,
0.0133819580078125,
-0.0487060546875,
0.0189666748046875,
0.00231170654296875,
-0.056671142578125,
-0.04638671875,
0.020599365234375,
-0.08892822265625,
-0.0147247314453125,
0.10369873046875,
-0.0121612548828125,
-0.0367431640625,
-0.017608642578125,
-0.0234527587890625,
0.0347900390625,
-0.025390625,
0.05328369140625,
0.062164306640625,
0.01074981689453125,
-0.01264190673828125,
-0.08428955078125,
0.036407470703125,
-0.005016326904296875,
-0.048736572265625,
0.004852294921875,
0.038055419921875,
0.04296875,
-0.0036678314208984375,
0.044921875,
-0.035736083984375,
0.0263519287109375,
0.01053619384765625,
0.0241851806640625,
-0.00789642333984375,
-0.0006031990051269531,
-0.0196533203125,
0.0006799697875976562,
0.012725830078125,
-0.01198577880859375
]
] |
FelixChao/vicuna-7B-chemical | 2023-08-25T12:15:59.000Z | [
"transformers",
"pytorch",
"llama",
"text-generation",
"chemistry",
"en",
"dataset:andersonbcdefg/chemistry",
"arxiv:1910.09700",
"license:apache-2.0",
"endpoints_compatible",
"has_space",
"text-generation-inference",
"region:us"
] | text-generation | FelixChao | null | null | FelixChao/vicuna-7B-chemical | 0 | 5,903 | transformers | 2023-08-09T04:36:39 | ---
license: apache-2.0
datasets:
- andersonbcdefg/chemistry
language:
- en
metrics:
- bleu
- rouge
pipeline_tag: text-generation
tags:
- chemistry
---
# Model Card for Model ID
<!-- Provide a quick summary of what the model is/does. -->
This modelcard aims to be a base template for new models. It has been generated using [this raw template](https://github.com/huggingface/huggingface_hub/blob/main/src/huggingface_hub/templates/modelcard_template.md?plain=1).
## Model Details
### Model Description
<!-- Provide a longer summary of what this model is. -->
This model is currently **under testing**. It can answer some **basic chemical questions** and is better than the base-sharded model.
Through some **chemical reactions**, it can clearly understand **the reactant and corresponding products**. **(For most, it is correct😅)**
**Just have fun testing it, and ask some interesting questions!!**
- **Developed by:** FelixChao
- **Shared by [optional]:** CleverShovel/vicuna-7b-v1.3-sharded-bf16
- **Finetuned from model [optional]:** CleverShovel/vicuna-7b-v1.3-sharded-bf16
## How to Get Started with the Model
Use the code below to get started with the model.
### For Pipeline
```python
# Use a pipeline as a high-level helper
from transformers import pipeline
pipe = pipeline("text-generation", model="FelixChao/vicuna-7B-chemical")
```
### For Model_Loading
```python
# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM
tokenizer = AutoTokenizer.from_pretrained("FelixChao/vicuna-7B-chemical")
model = AutoModelForCausalLM.from_pretrained("FelixChao/vicuna-7B-chemical")
```
## Training Details
### Training Data
<!-- This should link to a Data Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
[More Information Needed]
### Training Procedure
<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
#### Preprocessing [optional]
[More Information Needed]
#### Training Hyperparameters
- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
#### Speeds, Sizes, Times [optional]
<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
[More Information Needed]
## Evaluation
<!-- This section describes the evaluation protocols and provides the results. -->
### Testing Data, Factors & Metrics
#### Testing Data
<!-- This should link to a Data Card if possible. -->
[More Information Needed]
#### Factors
<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
[More Information Needed]
#### Metrics
<!-- These are the evaluation metrics being used, ideally with a description of why. -->
[More Information Needed]
### Results
[More Information Needed]
#### Summary
## Model Examination [optional]
<!-- Relevant interpretability work for the model goes here -->
[More Information Needed]
## Environmental Impact
<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
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).
- **Hardware Type:** [More Information Needed]
- **Hours used:** [More Information Needed]
- **Cloud Provider:** [More Information Needed]
- **Compute Region:** [More Information Needed]
- **Carbon Emitted:** [More Information Needed]
## Technical Specifications [optional]
### Model Architecture and Objective
[More Information Needed]
### Compute Infrastructure
[More Information Needed]
#### Hardware
[More Information Needed]
#### Software
[More Information Needed]
## Citation [optional]
<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
**BibTeX:**
[More Information Needed]
**APA:**
[More Information Needed]
## Glossary [optional]
<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
[More Information Needed]
## More Information [optional]
[More Information Needed]
## Model Card Authors [optional]
[More Information Needed]
## Model Card Contact
[More Information Needed]
| 4,654 | [
[
-0.0465087890625,
-0.04931640625,
0.03936767578125,
0.00434112548828125,
-0.0181121826171875,
-0.01995849609375,
0.0052642822265625,
-0.033477783203125,
0.0191497802734375,
0.04217529296875,
-0.047088623046875,
-0.045196533203125,
-0.040771484375,
-0.001544952392578125,
-0.0160369873046875,
0.09210205078125,
-0.00551605224609375,
0.0169830322265625,
-0.02020263671875,
0.0161895751953125,
-0.0266265869140625,
-0.04095458984375,
-0.0469970703125,
-0.034820556640625,
0.032928466796875,
0.023834228515625,
0.042938232421875,
0.0672607421875,
0.0477294921875,
0.027679443359375,
-0.0189208984375,
-0.00981903076171875,
-0.0247650146484375,
-0.01727294921875,
-0.02215576171875,
-0.022369384765625,
-0.061767578125,
0.00955963134765625,
0.039886474609375,
0.040771484375,
-0.00872039794921875,
0.0301361083984375,
-0.005275726318359375,
0.025146484375,
-0.046905517578125,
0.02532958984375,
-0.0419921875,
0.0016765594482421875,
-0.01129150390625,
-0.0008220672607421875,
-0.00711822509765625,
0.0010051727294921875,
-0.01363372802734375,
-0.052825927734375,
0.0243988037109375,
0.0160980224609375,
0.088623046875,
0.01904296875,
-0.0289459228515625,
-0.01174163818359375,
-0.0550537109375,
0.048248291015625,
-0.053955078125,
0.017852783203125,
0.026336669921875,
0.0298309326171875,
-0.0021991729736328125,
-0.0675048828125,
-0.045623779296875,
-0.0121917724609375,
-0.00406646728515625,
0.030853271484375,
-0.00894927978515625,
0.013641357421875,
0.03668212890625,
0.049072265625,
-0.040313720703125,
-0.0009446144104003906,
-0.0416259765625,
-0.0227508544921875,
0.061126708984375,
0.039764404296875,
0.0104217529296875,
-0.0262451171875,
-0.044097900390625,
-0.0241546630859375,
-0.01354217529296875,
0.0005588531494140625,
0.03173828125,
0.0240020751953125,
-0.038421630859375,
0.0498046875,
-0.01551055908203125,
0.03240966796875,
0.00431060791015625,
0.00994110107421875,
0.021697998046875,
-0.04681396484375,
-0.031646728515625,
-0.0103302001953125,
0.06365966796875,
0.022705078125,
-0.0175933837890625,
0.01372528076171875,
-0.021453857421875,
-0.0027599334716796875,
0.027008056640625,
-0.08074951171875,
-0.03460693359375,
0.03643798828125,
-0.038177490234375,
-0.03448486328125,
0.0113677978515625,
-0.0706787109375,
0.00362396240234375,
-0.03082275390625,
0.03887939453125,
-0.023681640625,
-0.0213165283203125,
-0.00478363037109375,
-0.033050537109375,
0.0171661376953125,
0.02056884765625,
-0.06005859375,
0.040618896484375,
0.0416259765625,
0.05218505859375,
-0.006839752197265625,
-0.0194091796875,
0.0017719268798828125,
0.0012655258178710938,
-0.0062103271484375,
0.055267333984375,
-0.024505615234375,
-0.03509521484375,
-0.0273895263671875,
0.0206298828125,
0.0025882720947265625,
-0.0269012451171875,
0.0443115234375,
-0.0285797119140625,
0.0206146240234375,
-0.0288848876953125,
-0.04425048828125,
-0.0211334228515625,
0.0175933837890625,
-0.06256103515625,
0.0888671875,
0.013153076171875,
-0.072265625,
0.0075836181640625,
-0.07061767578125,
-0.0152740478515625,
0.0124053955078125,
0.010284423828125,
-0.043792724609375,
-0.003787994384765625,
-0.01416015625,
0.03179931640625,
-0.03521728515625,
0.0136871337890625,
-0.03118896484375,
-0.0094146728515625,
-0.007106781005859375,
-0.0035247802734375,
0.08966064453125,
0.035797119140625,
-0.02056884765625,
0.014923095703125,
-0.0692138671875,
0.00907135009765625,
0.0264892578125,
-0.0239105224609375,
0.007495880126953125,
-0.034271240234375,
0.04644775390625,
0.0122222900390625,
0.02685546875,
-0.029296875,
0.0077056884765625,
0.0009250640869140625,
0.0296630859375,
0.043487548828125,
0.031341552734375,
0.0113983154296875,
-0.02783203125,
0.0478515625,
0.00063323974609375,
0.03643798828125,
0.004070281982421875,
-0.039825439453125,
-0.0631103515625,
-0.00708770751953125,
0.02215576171875,
0.047210693359375,
-0.0299072265625,
0.06304931640625,
0.004852294921875,
-0.0787353515625,
-0.01349639892578125,
0.00838470458984375,
0.03448486328125,
0.04437255859375,
0.03240966796875,
-0.028472900390625,
-0.0545654296875,
-0.068115234375,
0.0045928955078125,
-0.01120758056640625,
0.0151519775390625,
0.01470184326171875,
0.07647705078125,
-0.024017333984375,
0.05810546875,
-0.054412841796875,
-0.005275726318359375,
-0.0174713134765625,
0.0012416839599609375,
0.01129150390625,
0.049530029296875,
0.047088623046875,
-0.06280517578125,
-0.018310546875,
-0.019378662109375,
-0.044158935546875,
0.003917694091796875,
0.00604248046875,
-0.0262603759765625,
-0.017364501953125,
0.02630615234375,
-0.04766845703125,
0.0312347412109375,
0.032379150390625,
-0.021331787109375,
0.05462646484375,
-0.02081298828125,
0.0042572021484375,
-0.09088134765625,
0.03533935546875,
0.0030155181884765625,
-0.004253387451171875,
-0.034210205078125,
0.0171051025390625,
-0.003986358642578125,
-0.0283966064453125,
-0.055938720703125,
0.0576171875,
-0.0280303955078125,
0.0052490234375,
-0.0206451416015625,
-0.020477294921875,
0.017059326171875,
0.042144775390625,
0.017486572265625,
0.0374755859375,
0.0377197265625,
-0.056854248046875,
0.031585693359375,
0.018096923828125,
-0.004253387451171875,
0.034942626953125,
-0.06500244140625,
0.0113525390625,
-0.0029144287109375,
0.02349853515625,
-0.0638427734375,
-0.03179931640625,
0.024017333984375,
-0.0300750732421875,
0.032623291015625,
-0.0084075927734375,
-0.046783447265625,
-0.03375244140625,
-0.00153350830078125,
0.0287933349609375,
0.040374755859375,
-0.0178070068359375,
0.041534423828125,
0.035980224609375,
0.016357421875,
-0.011077880859375,
-0.042755126953125,
-0.0126800537109375,
-0.0276031494140625,
-0.0286102294921875,
0.037567138671875,
-0.02764892578125,
-0.00811767578125,
0.0005884170532226562,
0.01129913330078125,
-0.03173828125,
0.0146942138671875,
0.02557373046875,
0.014801025390625,
0.00429534912109375,
0.0005774497985839844,
-0.0081329345703125,
-0.020050048828125,
0.0209503173828125,
-0.00165557861328125,
0.03350830078125,
-0.01390838623046875,
0.0047149658203125,
-0.0523681640625,
0.0292816162109375,
0.041229248046875,
-0.014495849609375,
0.05078125,
0.05755615234375,
-0.053863525390625,
-0.0008859634399414062,
-0.0209503173828125,
-0.02105712890625,
-0.03436279296875,
0.0298309326171875,
-0.021331787109375,
-0.0135650634765625,
0.04754638671875,
0.005229949951171875,
-0.001354217529296875,
0.07763671875,
0.046600341796875,
-0.01082611083984375,
0.0653076171875,
0.05364990234375,
0.0126800537109375,
0.03533935546875,
-0.047088623046875,
-0.001720428466796875,
-0.06329345703125,
-0.030731201171875,
-0.059356689453125,
-0.0021572113037109375,
-0.035400390625,
-0.0212554931640625,
0.0160980224609375,
0.0120086669921875,
-0.04473876953125,
0.05279541015625,
-0.04425048828125,
0.0160980224609375,
0.035491943359375,
0.00969696044921875,
0.0004935264587402344,
-0.0148162841796875,
-0.007358551025390625,
0.003147125244140625,
-0.052154541015625,
-0.035308837890625,
0.07550048828125,
0.041259765625,
0.040679931640625,
-0.009185791015625,
0.056060791015625,
0.015716552734375,
0.021484375,
-0.044525146484375,
0.03546142578125,
0.008636474609375,
-0.07769775390625,
-0.0030803680419921875,
-0.023773193359375,
-0.06146240234375,
0.0085906982421875,
-0.031768798828125,
-0.04107666015625,
0.01445770263671875,
0.022003173828125,
-0.02886962890625,
0.02197265625,
-0.047943115234375,
0.08892822265625,
-0.04412841796875,
-0.0302276611328125,
-0.00611114501953125,
-0.043914794921875,
0.0283050537109375,
-0.0014801025390625,
0.019195556640625,
-0.002925872802734375,
-0.0017461776733398438,
0.06976318359375,
-0.062744140625,
0.061920166015625,
-0.0251312255859375,
0.0197296142578125,
0.0312042236328125,
-0.023193359375,
0.02288818359375,
-0.01025390625,
-0.013153076171875,
0.026885986328125,
0.018402099609375,
-0.04095458984375,
-0.033843994140625,
0.04302978515625,
-0.057159423828125,
-0.0208587646484375,
-0.04644775390625,
-0.03436279296875,
0.004535675048828125,
0.032928466796875,
0.0400390625,
0.009429931640625,
-0.0176239013671875,
0.005100250244140625,
0.0489501953125,
-0.0169830322265625,
0.015869140625,
0.0203094482421875,
-0.0123748779296875,
-0.0423583984375,
0.05535888671875,
0.00841522216796875,
0.019683837890625,
0.0294342041015625,
0.0257720947265625,
-0.035736083984375,
-0.037841796875,
-0.02886962890625,
0.0202789306640625,
-0.053955078125,
-0.014190673828125,
-0.0609130859375,
-0.02899169921875,
-0.041290283203125,
0.005123138427734375,
-0.039794921875,
-0.0171356201171875,
-0.039337158203125,
-0.0187835693359375,
0.035736083984375,
0.041961669921875,
-0.01396942138671875,
0.043487548828125,
-0.043121337890625,
0.0169219970703125,
0.0119171142578125,
0.019866943359375,
0.0019474029541015625,
-0.033233642578125,
-0.0226593017578125,
0.01241302490234375,
-0.04205322265625,
-0.07916259765625,
0.02392578125,
0.00186920166015625,
0.049285888671875,
0.027435302734375,
0.0032787322998046875,
0.0511474609375,
-0.01800537109375,
0.06488037109375,
0.015167236328125,
-0.06768798828125,
0.0546875,
-0.02789306640625,
0.00691986083984375,
0.055938720703125,
0.049041748046875,
-0.01256561279296875,
0.0015077590942382812,
-0.08050537109375,
-0.06268310546875,
0.034332275390625,
0.022308349609375,
0.003879547119140625,
0.0020618438720703125,
0.04461669921875,
-0.0129852294921875,
0.0209503173828125,
-0.060394287109375,
-0.034271240234375,
-0.0207672119140625,
-0.0030193328857421875,
-0.0006680488586425781,
-0.004302978515625,
-0.01064300537109375,
-0.046234130859375,
0.078125,
0.01434326171875,
0.03277587890625,
0.00954437255859375,
0.0083770751953125,
0.00783538818359375,
-0.00421905517578125,
0.0267486572265625,
0.0295562744140625,
-0.045867919921875,
-0.007266998291015625,
0.0139007568359375,
-0.046630859375,
-0.002925872802734375,
0.01136016845703125,
-0.011627197265625,
-0.004009246826171875,
0.0164947509765625,
0.06298828125,
0.00957489013671875,
-0.0292816162109375,
0.0281219482421875,
0.0030117034912109375,
-0.02117919921875,
-0.032470703125,
0.0198822021484375,
0.0082550048828125,
-0.0035800933837890625,
-0.0015621185302734375,
0.0035495758056640625,
0.023162841796875,
-0.05322265625,
0.0145721435546875,
0.026092529296875,
-0.03643798828125,
-0.00908660888671875,
0.07806396484375,
0.0192718505859375,
-0.027130126953125,
0.04412841796875,
-0.025970458984375,
-0.038482666015625,
0.0767822265625,
0.040771484375,
0.0628662109375,
-0.012237548828125,
-0.0017948150634765625,
0.0545654296875,
0.019561767578125,
0.01087188720703125,
0.0301513671875,
-0.010650634765625,
-0.035797119140625,
0.0108795166015625,
-0.048858642578125,
-0.03271484375,
0.0255584716796875,
-0.0657958984375,
0.04534912109375,
-0.062347412109375,
-0.0204010009765625,
0.021453857421875,
0.034515380859375,
-0.08294677734375,
0.047271728515625,
-0.002841949462890625,
0.07696533203125,
-0.08489990234375,
0.059417724609375,
0.058746337890625,
-0.0596923828125,
-0.0596923828125,
-0.0251922607421875,
0.00525665283203125,
-0.048004150390625,
0.0192718505859375,
0.0015544891357421875,
0.0116119384765625,
0.0010318756103515625,
-0.0531005859375,
-0.06378173828125,
0.1002197265625,
0.0070953369140625,
-0.042205810546875,
0.00899505615234375,
-0.0103302001953125,
0.04962158203125,
-0.034454345703125,
0.05133056640625,
0.030364990234375,
0.049072265625,
0.01548004150390625,
-0.051300048828125,
0.0171051025390625,
-0.037506103515625,
0.00824737548828125,
-0.007587432861328125,
-0.06256103515625,
0.0712890625,
-0.0132293701171875,
-0.00644683837890625,
0.0116729736328125,
0.040374755859375,
0.01800537109375,
0.0333251953125,
0.03228759765625,
0.053192138671875,
0.0633544921875,
0.00305938720703125,
0.09442138671875,
-0.034271240234375,
0.043914794921875,
0.09320068359375,
-0.00458526611328125,
0.060455322265625,
0.022125244140625,
-0.030548095703125,
0.03033447265625,
0.0760498046875,
-0.034332275390625,
0.024688720703125,
0.02471923828125,
0.001171112060546875,
-0.0204925537109375,
-0.01087188720703125,
-0.041107177734375,
0.02191162109375,
0.0187835693359375,
-0.042327880859375,
-0.0162200927734375,
-0.017913818359375,
0.005901336669921875,
-0.018951416015625,
-0.0181427001953125,
0.043060302734375,
-0.00341796875,
-0.040802001953125,
0.03271484375,
0.0251312255859375,
0.0284271240234375,
-0.054718017578125,
-0.013671875,
-0.0013074874877929688,
0.005329132080078125,
-0.0361328125,
-0.04443359375,
0.0255126953125,
-0.007434844970703125,
-0.033905029296875,
-0.00191497802734375,
0.037139892578125,
-0.0254974365234375,
-0.056854248046875,
0.024322509765625,
0.011749267578125,
0.03460693359375,
-0.01393890380859375,
-0.08624267578125,
0.0118408203125,
-0.0038890838623046875,
-0.0013599395751953125,
0.01239776611328125,
0.0017147064208984375,
0.0022907257080078125,
0.030120849609375,
0.0552978515625,
-0.003673553466796875,
-0.005947113037109375,
-0.005229949951171875,
0.06524658203125,
-0.05535888671875,
-0.0399169921875,
-0.03350830078125,
0.054412841796875,
-0.00980377197265625,
-0.039306640625,
0.051483154296875,
0.07196044921875,
0.0587158203125,
-0.00218963623046875,
0.06329345703125,
-0.00949859619140625,
0.0170745849609375,
-0.0296173095703125,
0.05072021484375,
-0.0445556640625,
0.0010509490966796875,
-0.030853271484375,
-0.0804443359375,
0.001636505126953125,
0.045196533203125,
-0.0110626220703125,
0.0192108154296875,
0.05352783203125,
0.0550537109375,
-0.016510009765625,
0.0132293701171875,
0.0006380081176757812,
0.023345947265625,
0.024444580078125,
0.03277587890625,
0.0369873046875,
-0.06201171875,
0.017578125,
-0.056976318359375,
-0.0224761962890625,
-0.00847625732421875,
-0.07672119140625,
-0.050445556640625,
-0.04962158203125,
-0.05072021484375,
-0.03875732421875,
-0.006534576416015625,
0.0552978515625,
0.06622314453125,
-0.054595947265625,
-0.026824951171875,
-0.0180816650390625,
-0.0009441375732421875,
-0.02020263671875,
-0.021209716796875,
0.031646728515625,
-0.005474090576171875,
-0.05841064453125,
0.0032634735107421875,
-0.014129638671875,
0.0263214111328125,
-0.0239105224609375,
-0.00717926025390625,
-0.015838623046875,
0.0010528564453125,
0.0203399658203125,
0.0306396484375,
-0.0345458984375,
-0.0103607177734375,
-0.0156402587890625,
-0.0053863525390625,
-0.00331878662109375,
0.048736572265625,
-0.03033447265625,
0.02862548828125,
0.04205322265625,
0.0247650146484375,
0.06634521484375,
0.00553131103515625,
0.03057861328125,
-0.023193359375,
0.0004119873046875,
0.0172882080078125,
0.036163330078125,
0.01480865478515625,
-0.04754638671875,
0.0340576171875,
0.035736083984375,
-0.041015625,
-0.04833984375,
-0.0013322830200195312,
-0.0989990234375,
-0.01434326171875,
0.08721923828125,
0.00038623809814453125,
-0.034210205078125,
-0.005908966064453125,
-0.0179595947265625,
0.022064208984375,
-0.018341064453125,
0.03680419921875,
0.0565185546875,
-0.01837158203125,
0.0039825439453125,
-0.035125732421875,
0.034881591796875,
0.0079193115234375,
-0.0841064453125,
-0.00997161865234375,
0.038848876953125,
0.04742431640625,
0.0124053955078125,
0.0364990234375,
-0.0169219970703125,
0.0147705078125,
0.0245361328125,
0.0282135009765625,
-0.01544189453125,
-0.032012939453125,
-0.0285797119140625,
-0.002166748046875,
-0.0177001953125,
-0.022796630859375
]
] |
TheBloke/CodeLlama-13B-Instruct-fp16 | 2023-08-25T11:13:46.000Z | [
"transformers",
"safetensors",
"llama",
"text-generation",
"llama-2",
"codellama",
"custom_code",
"license:llama2",
"endpoints_compatible",
"has_space",
"text-generation-inference",
"region:us"
] | text-generation | TheBloke | null | null | TheBloke/CodeLlama-13B-Instruct-fp16 | 28 | 5,900 | transformers | 2023-08-24T16:20:49 | ---
license: llama2
tags:
- llama-2
- codellama
---
<!-- header start -->
<!-- 200823 -->
<div style="width: auto; margin-left: auto; margin-right: auto">
<img src="https://i.imgur.com/EBdldam.jpg" alt="TheBlokeAI" style="width: 100%; min-width: 400px; display: block; margin: auto;">
</div>
<div style="display: flex; justify-content: space-between; width: 100%;">
<div style="display: flex; flex-direction: column; align-items: flex-start;">
<p style="margin-top: 0.5em; margin-bottom: 0em;"><a href="https://discord.gg/theblokeai">Chat & support: TheBloke's Discord server</a></p>
</div>
<div style="display: flex; flex-direction: column; align-items: flex-end;">
<p style="margin-top: 0.5em; margin-bottom: 0em;"><a href="https://www.patreon.com/TheBlokeAI">Want to contribute? TheBloke's Patreon page</a></p>
</div>
</div>
<div style="text-align:center; margin-top: 0em; margin-bottom: 0em"><p style="margin-top: 0.25em; margin-bottom: 0em;">TheBloke's LLM work is generously supported by a grant from <a href="https://a16z.com">andreessen horowitz (a16z)</a></p></div>
<hr style="margin-top: 1.0em; margin-bottom: 1.0em;">
<!-- header end -->
# CodeLlama 13B-Instruct fp16
- Model creator: [Meta](https://ai.meta.com/llama/)
## Description
This is Transformers/HF format fp16 weights for CodeLlama 13B-Instruct. It is the result of downloading CodeLlama 13B-Instruct from [Meta](https://ai.meta.com/blog/code-llama-large-language-model-coding/) and converting to HF using `convert_llama_weights_to_hf.py`.
Quantisations will be coming shortly.
Please note that due to a change in the RoPE Theta value, for correct results you must load these FP16 models with `trust_remote_code=True`
Credit to @emozilla for creating the necessary modelling code to achieve this!
## Prompt template: TBC
<!-- footer start -->
<!-- 200823 -->
## Discord
For further support, and discussions on these models and AI in general, join us at:
[TheBloke AI's Discord server](https://discord.gg/theblokeai)
## Thanks, and how to contribute.
Thanks to the [chirper.ai](https://chirper.ai) team!
I've had a lot of people ask if they can contribute. I enjoy providing models and helping people, and would love to be able to spend even more time doing it, as well as expanding into new projects like fine tuning/training.
If you're able and willing to contribute it will be most gratefully received and will help me to keep providing more models, and to start work on new AI projects.
Donaters will get priority support on any and all AI/LLM/model questions and requests, access to a private Discord room, plus other benefits.
* Patreon: https://patreon.com/TheBlokeAI
* Ko-Fi: https://ko-fi.com/TheBlokeAI
**Special thanks to**: Aemon Algiz.
**Patreon special mentions**: Sam, theTransient, Jonathan Leane, Steven Wood, webtim, Johann-Peter Hartmann, Geoffrey Montalvo, Gabriel Tamborski, Willem Michiel, John Villwock, Derek Yates, Mesiah Bishop, Eugene Pentland, Pieter, Chadd, Stephen Murray, Daniel P. Andersen, terasurfer, Brandon Frisco, Thomas Belote, Sid, Nathan LeClaire, Magnesian, Alps Aficionado, Stanislav Ovsiannikov, Alex, Joseph William Delisle, Nikolai Manek, Michael Davis, Junyu Yang, K, J, Spencer Kim, Stefan Sabev, Olusegun Samson, transmissions 11, Michael Levine, Cory Kujawski, Rainer Wilmers, zynix, Kalila, Luke @flexchar, Ajan Kanaga, Mandus, vamX, Ai Maven, Mano Prime, Matthew Berman, subjectnull, Vitor Caleffi, Clay Pascal, biorpg, alfie_i, 阿明, Jeffrey Morgan, ya boyyy, Raymond Fosdick, knownsqashed, Olakabola, Leonard Tan, ReadyPlayerEmma, Enrico Ros, Dave, Talal Aujan, Illia Dulskyi, Sean Connelly, senxiiz, Artur Olbinski, Elle, Raven Klaugh, Fen Risland, Deep Realms, Imad Khwaja, Fred von Graf, Will Dee, usrbinkat, SuperWojo, Alexandros Triantafyllidis, Swaroop Kallakuri, Dan Guido, John Detwiler, Pedro Madruga, Iucharbius, Viktor Bowallius, Asp the Wyvern, Edmond Seymore, Trenton Dambrowitz, Space Cruiser, Spiking Neurons AB, Pyrater, LangChain4j, Tony Hughes, Kacper Wikieł, Rishabh Srivastava, David Ziegler, Luke Pendergrass, Andrey, Gabriel Puliatti, Lone Striker, Sebastain Graf, Pierre Kircher, Randy H, NimbleBox.ai, Vadim, danny, Deo Leter
Thank you to all my generous patrons and donaters!
And thank you again to a16z for their generous grant.
<!-- footer end -->
# Original model card
# Code Llama
## **Model Details**
**Model Developers** Meta AI
**Variations** Code Llama comes in three model sizes, and three variants:
1) Code Llama: our base models designed for general code synthesis and understanding
2) Code Llama - Python: designed specifically for Python
3) Code Llama - Instruct: for instruction following and safer deployment
All variants are available in sizes of 7B, 13B and 34B parameters.
**Input** Models input text only.
**Output** Models output text only.
**Model Architecture** Code Llama and its variants are autoregressive language models using optimized transformer architectures. Code Llama 7B and 13B additionally support infilling text generation. All models were fine-tuned with up to 16K tokens, and support up to 100K tokens at inference time.
**Model Dates** Code Llama and its variants have been trained between January 2023 and July 2023.
**Status** This is a static model trained on an offline dataset. Future versions of Code Llama - Instruct will be released as we improve model safety with community feedback.
**Licence** A custom commercial license is available at: [https://ai.meta.com/resources/models-and-libraries/llama-downloads/](https://ai.meta.com/resources/models-and-libraries/llama-downloads/).
**Research Paper** More information can be found in the paper "[Code Llama: Open Foundation Models for Code](https://ai.meta.com/research/publications/code-llama-open-foundation-models-for-code/)".
**Where to send comments** Instructions on how to provide feedback or comments on the model can be found in the model [README](README.md), or by opening an issue in the GitHub repository ([https://github.com/facebookresearch/codellama/](https://github.com/facebookresearch/codellama/)).
## **Intended Use**
**Intended Use Cases** Code Llama and its variants is intended for commercial and research use in English and relevant programming languages. The base model Code Llama can be adapted for a variety of code synthesis and understanding tasks, Code Llama - Python is designed specifically to handle the Python programming language, and Code Llama - Instruct is intended to be safer to use for code assistant and generation applications.
**Out-of-Scope Uses** Use in any manner that violates applicable laws or regulations (including trade compliance laws). Use in languages other than English. Use in any other way that is prohibited by the Acceptable Use Policy and Licensing Agreement for Code Llama and its variants.
## **Hardware and Software**
**Training Factors**
We used custom training libraries. The training and fine-tuning of the released models have been performed Meta’s Research Super Cluster.
**Carbon Footprint** In aggregate, training all 9 Code Llama models required 400K GPU hours of computation on hardware of type A100-80GB (TDP of 350-400W). Estimated total emissions were 65.3 tCO2eq, 100% of which were offset by Meta’s sustainability program.
**Training data**
All experiments reported here and the released models have been trained and fine-tuned using the same data as Llama 2 with different weights (see Section 2 and Table 1 in the [research paper](https://ai.meta.com/research/publications/code-llama-open-foundation-models-for-code/) for details).
Code Llama - Instruct uses additional instruction fine-tuning data.
**Evaluation Results**
See evaluations for the main models and detailed ablations in Section 3 and safety evaluations in Section 4 of the research paper.
## **Ethical Considerations and Limitations**
Code Llama and its variants are a new technology that carries risks with use. Testing conducted to date has been in English, and has not covered, nor could it cover all scenarios. For these reasons, as with all LLMs, Code Llama’s potential outputs cannot be predicted in advance, and the model may in some instances produce inaccurate or objectionable responses to user prompts. Therefore, before deploying any applications of Code Llama, developers should perform safety testing and tuning tailored to their specific applications of the model.
Please see the Responsible Use Guide available available at [https://ai.meta.com/llama/responsible-user-guide](https://ai.meta.com/llama/responsible-user-guide).
| 8,618 | [
[
-0.03216552734375,
-0.03997802734375,
0.015655517578125,
0.01049041748046875,
-0.0158538818359375,
0.01016998291015625,
0.0006384849548339844,
-0.0531005859375,
0.0364990234375,
0.0176544189453125,
-0.052947998046875,
-0.0311431884765625,
-0.032806396484375,
0.00995635986328125,
-0.037994384765625,
0.0762939453125,
0.0126800537109375,
-0.0205078125,
-0.00513458251953125,
0.006343841552734375,
-0.0298309326171875,
-0.0255279541015625,
-0.0291290283203125,
-0.03936767578125,
0.03045654296875,
0.01605224609375,
0.057373046875,
0.0438232421875,
0.040069580078125,
0.0285797119140625,
-0.0178985595703125,
0.005931854248046875,
-0.03955078125,
-0.030731201171875,
0.004116058349609375,
-0.024017333984375,
-0.057952880859375,
-0.01047515869140625,
0.0226287841796875,
0.0228271484375,
-0.0144805908203125,
0.035186767578125,
-0.003429412841796875,
0.04229736328125,
-0.0296478271484375,
0.009124755859375,
-0.04290771484375,
0.00027370452880859375,
-0.0008120536804199219,
0.003765106201171875,
-0.004055023193359375,
-0.0222930908203125,
-0.0198211669921875,
-0.06640625,
-0.00833892822265625,
0.002765655517578125,
0.08758544921875,
0.02996826171875,
-0.0184326171875,
-0.001079559326171875,
-0.047332763671875,
0.05517578125,
-0.06768798828125,
0.019500732421875,
0.0231475830078125,
0.0088653564453125,
-0.001659393310546875,
-0.07147216796875,
-0.058807373046875,
-0.011138916015625,
-0.00395965576171875,
0.0200347900390625,
-0.044647216796875,
-0.0059814453125,
0.01068115234375,
0.032318115234375,
-0.0347900390625,
0.008880615234375,
-0.043701171875,
-0.006534576416015625,
0.06396484375,
0.004123687744140625,
0.02459716796875,
-0.0099334716796875,
-0.0258941650390625,
-0.014251708984375,
-0.05859375,
0.0117645263671875,
0.0303802490234375,
0.0009198188781738281,
-0.06439208984375,
0.0543212890625,
-0.00904083251953125,
0.032073974609375,
0.020965576171875,
-0.01873779296875,
0.031768798828125,
-0.04876708984375,
-0.0245208740234375,
-0.01332855224609375,
0.07025146484375,
0.035797119140625,
0.0017995834350585938,
0.01204681396484375,
-0.00958251953125,
0.00612640380859375,
0.016845703125,
-0.0592041015625,
-0.01192474365234375,
0.0302276611328125,
-0.054840087890625,
-0.036956787109375,
-0.0054779052734375,
-0.06414794921875,
-0.019439697265625,
-0.0060882568359375,
0.0250091552734375,
-0.01105499267578125,
-0.038787841796875,
0.0180816650390625,
-0.0027065277099609375,
0.03173828125,
0.0257720947265625,
-0.05499267578125,
0.0059356689453125,
0.037841796875,
0.054473876953125,
0.0196990966796875,
-0.018951416015625,
-0.01180267333984375,
0.01123809814453125,
-0.0209808349609375,
0.040069580078125,
-0.025543212890625,
-0.040924072265625,
-0.0117034912109375,
0.007511138916015625,
0.009033203125,
-0.0221710205078125,
0.02508544921875,
-0.0251007080078125,
-0.004245758056640625,
-0.0074920654296875,
-0.02154541015625,
-0.02471923828125,
0.0099334716796875,
-0.044097900390625,
0.0662841796875,
0.01776123046875,
-0.049285888671875,
0.0024890899658203125,
-0.054046630859375,
-0.019561767578125,
0.0009551048278808594,
-0.00669097900390625,
-0.04034423828125,
-0.006744384765625,
0.0162353515625,
0.02142333984375,
-0.035400390625,
0.0196990966796875,
-0.0237884521484375,
-0.0302276611328125,
0.0146484375,
-0.027130126953125,
0.074462890625,
0.0242919921875,
-0.046630859375,
0.00885009765625,
-0.062744140625,
-0.0160369873046875,
0.036346435546875,
-0.033294677734375,
0.0214385986328125,
-0.0014963150024414062,
0.0009121894836425781,
0.0027790069580078125,
0.035400390625,
-0.032806396484375,
0.02459716796875,
-0.02801513671875,
0.04119873046875,
0.0628662109375,
-0.0028743743896484375,
0.0303955078125,
-0.051025390625,
0.048919677734375,
-0.01227569580078125,
0.0283203125,
-0.0094757080078125,
-0.05078125,
-0.067138671875,
-0.0252685546875,
0.007022857666015625,
0.040496826171875,
-0.04058837890625,
0.05499267578125,
-0.01018524169921875,
-0.062744140625,
-0.041015625,
0.006954193115234375,
0.025360107421875,
0.026153564453125,
0.030853271484375,
-0.01263427734375,
-0.05419921875,
-0.058624267578125,
0.01224517822265625,
-0.0291900634765625,
-0.003170013427734375,
0.0292205810546875,
0.0587158203125,
-0.034942626953125,
0.055694580078125,
-0.03363037109375,
-0.0248565673828125,
-0.0240936279296875,
-0.0244598388671875,
0.037750244140625,
0.057525634765625,
0.051605224609375,
-0.05535888671875,
-0.023773193359375,
0.0117034912109375,
-0.056671142578125,
0.0009765625,
-0.01593017578125,
-0.010040283203125,
0.0117645263671875,
0.020660400390625,
-0.0665283203125,
0.052734375,
0.05230712890625,
-0.0255584716796875,
0.03790283203125,
-0.0128936767578125,
0.0013151168823242188,
-0.0782470703125,
0.02001953125,
-0.00208282470703125,
0.002361297607421875,
-0.0430908203125,
0.006656646728515625,
-0.0193634033203125,
-0.00664520263671875,
-0.041748046875,
0.040863037109375,
-0.030303955078125,
0.00022327899932861328,
-0.003997802734375,
-0.0086517333984375,
0.0035686492919921875,
0.04681396484375,
-0.017333984375,
0.058868408203125,
0.038848876953125,
-0.0401611328125,
0.02752685546875,
0.03546142578125,
-0.03143310546875,
0.017242431640625,
-0.0814208984375,
0.015869140625,
0.00585174560546875,
0.0355224609375,
-0.0765380859375,
-0.0142364501953125,
0.037872314453125,
-0.055389404296875,
0.016571044921875,
-0.004932403564453125,
-0.0282745361328125,
-0.039703369140625,
-0.0290985107421875,
0.036865234375,
0.057525634765625,
-0.03912353515625,
0.043212890625,
0.032562255859375,
0.01528167724609375,
-0.05517578125,
-0.05859375,
-0.004695892333984375,
-0.025177001953125,
-0.046875,
0.0362548828125,
-0.0237884521484375,
-0.0199127197265625,
-0.00469970703125,
-0.00279998779296875,
0.002445220947265625,
0.01953125,
0.032012939453125,
0.02679443359375,
-0.011993408203125,
-0.022064208984375,
-0.002483367919921875,
-0.0013179779052734375,
-0.0066986083984375,
-0.0152740478515625,
0.060516357421875,
-0.027496337890625,
-0.0260772705078125,
-0.06427001953125,
0.0104827880859375,
0.044586181640625,
-0.0200347900390625,
0.053619384765625,
0.034088134765625,
-0.033172607421875,
0.00675201416015625,
-0.037384033203125,
-0.0107574462890625,
-0.04559326171875,
0.0192413330078125,
-0.00995635986328125,
-0.05413818359375,
0.042236328125,
0.0183868408203125,
0.019989013671875,
0.040130615234375,
0.0491943359375,
-0.01525115966796875,
0.061187744140625,
0.05828857421875,
-0.0196533203125,
0.04083251953125,
-0.06500244140625,
0.01288604736328125,
-0.051788330078125,
-0.0347900390625,
-0.0462646484375,
-0.03955078125,
-0.050872802734375,
-0.042510986328125,
0.030914306640625,
0.0032634735107421875,
-0.0478515625,
0.037200927734375,
-0.046539306640625,
0.023468017578125,
0.038787841796875,
0.01290130615234375,
0.01399993896484375,
0.002349853515625,
0.00736236572265625,
0.0168914794921875,
-0.051361083984375,
-0.036651611328125,
0.0821533203125,
0.0286712646484375,
0.0516357421875,
0.004619598388671875,
0.05804443359375,
0.01558685302734375,
0.0170745849609375,
-0.04046630859375,
0.04083251953125,
0.004344940185546875,
-0.060516357421875,
-0.01473236083984375,
-0.0132293701171875,
-0.075439453125,
0.00839996337890625,
-0.0188751220703125,
-0.053955078125,
0.0299530029296875,
0.00004744529724121094,
-0.0311431884765625,
0.037994384765625,
-0.0234832763671875,
0.048004150390625,
-0.018218994140625,
-0.020538330078125,
-0.0179290771484375,
-0.053863525390625,
0.0245513916015625,
0.003063201904296875,
0.0298614501953125,
-0.01023101806640625,
-0.012939453125,
0.05023193359375,
-0.045928955078125,
0.0833740234375,
0.00298309326171875,
-0.0189056396484375,
0.048919677734375,
-0.0018033981323242188,
0.03826904296875,
0.0119171142578125,
-0.0149078369140625,
0.045318603515625,
-0.0123138427734375,
-0.01450347900390625,
-0.006786346435546875,
0.039398193359375,
-0.0863037109375,
-0.04669189453125,
-0.021453857421875,
-0.034027099609375,
0.02801513671875,
0.0228424072265625,
0.030487060546875,
0.0180816650390625,
0.005550384521484375,
0.036834716796875,
0.0232696533203125,
-0.035308837890625,
0.0491943359375,
0.01947021484375,
-0.0055694580078125,
-0.044647216796875,
0.06781005859375,
0.0010995864868164062,
0.01242828369140625,
0.0293121337890625,
0.01348876953125,
-0.01506805419921875,
-0.0309295654296875,
-0.030364990234375,
0.039276123046875,
-0.04400634765625,
-0.041961669921875,
-0.03424072265625,
-0.01464080810546875,
-0.0306396484375,
-0.0294342041015625,
-0.03448486328125,
-0.0301666259765625,
-0.055755615234375,
-0.00925445556640625,
0.044403076171875,
0.0535888671875,
-0.01242828369140625,
0.03436279296875,
-0.046112060546875,
0.0243682861328125,
0.004520416259765625,
0.0125732421875,
0.0098876953125,
-0.05029296875,
-0.013153076171875,
0.0165557861328125,
-0.043243408203125,
-0.049041748046875,
0.041534423828125,
0.008056640625,
0.044586181640625,
0.0188751220703125,
0.004253387451171875,
0.0584716796875,
-0.031524658203125,
0.08013916015625,
0.037445068359375,
-0.07537841796875,
0.043304443359375,
-0.0335693359375,
0.01511383056640625,
0.0263671875,
0.032470703125,
-0.018646240234375,
-0.0296478271484375,
-0.060791015625,
-0.060150146484375,
0.0494384765625,
0.0189056396484375,
0.0177154541015625,
0.00995635986328125,
0.0299530029296875,
-0.01056671142578125,
0.0197296142578125,
-0.08740234375,
-0.03277587890625,
-0.0261383056640625,
-0.00914764404296875,
-0.0024871826171875,
0.0010976791381835938,
-0.0162353515625,
-0.03289794921875,
0.05859375,
-0.01348114013671875,
0.0469970703125,
0.0149078369140625,
0.007740020751953125,
-0.0244598388671875,
0.0004341602325439453,
0.0516357421875,
0.062469482421875,
-0.0016880035400390625,
-0.014678955078125,
0.0145111083984375,
-0.0310821533203125,
0.0102691650390625,
-0.0031719207763671875,
-0.0236968994140625,
-0.0198974609375,
0.0312347412109375,
0.053955078125,
0.005828857421875,
-0.044921875,
0.036773681640625,
0.01018524169921875,
-0.028961181640625,
-0.03369140625,
0.01450347900390625,
0.02471923828125,
0.042755126953125,
0.032257080078125,
0.0033397674560546875,
-0.0003979206085205078,
-0.030548095703125,
0.00238800048828125,
0.034942626953125,
-0.00215911865234375,
-0.03106689453125,
0.08447265625,
0.01410675048828125,
-0.036163330078125,
0.03863525390625,
0.01293182373046875,
-0.03289794921875,
0.087158203125,
0.048431396484375,
0.061126708984375,
-0.004467010498046875,
0.01248931884765625,
0.04217529296875,
0.03826904296875,
0.004467010498046875,
0.024169921875,
0.0010967254638671875,
-0.03851318359375,
-0.01444244384765625,
-0.046722412109375,
-0.0274810791015625,
0.01525115966796875,
-0.038848876953125,
0.0380859375,
-0.05706787109375,
-0.0081787109375,
-0.02227783203125,
0.007488250732421875,
-0.042510986328125,
0.009521484375,
0.0151519775390625,
0.07147216796875,
-0.044342041015625,
0.060577392578125,
0.03912353515625,
-0.04974365234375,
-0.07122802734375,
-0.0109405517578125,
-0.0009641647338867188,
-0.06976318359375,
0.0362548828125,
0.0152130126953125,
-0.00630950927734375,
0.01288604736328125,
-0.06982421875,
-0.07708740234375,
0.11602783203125,
0.0226593017578125,
-0.045440673828125,
-0.0019073486328125,
0.007904052734375,
0.0362548828125,
-0.02728271484375,
0.0293121337890625,
0.036956787109375,
0.038604736328125,
0.0009551048278808594,
-0.07757568359375,
0.0096282958984375,
-0.031982421875,
0.0025348663330078125,
-0.01073455810546875,
-0.089599609375,
0.0660400390625,
-0.0282745361328125,
-0.0021266937255859375,
0.02850341796875,
0.05755615234375,
0.041412353515625,
0.0227508544921875,
0.03643798828125,
0.031951904296875,
0.04791259765625,
0.0010395050048828125,
0.0845947265625,
-0.0452880859375,
0.033966064453125,
0.04559326171875,
-0.00620269775390625,
0.05438232421875,
0.025726318359375,
-0.037017822265625,
0.039520263671875,
0.046600341796875,
-0.0196533203125,
0.0302276611328125,
0.01558685302734375,
-0.01001739501953125,
-0.00394439697265625,
-0.015777587890625,
-0.061798095703125,
0.024200439453125,
0.02020263671875,
-0.0183258056640625,
0.011932373046875,
-0.01242828369140625,
0.004917144775390625,
-0.0147705078125,
-0.017181396484375,
0.03955078125,
0.017608642578125,
-0.0281524658203125,
0.083740234375,
-0.0035190582275390625,
0.07122802734375,
-0.05694580078125,
-0.005916595458984375,
-0.036956787109375,
0.01529693603515625,
-0.032318115234375,
-0.040374755859375,
0.002269744873046875,
0.005924224853515625,
-0.007389068603515625,
-0.0113983154296875,
0.041717529296875,
-0.002513885498046875,
-0.046661376953125,
0.036041259765625,
0.008636474609375,
0.0128021240234375,
0.0281829833984375,
-0.06378173828125,
0.03448486328125,
0.01800537109375,
-0.0345458984375,
0.0185394287109375,
0.01180267333984375,
0.016571044921875,
0.0625,
0.051910400390625,
-0.0084381103515625,
0.01346588134765625,
-0.0171356201171875,
0.0816650390625,
-0.035888671875,
-0.0298309326171875,
-0.0628662109375,
0.057586669921875,
0.0178680419921875,
-0.025604248046875,
0.05340576171875,
0.035125732421875,
0.06793212890625,
-0.01055908203125,
0.054901123046875,
-0.0259552001953125,
0.009033203125,
-0.0240325927734375,
0.0640869140625,
-0.07403564453125,
0.0285186767578125,
-0.0333251953125,
-0.05999755859375,
-0.0164642333984375,
0.06396484375,
0.0149078369140625,
0.013702392578125,
0.026336669921875,
0.07012939453125,
0.01171875,
-0.00518035888671875,
0.0196380615234375,
0.0243682861328125,
0.040069580078125,
0.05853271484375,
0.06451416015625,
-0.05548095703125,
0.0594482421875,
-0.043304443359375,
-0.008056640625,
-0.0236053466796875,
-0.06884765625,
-0.05670166015625,
-0.03033447265625,
-0.0298309326171875,
-0.0308837890625,
-0.01256561279296875,
0.07061767578125,
0.050262451171875,
-0.044769287109375,
-0.045867919921875,
-0.00911712646484375,
0.017242431640625,
-0.0189056396484375,
-0.01251983642578125,
0.0169525146484375,
0.0170440673828125,
-0.059661865234375,
0.041412353515625,
-0.0012302398681640625,
0.023529052734375,
-0.0142669677734375,
-0.023834228515625,
-0.0361328125,
0.0030574798583984375,
0.028289794921875,
0.031646728515625,
-0.053466796875,
-0.017303466796875,
0.004390716552734375,
0.004119873046875,
0.0160675048828125,
0.0303802490234375,
-0.0494384765625,
-0.002155303955078125,
0.03900146484375,
0.038665771484375,
0.03973388671875,
-0.0030689239501953125,
0.0178070068359375,
-0.02728271484375,
0.0240325927734375,
0.006862640380859375,
0.035614013671875,
0.0023288726806640625,
-0.042236328125,
0.0587158203125,
0.0217742919921875,
-0.05242919921875,
-0.076171875,
-0.005435943603515625,
-0.08404541015625,
-0.0225677490234375,
0.087158203125,
-0.0010099411010742188,
-0.0222320556640625,
0.01290130615234375,
-0.01959228515625,
0.02606201171875,
-0.0302886962890625,
0.039276123046875,
0.022491455078125,
-0.00666046142578125,
-0.0084381103515625,
-0.045318603515625,
0.01495361328125,
0.019561767578125,
-0.0654296875,
-0.010467529296875,
0.04144287109375,
0.022491455078125,
0.029388427734375,
0.05908203125,
-0.00862884521484375,
0.0301971435546875,
0.006191253662109375,
0.0214691162109375,
-0.00995635986328125,
-0.026824951171875,
-0.0288848876953125,
0.0027065277099609375,
-0.016632080078125,
-0.014312744140625
]
] |
Danielbrdz/CodeBarcenas-7b | 2023-09-03T22:50:29.000Z | [
"transformers",
"pytorch",
"llama",
"text-generation",
"en",
"license:llama2",
"endpoints_compatible",
"has_space",
"text-generation-inference",
"region:us"
] | text-generation | Danielbrdz | null | null | Danielbrdz/CodeBarcenas-7b | 0 | 5,899 | transformers | 2023-09-03T22:10:59 | ---
license: llama2
language:
- en
---
CodeBarcenas
Model specialized in the Python language
Based on the model: WizardLM/WizardCoder-Python-7B-V1.0
And trained with the dataset: mlabonne/Evol-Instruct-Python-26k
Made with ❤️ in Guadalupe, Nuevo Leon, Mexico 🇲🇽 | 262 | [
[
-0.020538330078125,
-0.035186767578125,
-0.008392333984375,
0.032623291015625,
0.0021648406982421875,
-0.0018758773803710938,
0.0202789306640625,
-0.0139617919921875,
0.01328277587890625,
0.059478759765625,
-0.025726318359375,
-0.04107666015625,
-0.003795623779296875,
0.0200347900390625,
-0.0282440185546875,
0.06964111328125,
0.0103912353515625,
0.04400634765625,
0.023590087890625,
-0.008148193359375,
-0.049072265625,
-0.04534912109375,
-0.0439453125,
-0.05670166015625,
0.04718017578125,
0.00927734375,
0.05157470703125,
0.023834228515625,
0.01352691650390625,
0.01490020751953125,
-0.004894256591796875,
-0.00318145751953125,
-0.032501220703125,
-0.017791748046875,
-0.029327392578125,
-0.04913330078125,
-0.037628173828125,
-0.0282440185546875,
-0.00887298583984375,
0.03839111328125,
-0.021484375,
0.0121002197265625,
-0.026458740234375,
0.04791259765625,
-0.026519775390625,
0.03240966796875,
-0.040130615234375,
0.00537109375,
-0.0053558349609375,
0.0023479461669921875,
-0.01062774658203125,
-0.0428466796875,
0.0095977783203125,
-0.038238525390625,
0.03521728515625,
0.00420379638671875,
0.045135498046875,
0.01284027099609375,
-0.052001953125,
-0.04193115234375,
-0.032135009765625,
0.01238250732421875,
-0.03839111328125,
0.0186004638671875,
0.0439453125,
0.032867431640625,
-0.033172607421875,
-0.045684814453125,
-0.01190948486328125,
-0.0001811981201171875,
0.00738525390625,
-0.01041412353515625,
-0.00817108154296875,
0.0020236968994140625,
0.01678466796875,
0.001842498779296875,
-0.052825927734375,
0.00787353515625,
-0.07025146484375,
-0.0198516845703125,
0.0421142578125,
0.0479736328125,
0.0064239501953125,
0.017364501953125,
0.0052490234375,
0.01837158203125,
-0.05517578125,
0.0093841552734375,
0.033721923828125,
0.00926971435546875,
0.010528564453125,
0.04925537109375,
-0.0265045166015625,
0.079345703125,
-0.020355224609375,
0.00036597251892089844,
0.030670166015625,
0.003246307373046875,
-0.0509033203125,
0.0175933837890625,
0.034942626953125,
0.01262664794921875,
0.047515869140625,
-0.014007568359375,
-0.03497314453125,
0.0030956268310546875,
0.0213623046875,
-0.05438232421875,
-0.05157470703125,
0.01456451416015625,
-0.039520263671875,
-0.0282440185546875,
0.011199951171875,
-0.030181884765625,
-0.031341552734375,
-0.01264190673828125,
0.02252197265625,
-0.0364990234375,
-0.0335693359375,
0.0247802734375,
-0.005970001220703125,
0.0014162063598632812,
0.0273895263671875,
-0.06646728515625,
0.0282745361328125,
0.03131103515625,
0.054290771484375,
0.00336456298828125,
-0.027862548828125,
-0.004512786865234375,
-0.0053558349609375,
-0.05902099609375,
0.05438232421875,
-0.0161590576171875,
-0.015106201171875,
0.01519012451171875,
0.01959228515625,
-0.005084991455078125,
-0.037841796875,
0.011444091796875,
-0.07244873046875,
0.0020961761474609375,
0.01446533203125,
-0.0467529296875,
-0.045257568359375,
0.016998291015625,
-0.07586669921875,
0.054168701171875,
0.031585693359375,
-0.053619384765625,
0.0310516357421875,
-0.034210205078125,
-0.01126861572265625,
0.0103759765625,
0.0025920867919921875,
-0.016998291015625,
-0.0011272430419921875,
-0.0023250579833984375,
0.01146697998046875,
-0.01324462890625,
0.02764892578125,
-0.004917144775390625,
-0.0283966064453125,
0.030120849609375,
-0.03192138671875,
0.0953369140625,
0.042724609375,
0.01033782958984375,
0.0146942138671875,
-0.0982666015625,
-0.00445556640625,
0.0006418228149414062,
-0.04119873046875,
-0.02557373046875,
-0.018402099609375,
0.037384033203125,
0.0084228515625,
0.035614013671875,
-0.0242919921875,
0.034515380859375,
-0.046844482421875,
0.0127410888671875,
0.018310546875,
-0.005954742431640625,
0.01554107666015625,
-0.016204833984375,
0.054962158203125,
-0.0175018310546875,
0.0003952980041503906,
-0.0123291015625,
-0.031341552734375,
-0.04461669921875,
-0.0247650146484375,
0.021484375,
0.056243896484375,
-0.0445556640625,
0.032745361328125,
0.0025157928466796875,
-0.053558349609375,
0.0088958740234375,
-0.0032062530517578125,
0.01507568359375,
0.0016393661499023438,
0.0088043212890625,
0.01129150390625,
-0.05029296875,
-0.059844970703125,
0.008636474609375,
-0.00481414794921875,
-0.0115509033203125,
-0.0208587646484375,
0.0596923828125,
-0.0108642578125,
0.07598876953125,
-0.03021240234375,
-0.01381683349609375,
-0.03106689453125,
-0.0192718505859375,
0.07171630859375,
0.04364013671875,
0.05535888671875,
-0.03851318359375,
-0.044677734375,
0.00537872314453125,
-0.051513671875,
-0.01192474365234375,
0.00937652587890625,
-0.005184173583984375,
0.0093231201171875,
0.0259552001953125,
-0.032470703125,
0.07281494140625,
0.0287017822265625,
-0.0445556640625,
0.055389404296875,
-0.03857421875,
0.00193023681640625,
-0.08026123046875,
-0.0111541748046875,
-0.016326904296875,
-0.0167236328125,
-0.05706787109375,
-0.017364501953125,
0.02801513671875,
-0.00418853759765625,
-0.0283966064453125,
0.032867431640625,
-0.03765869140625,
0.007904052734375,
-0.0278167724609375,
-0.049163818359375,
-0.007633209228515625,
0.05023193359375,
0.0227508544921875,
0.04718017578125,
0.06671142578125,
-0.0341796875,
0.07452392578125,
0.021728515625,
-0.0421142578125,
0.025665283203125,
-0.06341552734375,
-0.0059661865234375,
0.02154541015625,
-0.00402069091796875,
-0.04193115234375,
-0.02020263671875,
0.0198822021484375,
-0.020294189453125,
0.00759124755859375,
-0.0263671875,
-0.048980712890625,
-0.051910400390625,
0.0052490234375,
0.01419830322265625,
0.0195465087890625,
-0.046783447265625,
0.0227813720703125,
0.01105499267578125,
0.0089874267578125,
-0.046600341796875,
-0.045745849609375,
0.00888824462890625,
-0.0157318115234375,
-0.044281005859375,
-0.00788116455078125,
0.00902557373046875,
-0.02447509765625,
-0.023651123046875,
-0.0017480850219726562,
-0.045684814453125,
-0.011199951171875,
0.0276947021484375,
0.027099609375,
-0.018707275390625,
0.0208282470703125,
-0.0004229545593261719,
0.01036834716796875,
0.0022640228271484375,
-0.0158233642578125,
0.06768798828125,
-0.01275634765625,
-0.00568389892578125,
-0.021759033203125,
0.004093170166015625,
0.038818359375,
-0.009185791015625,
0.0771484375,
0.00592041015625,
-0.0234375,
-0.0335693359375,
-0.0182037353515625,
0.0171966552734375,
-0.035186767578125,
0.0416259765625,
-0.048004150390625,
-0.036956787109375,
0.06524658203125,
0.010284423828125,
-0.0205841064453125,
0.00960540771484375,
0.0679931640625,
0.0287628173828125,
0.0582275390625,
0.043701171875,
-0.00399017333984375,
0.0384521484375,
-0.0243377685546875,
-0.0107269287109375,
-0.0190887451171875,
-0.048065185546875,
-0.047576904296875,
0.0277557373046875,
-0.0234222412109375,
-0.00972747802734375,
0.0022945404052734375,
0.0189971923828125,
-0.059967041015625,
0.05712890625,
-0.0440673828125,
0.0275115966796875,
0.043914794921875,
0.00937652587890625,
0.01325225830078125,
0.00396728515625,
-0.0078277587890625,
0.025787353515625,
-0.07037353515625,
-0.038055419921875,
0.0819091796875,
0.016387939453125,
0.1109619140625,
-0.0032253265380859375,
0.024627685546875,
0.0287628173828125,
-0.0034008026123046875,
-0.031280517578125,
0.014404296875,
0.0245361328125,
-0.07183837890625,
-0.0032672882080078125,
-0.01165771484375,
-0.0997314453125,
0.0245361328125,
0.00183868408203125,
-0.049407958984375,
0.022674560546875,
-0.0012140274047851562,
-0.00439453125,
0.0191497802734375,
-0.05072021484375,
0.0672607421875,
-0.011474609375,
0.00921630859375,
-0.01149749755859375,
-0.0190277099609375,
0.0311279296875,
-0.006862640380859375,
0.0099945068359375,
-0.00566864013671875,
0.012939453125,
0.058868408203125,
-0.06396484375,
0.03668212890625,
-0.0032901763916015625,
-0.010894775390625,
0.00785064697265625,
0.0335693359375,
0.0206298828125,
0.01059722900390625,
-0.0102691650390625,
0.0194549560546875,
0.024200439453125,
-0.022430419921875,
-0.0228118896484375,
0.03973388671875,
-0.0596923828125,
-0.016815185546875,
-0.0467529296875,
-0.04534912109375,
0.005809783935546875,
0.0259246826171875,
0.02398681640625,
0.0670166015625,
-0.032012939453125,
-0.0026111602783203125,
0.0438232421875,
-0.022369384765625,
0.0341796875,
0.0643310546875,
-0.0517578125,
-0.052825927734375,
0.060882568359375,
0.02545166015625,
-0.0253143310546875,
0.01145172119140625,
-0.00621795654296875,
-0.0032482147216796875,
-0.042755126953125,
-0.02801513671875,
0.0018262863159179688,
-0.061676025390625,
-0.040985107421875,
-0.035369873046875,
-0.03570556640625,
-0.02764892578125,
-0.01509857177734375,
-0.0272674560546875,
-0.0311279296875,
-0.0260467529296875,
-0.006870269775390625,
0.049102783203125,
0.08099365234375,
0.006839752197265625,
0.028350830078125,
-0.042633056640625,
0.01184844970703125,
0.017425537109375,
0.032928466796875,
-0.004009246826171875,
-0.031097412109375,
-0.065185546875,
-0.0179443359375,
0.00753021240234375,
-0.08258056640625,
0.06707763671875,
-0.0031261444091796875,
0.0594482421875,
0.0265350341796875,
-0.01065826416015625,
-0.0001291036605834961,
-0.031890869140625,
0.0479736328125,
0.02886962890625,
-0.0478515625,
0.048187255859375,
-0.0207061767578125,
0.0216217041015625,
0.016510009765625,
0.012786865234375,
-0.0203399658203125,
-0.0261383056640625,
-0.018341064453125,
-0.032012939453125,
0.0633544921875,
0.028900146484375,
-0.00046324729919433594,
0.0033168792724609375,
0.0164031982421875,
0.034423828125,
0.0258026123046875,
-0.05450439453125,
-0.0259552001953125,
-0.060028076171875,
-0.031402587890625,
0.028106689453125,
0.00492095947265625,
-0.0003178119659423828,
-0.03265380859375,
0.03851318359375,
-0.004360198974609375,
0.006092071533203125,
-0.0005640983581542969,
-0.0270538330078125,
0.0250701904296875,
-0.0087890625,
0.0462646484375,
0.078369140625,
-0.0229644775390625,
0.004833221435546875,
0.001880645751953125,
-0.03936767578125,
0.02581787109375,
-0.004608154296875,
-0.00879669189453125,
-0.0057525634765625,
0.0399169921875,
0.058380126953125,
-0.0347900390625,
-0.045989990234375,
0.0135345458984375,
0.0007157325744628906,
-0.0008907318115234375,
-0.047607421875,
0.0404052734375,
-0.00024402141571044922,
0.030029296875,
0.04345703125,
0.0216064453125,
0.01415252685546875,
-0.00443267822265625,
-0.0014276504516601562,
0.01322174072265625,
-0.038116455078125,
-0.01461029052734375,
0.0570068359375,
0.002170562744140625,
-0.051300048828125,
0.0399169921875,
0.0024356842041015625,
-0.048736572265625,
0.0791015625,
0.053802490234375,
0.042572021484375,
-0.0028858184814453125,
0.0058135986328125,
0.04547119140625,
0.031890869140625,
-0.01026153564453125,
0.028533935546875,
0.00572967529296875,
-0.06268310546875,
0.004985809326171875,
-0.029388427734375,
-0.0157318115234375,
-0.00986480712890625,
-0.045074462890625,
0.03826904296875,
-0.032928466796875,
-0.00691986083984375,
-0.0200653076171875,
0.004863739013671875,
-0.05963134765625,
0.020721435546875,
0.0010995864868164062,
0.1083984375,
-0.041290283203125,
0.11944580078125,
0.04119873046875,
-0.05816650390625,
-0.039337158203125,
-0.046356201171875,
-0.0272216796875,
-0.0665283203125,
0.0819091796875,
0.003986358642578125,
-0.00620269775390625,
0.005947113037109375,
-0.05511474609375,
-0.0560302734375,
0.07586669921875,
0.01194000244140625,
-0.0423583984375,
0.002040863037109375,
-0.0031299591064453125,
0.048736572265625,
-0.044921875,
0.0289306640625,
0.042633056640625,
0.006938934326171875,
-0.0195465087890625,
-0.0843505859375,
-0.0279388427734375,
-0.038726806640625,
0.0282440185546875,
-0.0213623046875,
-0.037872314453125,
0.08209228515625,
0.00786590576171875,
0.0268402099609375,
0.0276031494140625,
0.036865234375,
0.01514434814453125,
0.01079559326171875,
0.01203155517578125,
0.02862548828125,
0.05718994140625,
0.0022792816162109375,
0.056488037109375,
-0.030303955078125,
0.0477294921875,
0.07293701171875,
-0.0086822509765625,
0.01055908203125,
-0.002994537353515625,
-0.033172607421875,
0.053802490234375,
0.0743408203125,
-0.0350341796875,
0.052459716796875,
0.0226898193359375,
-0.0118255615234375,
0.00543975830078125,
0.025421142578125,
-0.037261962890625,
0.0163726806640625,
0.04022216796875,
-0.0199737548828125,
-0.006008148193359375,
0.030517578125,
-0.0012874603271484375,
-0.01213836669921875,
-0.055633544921875,
0.05328369140625,
-0.0181121826171875,
-0.0306854248046875,
0.058380126953125,
0.008056640625,
0.05438232421875,
-0.05792236328125,
0.0110015869140625,
-0.0291748046875,
0.01279449462890625,
-0.0085296630859375,
-0.056396484375,
0.00745391845703125,
-0.003978729248046875,
-0.0197601318359375,
0.03204345703125,
0.051361083984375,
-0.037384033203125,
-0.061920166015625,
0.023590087890625,
0.006866455078125,
0.0166473388671875,
-0.01114654541015625,
-0.0482177734375,
-0.0025959014892578125,
-0.01189422607421875,
-0.030120849609375,
-0.00933074951171875,
0.03924560546875,
-0.012451171875,
0.059295654296875,
0.018310546875,
0.01145172119140625,
0.0350341796875,
0.01558685302734375,
0.06109619140625,
-0.053131103515625,
-0.053802490234375,
-0.04315185546875,
0.00434112548828125,
-0.00566864013671875,
-0.0297393798828125,
0.06378173828125,
0.06756591796875,
0.04388427734375,
-0.03948974609375,
0.0638427734375,
-0.0235443115234375,
0.0105743408203125,
-0.0199127197265625,
0.061737060546875,
-0.0115509033203125,
-0.0035152435302734375,
0.002254486083984375,
-0.07244873046875,
-0.0357666015625,
0.05328369140625,
-0.0006818771362304688,
-0.0163421630859375,
0.0406494140625,
0.091796875,
0.00986480712890625,
-0.02056884765625,
0.04937744140625,
0.006313323974609375,
0.0220184326171875,
0.045501708984375,
0.05865478515625,
-0.0296630859375,
0.043914794921875,
-0.0168304443359375,
-0.01453399658203125,
-0.01309967041015625,
-0.02398681640625,
-0.099853515625,
-0.0294036865234375,
-0.0077056884765625,
-0.05792236328125,
-0.009185791015625,
0.0885009765625,
0.047393798828125,
-0.090576171875,
-0.041748046875,
-0.037109375,
0.0252227783203125,
-0.01413726806640625,
-0.015350341796875,
0.043853759765625,
-0.02099609375,
-0.051025390625,
0.023956298828125,
0.006725311279296875,
0.0008492469787597656,
-0.04241943359375,
-0.021697998046875,
0.0016307830810546875,
0.025543212890625,
0.0253143310546875,
0.0178070068359375,
-0.043121337890625,
0.007083892822265625,
-0.0037555694580078125,
-0.03460693359375,
0.01526641845703125,
0.061737060546875,
-0.06744384765625,
0.04339599609375,
0.052398681640625,
-0.0036869049072265625,
0.039947509765625,
-0.030914306640625,
0.05023193359375,
-0.006534576416015625,
0.033294677734375,
-0.0128631591796875,
0.044281005859375,
0.007709503173828125,
-0.0162811279296875,
0.048614501953125,
0.0253143310546875,
-0.046051025390625,
-0.03607177734375,
-0.0148162841796875,
-0.056365966796875,
-0.038665771484375,
0.054901123046875,
0.00024402141571044922,
-0.0299072265625,
-0.049346923828125,
-0.05511474609375,
0.0293121337890625,
-0.0216064453125,
0.0305938720703125,
0.0333251953125,
-0.011322021484375,
-0.00766754150390625,
-0.0288543701171875,
0.025421142578125,
-0.03521728515625,
-0.031585693359375,
-0.0210418701171875,
0.0367431640625,
0.043212890625,
-0.00830078125,
0.037872314453125,
0.01482391357421875,
0.040557861328125,
0.055328369140625,
0.045623779296875,
-0.03607177734375,
-0.022705078125,
-0.031402587890625,
0.0224761962890625,
-0.0105438232421875,
-0.00274658203125
]
] |
lllyasviel/control_v11p_sd15_softedge | 2023-05-04T18:50:55.000Z | [
"diffusers",
"art",
"controlnet",
"stable-diffusion",
"controlnet-v1-1",
"image-to-image",
"arxiv:2302.05543",
"license:openrail",
"has_space",
"diffusers:ControlNetModel",
"region:us"
] | image-to-image | lllyasviel | null | null | lllyasviel/control_v11p_sd15_softedge | 6 | 5,897 | diffusers | 2023-04-14T19:24:54 | ---
license: openrail
base_model: runwayml/stable-diffusion-v1-5
tags:
- art
- controlnet
- stable-diffusion
- controlnet-v1-1
- image-to-image
duplicated_from: ControlNet-1-1-preview/control_v11p_sd15_softedge
---
# Controlnet - v1.1 - *Soft Edge Version*
**Controlnet v1.1** is the successor model of [Controlnet v1.0](https://huggingface.co/lllyasviel/ControlNet)
and was released in [lllyasviel/ControlNet-v1-1](https://huggingface.co/lllyasviel/ControlNet-v1-1) by [Lvmin Zhang](https://huggingface.co/lllyasviel).
This checkpoint is a conversion of [the original checkpoint](https://huggingface.co/lllyasviel/ControlNet-v1-1/blob/main/control_v11p_sd15_softedge.pth) into `diffusers` format.
It can be used in combination with **Stable Diffusion**, such as [runwayml/stable-diffusion-v1-5](https://huggingface.co/runwayml/stable-diffusion-v1-5).
For more details, please also have a look at the [🧨 Diffusers docs](https://huggingface.co/docs/diffusers/api/pipelines/stable_diffusion/controlnet).
ControlNet is a neural network structure to control diffusion models by adding extra conditions.
This checkpoint corresponds to the ControlNet conditioned on **Soft edges**.
## Model Details
- **Developed by:** Lvmin Zhang, Maneesh Agrawala
- **Model type:** Diffusion-based text-to-image generation model
- **Language(s):** English
- **License:** [The CreativeML OpenRAIL M license](https://huggingface.co/spaces/CompVis/stable-diffusion-license) is an [Open RAIL M license](https://www.licenses.ai/blog/2022/8/18/naming-convention-of-responsible-ai-licenses), adapted from the work that [BigScience](https://bigscience.huggingface.co/) and [the RAIL Initiative](https://www.licenses.ai/) are jointly carrying in the area of responsible AI licensing. See also [the article about the BLOOM Open RAIL license](https://bigscience.huggingface.co/blog/the-bigscience-rail-license) on which our license is based.
- **Resources for more information:** [GitHub Repository](https://github.com/lllyasviel/ControlNet), [Paper](https://arxiv.org/abs/2302.05543).
- **Cite as:**
@misc{zhang2023adding,
title={Adding Conditional Control to Text-to-Image Diffusion Models},
author={Lvmin Zhang and Maneesh Agrawala},
year={2023},
eprint={2302.05543},
archivePrefix={arXiv},
primaryClass={cs.CV}
}
## Introduction
Controlnet was proposed in [*Adding Conditional Control to Text-to-Image Diffusion Models*](https://arxiv.org/abs/2302.05543) by
Lvmin Zhang, Maneesh Agrawala.
The abstract reads as follows:
*We present a neural network structure, ControlNet, to control pretrained large diffusion models to support additional input conditions.
The ControlNet learns task-specific conditions in an end-to-end way, and the learning is robust even when the training dataset is small (< 50k).
Moreover, training a ControlNet is as fast as fine-tuning a diffusion model, and the model can be trained on a personal devices.
Alternatively, if powerful computation clusters are available, the model can scale to large amounts (millions to billions) of data.
We report that large diffusion models like Stable Diffusion can be augmented with ControlNets to enable conditional inputs like edge maps, segmentation maps, keypoints, etc.
This may enrich the methods to control large diffusion models and further facilitate related applications.*
## Example
It is recommended to use the checkpoint with [Stable Diffusion v1-5](https://huggingface.co/runwayml/stable-diffusion-v1-5) as the checkpoint
has been trained on it.
Experimentally, the checkpoint can be used with other diffusion models such as dreamboothed stable diffusion.
**Note**: If you want to process an image to create the auxiliary conditioning, external dependencies are required as shown below:
1. Install https://github.com/patrickvonplaten/controlnet_aux
```sh
$ pip install controlnet_aux==0.3.0
```
2. Let's install `diffusers` and related packages:
```
$ pip install diffusers transformers accelerate
```
3. Run code:
```python
import torch
import os
from huggingface_hub import HfApi
from pathlib import Path
from diffusers.utils import load_image
from PIL import Image
import numpy as np
from controlnet_aux import PidiNetDetector, HEDdetector
from diffusers import (
ControlNetModel,
StableDiffusionControlNetPipeline,
UniPCMultistepScheduler,
)
checkpoint = "lllyasviel/control_v11p_sd15_softedge"
image = load_image(
"https://huggingface.co/lllyasviel/control_v11p_sd15_softedge/resolve/main/images/input.png"
)
prompt = "royal chamber with fancy bed"
processor = HEDdetector.from_pretrained('lllyasviel/Annotators')
processor = PidiNetDetector.from_pretrained('lllyasviel/Annotators')
control_image = processor(image, safe=True)
control_image.save("./images/control.png")
controlnet = ControlNetModel.from_pretrained(checkpoint, torch_dtype=torch.float16)
pipe = StableDiffusionControlNetPipeline.from_pretrained(
"runwayml/stable-diffusion-v1-5", controlnet=controlnet, torch_dtype=torch.float16
)
pipe.scheduler = UniPCMultistepScheduler.from_config(pipe.scheduler.config)
pipe.enable_model_cpu_offload()
generator = torch.manual_seed(0)
image = pipe(prompt, num_inference_steps=30, generator=generator, image=control_image).images[0]
image.save('images/image_out.png')
```
## Other released checkpoints v1-1
The authors released 14 different checkpoints, each trained with [Stable Diffusion v1-5](https://huggingface.co/runwayml/stable-diffusion-v1-5)
on a different type of conditioning:
| Model Name | Control Image Overview| Condition Image | Control Image Example | Generated Image Example |
|---|---|---|---|---|
|[lllyasviel/control_v11p_sd15_canny](https://huggingface.co/lllyasviel/control_v11p_sd15_canny)<br/> | *Trained with canny edge detection* | A monochrome image with white edges on a black background.|<a href="https://huggingface.co/lllyasviel/control_v11p_sd15_canny/resolve/main/images/control.png"><img width="64" style="margin:0;padding:0;" src="https://huggingface.co/lllyasviel/control_v11p_sd15_canny/resolve/main/images/control.png"/></a>|<a href="https://huggingface.co/lllyasviel/control_v11p_sd15_canny/resolve/main/images/image_out.png"><img width="64" src="https://huggingface.co/lllyasviel/control_v11p_sd15_canny/resolve/main/images/image_out.png"/></a>|
|[lllyasviel/control_v11e_sd15_ip2p](https://huggingface.co/lllyasviel/control_v11e_sd15_ip2p)<br/> | *Trained with pixel to pixel instruction* | No condition .|<a href="https://huggingface.co/lllyasviel/control_v11e_sd15_ip2p/resolve/main/images/control.png"><img width="64" style="margin:0;padding:0;" src="https://huggingface.co/lllyasviel/control_v11e_sd15_ip2p/resolve/main/images/control.png"/></a>|<a href="https://huggingface.co/lllyasviel/control_v11e_sd15_ip2p/resolve/main/images/image_out.png"><img width="64" src="https://huggingface.co/lllyasviel/control_v11e_sd15_ip2p/resolve/main/images/image_out.png"/></a>|
|[lllyasviel/control_v11p_sd15_inpaint](https://huggingface.co/lllyasviel/control_v11p_sd15_inpaint)<br/> | Trained with image inpainting | No condition.|<a href="https://huggingface.co/lllyasviel/control_v11p_sd15_inpaint/resolve/main/images/control.png"><img width="64" style="margin:0;padding:0;" src="https://huggingface.co/lllyasviel/control_v11p_sd15_inpaint/resolve/main/images/control.png"/></a>|<a href="https://huggingface.co/lllyasviel/control_v11p_sd15_inpaint/resolve/main/images/output.png"><img width="64" src="https://huggingface.co/lllyasviel/control_v11p_sd15_inpaint/resolve/main/images/output.png"/></a>|
|[lllyasviel/control_v11p_sd15_mlsd](https://huggingface.co/lllyasviel/control_v11p_sd15_mlsd)<br/> | Trained with multi-level line segment detection | An image with annotated line segments.|<a href="https://huggingface.co/lllyasviel/control_v11p_sd15_mlsd/resolve/main/images/control.png"><img width="64" style="margin:0;padding:0;" src="https://huggingface.co/lllyasviel/control_v11p_sd15_mlsd/resolve/main/images/control.png"/></a>|<a href="https://huggingface.co/lllyasviel/control_v11p_sd15_mlsd/resolve/main/images/image_out.png"><img width="64" src="https://huggingface.co/lllyasviel/control_v11p_sd15_mlsd/resolve/main/images/image_out.png"/></a>|
|[lllyasviel/control_v11f1p_sd15_depth](https://huggingface.co/lllyasviel/control_v11f1p_sd15_depth)<br/> | Trained with depth estimation | An image with depth information, usually represented as a grayscale image.|<a href="https://huggingface.co/lllyasviel/control_v11f1p_sd15_depth/resolve/main/images/control.png"><img width="64" style="margin:0;padding:0;" src="https://huggingface.co/lllyasviel/control_v11f1p_sd15_depth/resolve/main/images/control.png"/></a>|<a href="https://huggingface.co/lllyasviel/control_v11f1p_sd15_depth/resolve/main/images/image_out.png"><img width="64" src="https://huggingface.co/lllyasviel/control_v11f1p_sd15_depth/resolve/main/images/image_out.png"/></a>|
|[lllyasviel/control_v11p_sd15_normalbae](https://huggingface.co/lllyasviel/control_v11p_sd15_normalbae)<br/> | Trained with surface normal estimation | An image with surface normal information, usually represented as a color-coded image.|<a href="https://huggingface.co/lllyasviel/control_v11p_sd15_normalbae/resolve/main/images/control.png"><img width="64" style="margin:0;padding:0;" src="https://huggingface.co/lllyasviel/control_v11p_sd15_normalbae/resolve/main/images/control.png"/></a>|<a href="https://huggingface.co/lllyasviel/control_v11p_sd15_normalbae/resolve/main/images/image_out.png"><img width="64" src="https://huggingface.co/lllyasviel/control_v11p_sd15_normalbae/resolve/main/images/image_out.png"/></a>|
|[lllyasviel/control_v11p_sd15_seg](https://huggingface.co/lllyasviel/control_v11p_sd15_seg)<br/> | Trained with image segmentation | An image with segmented regions, usually represented as a color-coded image.|<a href="https://huggingface.co/lllyasviel/control_v11p_sd15_seg/resolve/main/images/control.png"><img width="64" style="margin:0;padding:0;" src="https://huggingface.co/lllyasviel/control_v11p_sd15_seg/resolve/main/images/control.png"/></a>|<a href="https://huggingface.co/lllyasviel/control_v11p_sd15_seg/resolve/main/images/image_out.png"><img width="64" src="https://huggingface.co/lllyasviel/control_v11p_sd15_seg/resolve/main/images/image_out.png"/></a>|
|[lllyasviel/control_v11p_sd15_lineart](https://huggingface.co/lllyasviel/control_v11p_sd15_lineart)<br/> | Trained with line art generation | An image with line art, usually black lines on a white background.|<a href="https://huggingface.co/lllyasviel/control_v11p_sd15_lineart/resolve/main/images/control.png"><img width="64" style="margin:0;padding:0;" src="https://huggingface.co/lllyasviel/control_v11p_sd15_lineart/resolve/main/images/control.png"/></a>|<a href="https://huggingface.co/lllyasviel/control_v11p_sd15_lineart/resolve/main/images/image_out.png"><img width="64" src="https://huggingface.co/lllyasviel/control_v11p_sd15_lineart/resolve/main/images/image_out.png"/></a>|
|[lllyasviel/control_v11p_sd15s2_lineart_anime](https://huggingface.co/lllyasviel/control_v11p_sd15s2_lineart_anime)<br/> | Trained with anime line art generation | An image with anime-style line art.|<a href="https://huggingface.co/lllyasviel/control_v11p_sd15s2_lineart_anime/resolve/main/images/control.png"><img width="64" style="margin:0;padding:0;" src="https://huggingface.co/lllyasviel/control_v11p_sd15s2_lineart_anime/resolve/main/images/control.png"/></a>|<a href="https://huggingface.co/lllyasviel/control_v11p_sd15s2_lineart_anime/resolve/main/images/image_out.png"><img width="64" src="https://huggingface.co/lllyasviel/control_v11p_sd15s2_lineart_anime/resolve/main/images/image_out.png"/></a>|
|[lllyasviel/control_v11p_sd15_openpose](https://huggingface.co/lllyasviel/control_v11p_sd15s2_lineart_anime)<br/> | Trained with human pose estimation | An image with human poses, usually represented as a set of keypoints or skeletons.|<a href="https://huggingface.co/lllyasviel/control_v11p_sd15_openpose/resolve/main/images/control.png"><img width="64" style="margin:0;padding:0;" src="https://huggingface.co/lllyasviel/control_v11p_sd15_openpose/resolve/main/images/control.png"/></a>|<a href="https://huggingface.co/lllyasviel/control_v11p_sd15_openpose/resolve/main/images/image_out.png"><img width="64" src="https://huggingface.co/lllyasviel/control_v11p_sd15_openpose/resolve/main/images/image_out.png"/></a>|
|[lllyasviel/control_v11p_sd15_scribble](https://huggingface.co/lllyasviel/control_v11p_sd15_scribble)<br/> | Trained with scribble-based image generation | An image with scribbles, usually random or user-drawn strokes.|<a href="https://huggingface.co/lllyasviel/control_v11p_sd15_scribble/resolve/main/images/control.png"><img width="64" style="margin:0;padding:0;" src="https://huggingface.co/lllyasviel/control_v11p_sd15_scribble/resolve/main/images/control.png"/></a>|<a href="https://huggingface.co/lllyasviel/control_v11p_sd15_scribble/resolve/main/images/image_out.png"><img width="64" src="https://huggingface.co/lllyasviel/control_v11p_sd15_scribble/resolve/main/images/image_out.png"/></a>|
|[lllyasviel/control_v11p_sd15_softedge](https://huggingface.co/lllyasviel/control_v11p_sd15_softedge)<br/> | Trained with soft edge image generation | An image with soft edges, usually to create a more painterly or artistic effect.|<a href="https://huggingface.co/lllyasviel/control_v11p_sd15_softedge/resolve/main/images/control.png"><img width="64" style="margin:0;padding:0;" src="https://huggingface.co/lllyasviel/control_v11p_sd15_softedge/resolve/main/images/control.png"/></a>|<a href="https://huggingface.co/lllyasviel/control_v11p_sd15_softedge/resolve/main/images/image_out.png"><img width="64" src="https://huggingface.co/lllyasviel/control_v11p_sd15_softedge/resolve/main/images/image_out.png"/></a>|
|[lllyasviel/control_v11e_sd15_shuffle](https://huggingface.co/lllyasviel/control_v11e_sd15_shuffle)<br/> | Trained with image shuffling | An image with shuffled patches or regions.|<a href="https://huggingface.co/lllyasviel/control_v11e_sd15_shuffle/resolve/main/images/control.png"><img width="64" style="margin:0;padding:0;" src="https://huggingface.co/lllyasviel/control_v11e_sd15_shuffle/resolve/main/images/control.png"/></a>|<a href="https://huggingface.co/lllyasviel/control_v11e_sd15_shuffle/resolve/main/images/image_out.png"><img width="64" src="https://huggingface.co/lllyasviel/control_v11e_sd15_shuffle/resolve/main/images/image_out.png"/></a>|
|[lllyasviel/control_v11f1e_sd15_tile](https://huggingface.co/lllyasviel/control_v11f1e_sd15_tile)<br/> | Trained with image tiling | A blurry image or part of an image .|<a href="https://huggingface.co/lllyasviel/control_v11f1e_sd15_tile/resolve/main/images/original.png"><img width="64" style="margin:0;padding:0;" src="https://huggingface.co/lllyasviel/control_v11f1e_sd15_tile/resolve/main/images/original.png"/></a>|<a href="https://huggingface.co/lllyasviel/control_v11f1e_sd15_tile/resolve/main/images/output.png"><img width="64" src="https://huggingface.co/lllyasviel/control_v11f1e_sd15_tile/resolve/main/images/output.png"/></a>|
## Improvements in Soft Edge 1.1:
- Soft Edge 1.1 was called HED 1.0 in previous ControlNet.
- The training dataset of previous cnet 1.0 has several problems including (1) a small group of greyscale human images are duplicated thousands of times (!!), causing the previous model somewhat likely to generate grayscale human images; (2) some images has low quality, very blurry, or significant JPEG artifacts; (3) a small group of images has wrong paired prompts caused by a mistake in our data processing scripts. The new model fixed all problems of the training dataset and should be more reasonable in many cases.
- The Soft Edge 1.1 is significantly (in nealy 100% cases) better than HED 1.0. This is mainly because HED or PIDI estimator tend to hide a corrupted greyscale version of original image inside the soft edge map and the previous model HED 1.0 is over-fitted to restore that hidden corrupted image rather than perform boundary-aware diffusion. The training of Soft Edge 1.1 used 75% "safe" filtering to remove such hidden corrupted greyscale images insider control maps. This makes the Soft Edge 1.1 very robust. In out test, Soft Edge 1.1 is as usable as the depth model and has potential to be more frequently used.
## More information
For more information, please also have a look at the [Diffusers ControlNet Blog Post](https://huggingface.co/blog/controlnet) and have a look at the [official docs](https://github.com/lllyasviel/ControlNet-v1-1-nightly). | 16,762 | [
[
-0.044891357421875,
-0.04638671875,
0.0103607177734375,
0.04083251953125,
-0.01727294921875,
-0.020538330078125,
0.00238037109375,
-0.0400390625,
0.038818359375,
0.021026611328125,
-0.0577392578125,
-0.02880859375,
-0.055084228515625,
-0.0124359130859375,
-0.01042938232421875,
0.06402587890625,
-0.02239990234375,
-0.0011653900146484375,
0.00936126708984375,
-0.00569915771484375,
-0.0056304931640625,
-0.01214599609375,
-0.09356689453125,
-0.03759765625,
0.03558349609375,
0.002674102783203125,
0.04193115234375,
0.04168701171875,
0.03814697265625,
0.02923583984375,
-0.02862548828125,
0.004207611083984375,
-0.0214691162109375,
-0.015472412109375,
0.01263427734375,
-0.009185791015625,
-0.056182861328125,
0.00531768798828125,
0.053314208984375,
0.0240325927734375,
0.0013914108276367188,
-0.01702880859375,
0.01131439208984375,
0.051422119140625,
-0.037933349609375,
-0.008758544921875,
-0.01139068603515625,
0.0201416015625,
-0.0110321044921875,
0.00909423828125,
-0.01479339599609375,
-0.022735595703125,
0.00754547119140625,
-0.055999755859375,
-0.006229400634765625,
-0.0147857666015625,
0.102783203125,
0.02239990234375,
-0.034210205078125,
-0.005523681640625,
-0.019561767578125,
0.048858642578125,
-0.061676025390625,
0.00870513916015625,
0.0086212158203125,
0.01557159423828125,
-0.017547607421875,
-0.077392578125,
-0.03814697265625,
-0.0110321044921875,
-0.007808685302734375,
0.03521728515625,
-0.025238037109375,
0.0069122314453125,
0.019378662109375,
0.0164031982421875,
-0.028411865234375,
0.0209503173828125,
-0.024169921875,
-0.0304718017578125,
0.048980712890625,
-0.00023734569549560547,
0.0438232421875,
0.003826141357421875,
-0.0452880859375,
-0.004230499267578125,
-0.03057861328125,
0.026641845703125,
0.0168609619140625,
-0.00853729248046875,
-0.05712890625,
0.030426025390625,
-0.0037841796875,
0.0538330078125,
0.0323486328125,
-0.0157012939453125,
0.03875732421875,
-0.01309967041015625,
-0.0269927978515625,
-0.0201416015625,
0.0772705078125,
0.038818359375,
0.01320648193359375,
-0.0002110004425048828,
-0.01416778564453125,
-0.009521484375,
-0.0038394927978515625,
-0.0916748046875,
-0.01280975341796875,
0.01715087890625,
-0.04217529296875,
-0.026397705078125,
-0.0114288330078125,
-0.05340576171875,
-0.015869140625,
-0.00440216064453125,
0.0292205810546875,
-0.0465087890625,
-0.03900146484375,
0.0098876953125,
-0.033843994140625,
0.0413818359375,
0.045684814453125,
-0.034027099609375,
0.0162200927734375,
0.01454925537109375,
0.07708740234375,
-0.01983642578125,
-0.0125732421875,
-0.0206146240234375,
-0.00676727294921875,
-0.02337646484375,
0.0396728515625,
-0.01047515869140625,
-0.01000213623046875,
-0.0023097991943359375,
0.0273895263671875,
-0.01010894775390625,
-0.0252532958984375,
0.033966064453125,
-0.0260467529296875,
0.01363372802734375,
-0.0012378692626953125,
-0.0309600830078125,
-0.01305389404296875,
0.018035888671875,
-0.03656005859375,
0.0570068359375,
0.018798828125,
-0.079833984375,
0.0251617431640625,
-0.038970947265625,
-0.0197601318359375,
-0.0177459716796875,
0.01116943359375,
-0.057861328125,
-0.030609130859375,
0.00030350685119628906,
0.045440673828125,
0.001453399658203125,
-0.00919342041015625,
-0.036376953125,
-0.00431060791015625,
0.01210784912109375,
-0.00957489013671875,
0.093017578125,
0.011260986328125,
-0.047821044921875,
0.016021728515625,
-0.05206298828125,
0.004428863525390625,
0.0113525390625,
-0.0177459716796875,
0.006565093994140625,
-0.022613525390625,
0.0112762451171875,
0.0477294921875,
0.02801513671875,
-0.05120849609375,
0.01207733154296875,
-0.01934814453125,
0.036224365234375,
0.05126953125,
0.0162506103515625,
0.04559326171875,
-0.03924560546875,
0.041229248046875,
0.0208587646484375,
0.0240325927734375,
0.004703521728515625,
-0.038848876953125,
-0.0782470703125,
-0.044158935546875,
0.002300262451171875,
0.044921875,
-0.0611572265625,
0.058837890625,
0.00868988037109375,
-0.050933837890625,
-0.019195556640625,
0.006374359130859375,
0.038909912109375,
0.037567138671875,
0.023101806640625,
-0.03643798828125,
-0.025390625,
-0.07122802734375,
0.01068878173828125,
0.0181884765625,
0.00009751319885253906,
0.0146484375,
0.04986572265625,
-0.0090179443359375,
0.048980712890625,
-0.0191802978515625,
-0.0307769775390625,
-0.00799560546875,
-0.00713348388671875,
0.0237579345703125,
0.07843017578125,
0.058135986328125,
-0.059600830078125,
-0.047576904296875,
-0.00433349609375,
-0.0665283203125,
-0.003299713134765625,
-0.016326904296875,
-0.038970947265625,
0.018310546875,
0.04437255859375,
-0.051116943359375,
0.058746337890625,
0.03863525390625,
-0.04351806640625,
0.045166015625,
-0.0267333984375,
0.011474609375,
-0.0728759765625,
0.015960693359375,
0.0280914306640625,
-0.024658203125,
-0.04608154296875,
0.0078887939453125,
0.0093231201171875,
0.005279541015625,
-0.0552978515625,
0.055419921875,
-0.038238525390625,
0.01428985595703125,
-0.0228118896484375,
-0.00716400146484375,
0.005767822265625,
0.05078125,
0.0166015625,
0.037567138671875,
0.07391357421875,
-0.046478271484375,
0.0250244140625,
0.0313720703125,
-0.013214111328125,
0.06451416015625,
-0.06390380859375,
0.01096343994140625,
-0.01251983642578125,
0.043121337890625,
-0.07086181640625,
-0.019561767578125,
0.04815673828125,
-0.037811279296875,
0.043914794921875,
-0.022918701171875,
-0.018646240234375,
-0.0322265625,
-0.0264129638671875,
0.0128936767578125,
0.05999755859375,
-0.036956787109375,
0.0292205810546875,
0.01171112060546875,
0.01059722900390625,
-0.036865234375,
-0.0682373046875,
-0.005947113037109375,
-0.0283050537109375,
-0.062164306640625,
0.0382080078125,
-0.0107879638671875,
0.002155303955078125,
0.0012617111206054688,
0.0038394927978515625,
-0.024169921875,
-0.00007611513137817383,
0.0302276611328125,
0.0184478759765625,
-0.00666046142578125,
-0.01244354248046875,
0.00905609130859375,
-0.01384735107421875,
-0.004467010498046875,
-0.0276336669921875,
0.034759521484375,
0.0015773773193359375,
-0.014739990234375,
-0.074462890625,
0.01788330078125,
0.0430908203125,
-0.0033416748046875,
0.06768798828125,
0.07086181640625,
-0.033782958984375,
-0.0016117095947265625,
-0.02947998046875,
-0.013427734375,
-0.03973388671875,
-0.0033740997314453125,
-0.01904296875,
-0.0517578125,
0.050323486328125,
0.0036449432373046875,
-0.0013227462768554688,
0.050079345703125,
0.0294036865234375,
-0.0170745849609375,
0.067626953125,
0.04156494140625,
-0.007450103759765625,
0.060760498046875,
-0.058837890625,
-0.0125885009765625,
-0.0762939453125,
-0.0212554931640625,
-0.0244903564453125,
-0.05035400390625,
-0.027801513671875,
-0.024322509765625,
0.03466796875,
0.031951904296875,
-0.05572509765625,
0.034759521484375,
-0.045928955078125,
0.006134033203125,
0.02740478515625,
0.041656494140625,
-0.0111236572265625,
-0.01039886474609375,
-0.011627197265625,
0.0049591064453125,
-0.049530029296875,
-0.0213775634765625,
0.0450439453125,
0.042633056640625,
0.041534423828125,
-0.004863739013671875,
0.045745849609375,
0.0034122467041015625,
0.0232696533203125,
-0.04254150390625,
0.040771484375,
-0.0008101463317871094,
-0.042938232421875,
-0.01508331298828125,
-0.024688720703125,
-0.07794189453125,
0.006740570068359375,
-0.036956787109375,
-0.0555419921875,
0.026153564453125,
0.0187530517578125,
-0.00809478759765625,
0.036285400390625,
-0.052642822265625,
0.05859375,
-0.0013074874877929688,
-0.05029296875,
0.0038299560546875,
-0.06292724609375,
0.0158843994140625,
0.0214691162109375,
-0.0149688720703125,
0.00021409988403320312,
-0.01039886474609375,
0.06793212890625,
-0.06024169921875,
0.065673828125,
-0.04193115234375,
-0.0019464492797851562,
0.0282440185546875,
-0.0014438629150390625,
0.042449951171875,
-0.01134490966796875,
-0.018798828125,
0.006793975830078125,
-0.00604248046875,
-0.04473876953125,
-0.0264739990234375,
0.052520751953125,
-0.0535888671875,
-0.016357421875,
-0.0218505859375,
-0.019500732421875,
0.01558685302734375,
0.02001953125,
0.051361083984375,
0.03173828125,
0.01517486572265625,
0.0030670166015625,
0.050384521484375,
-0.0246124267578125,
0.052337646484375,
0.0029697418212890625,
-0.005466461181640625,
-0.040771484375,
0.0546875,
0.002197265625,
0.028564453125,
0.01323699951171875,
0.009979248046875,
-0.0179901123046875,
-0.0396728515625,
-0.035308837890625,
0.0338134765625,
-0.04632568359375,
-0.03302001953125,
-0.047821044921875,
-0.036102294921875,
-0.0290374755859375,
-0.03814697265625,
-0.0222320556640625,
-0.02056884765625,
-0.052703857421875,
0.01313018798828125,
0.05047607421875,
0.03985595703125,
-0.016448974609375,
0.045684814453125,
-0.021728515625,
0.017242431640625,
0.0206298828125,
0.0297088623046875,
-0.00524139404296875,
-0.045989990234375,
0.0018491744995117188,
0.0088653564453125,
-0.03759765625,
-0.059295654296875,
0.03668212890625,
0.006023406982421875,
0.035919189453125,
0.041259765625,
-0.017608642578125,
0.048309326171875,
-0.02557373046875,
0.044525146484375,
0.045135498046875,
-0.062225341796875,
0.038238525390625,
-0.030609130859375,
0.0219268798828125,
0.025787353515625,
0.041839599609375,
-0.031829833984375,
-0.0250244140625,
-0.058441162109375,
-0.05157470703125,
0.0438232421875,
0.017730712890625,
-0.004947662353515625,
0.027008056640625,
0.052947998046875,
-0.025390625,
0.0114593505859375,
-0.061614990234375,
-0.033233642578125,
-0.01995849609375,
0.0013818740844726562,
0.0034236907958984375,
0.0040740966796875,
-0.00482940673828125,
-0.03460693359375,
0.070068359375,
-0.0022678375244140625,
0.04351806640625,
0.03924560546875,
0.005756378173828125,
-0.012481689453125,
-0.0225677490234375,
0.040985107421875,
0.037841796875,
-0.007659912109375,
-0.0210723876953125,
0.006923675537109375,
-0.0300445556640625,
0.017181396484375,
-0.0010833740234375,
-0.027923583984375,
-0.006885528564453125,
0.028717041015625,
0.063232421875,
-0.01374053955078125,
-0.01311492919921875,
0.058837890625,
0.0036449432373046875,
-0.04095458984375,
-0.023223876953125,
0.004680633544921875,
0.010040283203125,
0.0362548828125,
0.01178741455078125,
0.0272369384765625,
0.006549835205078125,
-0.01165771484375,
0.0239105224609375,
0.0447998046875,
-0.04644775390625,
-0.012115478515625,
0.05694580078125,
0.005767822265625,
-0.01100921630859375,
0.0307159423828125,
-0.032958984375,
-0.056365966796875,
0.07171630859375,
0.03936767578125,
0.0560302734375,
-0.0086517333984375,
0.021148681640625,
0.05267333984375,
0.01479339599609375,
0.00719451904296875,
0.01519012451171875,
0.01027679443359375,
-0.0513916015625,
-0.0310821533203125,
-0.032012939453125,
-0.004009246826171875,
0.0124969482421875,
-0.03094482421875,
0.0343017578125,
-0.06048583984375,
-0.019317626953125,
-0.007259368896484375,
0.00986480712890625,
-0.05450439453125,
0.03265380859375,
0.005222320556640625,
0.095458984375,
-0.06365966796875,
0.06201171875,
0.045257568359375,
-0.03460693359375,
-0.0662841796875,
-0.0021152496337890625,
0.0012950897216796875,
-0.060211181640625,
0.047576904296875,
0.01398468017578125,
-0.004886627197265625,
0.0078125,
-0.0606689453125,
-0.042144775390625,
0.09820556640625,
0.0176544189453125,
-0.0168914794921875,
0.0072784423828125,
-0.03961181640625,
0.033599853515625,
-0.03131103515625,
0.0390625,
0.0325927734375,
0.040130615234375,
0.03289794921875,
-0.059478759765625,
0.0185394287109375,
-0.034881591796875,
0.00783538818359375,
0.0131378173828125,
-0.0767822265625,
0.0689697265625,
-0.0012750625610351562,
-0.01166534423828125,
0.019683837890625,
0.05712890625,
0.01551055908203125,
0.01326751708984375,
0.04937744140625,
0.060028076171875,
0.0245513916015625,
-0.00909423828125,
0.07440185546875,
-0.006122589111328125,
0.0226898193359375,
0.04803466796875,
0.0178680419921875,
0.040985107421875,
0.02679443359375,
-0.00439453125,
0.039581298828125,
0.06597900390625,
0.003326416015625,
0.033477783203125,
0.0367431640625,
-0.02203369140625,
-0.00968170166015625,
-0.006374359130859375,
-0.0283966064453125,
0.006702423095703125,
0.0214996337890625,
-0.0203704833984375,
-0.0181884765625,
0.020599365234375,
0.0232696533203125,
-0.01369476318359375,
-0.0347900390625,
0.052520751953125,
-0.007709503173828125,
-0.04052734375,
0.05731201171875,
-0.00518798828125,
0.08673095703125,
-0.05438232421875,
0.0025177001953125,
-0.02032470703125,
0.007221221923828125,
-0.03106689453125,
-0.0657958984375,
0.01328277587890625,
-0.0161590576171875,
0.0227203369140625,
-0.0296478271484375,
0.0577392578125,
-0.0293121337890625,
-0.03253173828125,
0.03851318359375,
0.009765625,
0.0292510986328125,
0.01299285888671875,
-0.0828857421875,
0.0188751220703125,
0.005615234375,
-0.03814697265625,
0.016204833984375,
0.02691650390625,
0.01543426513671875,
0.056671142578125,
0.0238037109375,
0.02740478515625,
0.01983642578125,
-0.0180511474609375,
0.079345703125,
-0.0214996337890625,
-0.0241851806640625,
-0.045135498046875,
0.060546875,
-0.0255584716796875,
-0.035003662109375,
0.041351318359375,
0.0211181640625,
0.054901123046875,
-0.0052337646484375,
0.0537109375,
-0.03155517578125,
0.01242828369140625,
-0.050811767578125,
0.06451416015625,
-0.06591796875,
-0.02423095703125,
-0.026275634765625,
-0.052703857421875,
-0.0213775634765625,
0.06536865234375,
-0.01119232177734375,
0.018402099609375,
0.041900634765625,
0.07550048828125,
-0.0168609619140625,
-0.0440673828125,
0.004222869873046875,
0.0086669921875,
0.02423095703125,
0.05609130859375,
0.0506591796875,
-0.050445556640625,
0.021270751953125,
-0.042449951171875,
-0.03765869140625,
-0.00800323486328125,
-0.07464599609375,
-0.06683349609375,
-0.05572509765625,
-0.054534912109375,
-0.05560302734375,
-0.0170135498046875,
0.05609130859375,
0.08770751953125,
-0.049285888671875,
-0.01201629638671875,
-0.025177001953125,
0.007450103759765625,
-0.01323699951171875,
-0.016265869140625,
0.0283966064453125,
-0.0076446533203125,
-0.0648193359375,
0.00028705596923828125,
0.015777587890625,
0.04296875,
-0.00794219970703125,
-0.03076171875,
-0.029815673828125,
-0.0200042724609375,
0.0191192626953125,
0.035308837890625,
-0.0325927734375,
-0.01470184326171875,
-0.0218658447265625,
-0.017120361328125,
0.00830841064453125,
0.0400390625,
-0.033538818359375,
0.0115203857421875,
0.03985595703125,
0.03167724609375,
0.06365966796875,
-0.01361846923828125,
0.01061248779296875,
-0.040985107421875,
0.041259765625,
0.002857208251953125,
0.035614013671875,
0.00836944580078125,
-0.0271453857421875,
0.033294677734375,
0.024383544921875,
-0.0576171875,
-0.0338134765625,
0.01336669921875,
-0.1033935546875,
-0.01065826416015625,
0.0758056640625,
-0.028076171875,
-0.03253173828125,
0.014556884765625,
-0.034423828125,
0.0274200439453125,
-0.02679443359375,
0.0159912109375,
0.0279693603515625,
-0.01453399658203125,
-0.029632568359375,
-0.0309600830078125,
0.048614501953125,
0.019683837890625,
-0.060455322265625,
-0.041839599609375,
0.04095458984375,
0.028839111328125,
0.0236358642578125,
0.0662841796875,
-0.007411956787109375,
0.0106658935546875,
-0.006683349609375,
0.019317626953125,
0.001789093017578125,
-0.0106353759765625,
-0.042022705078125,
-0.0084228515625,
-0.0160980224609375,
-0.0299530029296875
]
] |
KnutJaegersberg/LLongMA-3b-LIMA | 2023-09-03T18:51:00.000Z | [
"transformers",
"pytorch",
"llama",
"text-generation",
"custom_code",
"license:cc-by-nc-4.0",
"endpoints_compatible",
"has_space",
"text-generation-inference",
"region:us"
] | text-generation | KnutJaegersberg | null | null | KnutJaegersberg/LLongMA-3b-LIMA | 2 | 5,896 | transformers | 2023-09-03T13:54:25 | ---
license: cc-by-nc-4.0
---
Prompt example:
```
### Instruction:
How do you fine tune a large language model?
### Response:
``` | 132 | [
[
-0.029998779296875,
-0.06585693359375,
0.019927978515625,
-0.00537109375,
-0.02874755859375,
-0.00894927978515625,
-0.01678466796875,
0.01068878173828125,
-0.003139495849609375,
0.055450439453125,
-0.0540771484375,
-0.0222930908203125,
-0.01471710205078125,
-0.0007977485656738281,
-0.02642822265625,
0.071044921875,
-0.0028018951416015625,
0.0117645263671875,
0.01131439208984375,
0.02423095703125,
-0.06182861328125,
-0.005283355712890625,
-0.09149169921875,
-0.006687164306640625,
0.0308380126953125,
0.07061767578125,
0.03759765625,
0.057769775390625,
0.019134521484375,
0.0174407958984375,
-0.007610321044921875,
0.008819580078125,
-0.04730224609375,
0.01300048828125,
-0.004138946533203125,
-0.0218505859375,
-0.041259765625,
-0.012237548828125,
0.058502197265625,
0.06201171875,
0.0090179443359375,
0.03204345703125,
-0.01026153564453125,
0.0189361572265625,
-0.01715087890625,
0.01611328125,
-0.009521484375,
-0.0009264945983886719,
-0.003284454345703125,
-0.00559234619140625,
-0.045257568359375,
-0.042083740234375,
-0.0172119140625,
-0.041595458984375,
0.00798797607421875,
0.0211639404296875,
0.06109619140625,
0.01116180419921875,
-0.038665771484375,
0.006763458251953125,
-0.057373046875,
0.046905517578125,
-0.01517486572265625,
0.0145721435546875,
0.048583984375,
0.035400390625,
-0.01531982421875,
-0.05645751953125,
-0.03863525390625,
-0.0115966796875,
0.00295257568359375,
-0.00335693359375,
0.00586700439453125,
-0.0157470703125,
0.053985595703125,
0.01401519775390625,
-0.033905029296875,
0.00811004638671875,
-0.042694091796875,
-0.028106689453125,
0.03521728515625,
0.034881591796875,
0.0171356201171875,
0.0261688232421875,
0.020294189453125,
-0.0211334228515625,
-0.035430908203125,
-0.01348114013671875,
0.007659912109375,
0.0322265625,
-0.018585205078125,
0.05572509765625,
-0.01439666748046875,
0.06341552734375,
0.00331878662109375,
0.04278564453125,
-0.01059722900390625,
-0.03619384765625,
-0.0275115966796875,
-0.0135955810546875,
0.038787841796875,
0.033355712890625,
0.043487548828125,
-0.007053375244140625,
-0.0199432373046875,
-0.01551055908203125,
0.0132904052734375,
-0.0740966796875,
-0.031646728515625,
0.0089569091796875,
-0.047515869140625,
-0.01226806640625,
-0.0139312744140625,
-0.0736083984375,
-0.01450347900390625,
-0.0212554931640625,
0.0185089111328125,
0.00978851318359375,
-0.0213470458984375,
0.0310211181640625,
-0.022705078125,
0.0361328125,
0.0182647705078125,
-0.0797119140625,
0.04052734375,
0.0440673828125,
0.01776123046875,
0.045074462890625,
0.009185791015625,
-0.0579833984375,
-0.0154876708984375,
-0.033355712890625,
0.05816650390625,
-0.0273895263671875,
-0.041290283203125,
-0.0005207061767578125,
0.00850677490234375,
0.01904296875,
-0.034942626953125,
0.02996826171875,
-0.0310516357421875,
0.046173095703125,
-0.044342041015625,
-0.0321044921875,
-0.005054473876953125,
0.0182037353515625,
-0.042144775390625,
0.049560546875,
0.03399658203125,
-0.032196044921875,
-0.012969970703125,
-0.07086181640625,
0.00632476806640625,
0.00991058349609375,
0.004241943359375,
0.024810791015625,
0.01277923583984375,
0.00859832763671875,
0.0207672119140625,
-0.044158935546875,
-0.01166534423828125,
-0.041900634765625,
-0.001781463623046875,
0.019500732421875,
-0.02935791015625,
0.060394287109375,
0.02996826171875,
-0.0027904510498046875,
0.02508544921875,
-0.072265625,
0.01641845703125,
0.0257720947265625,
-0.018585205078125,
-0.005786895751953125,
-0.0276947021484375,
0.0350341796875,
-0.0146484375,
0.056304931640625,
-0.0399169921875,
0.06756591796875,
-0.0185546875,
0.0274658203125,
0.056488037109375,
0.0172271728515625,
0.01611328125,
0.002773284912109375,
0.038421630859375,
-0.01561737060546875,
0.0032958984375,
-0.0252685546875,
0.00455474853515625,
-0.06658935546875,
-0.0017290115356445312,
0.0285491943359375,
0.04278564453125,
-0.0277252197265625,
0.0126495361328125,
0.01016998291015625,
-0.009674072265625,
-0.00653076171875,
-0.002872467041015625,
0.0251617431640625,
0.04547119140625,
0.03717041015625,
0.0007171630859375,
-0.05615234375,
-0.055450439453125,
0.009063720703125,
-0.022705078125,
-0.004390716552734375,
0.0115203857421875,
0.0258636474609375,
-0.027557373046875,
0.036895751953125,
-0.060333251953125,
0.0499267578125,
-0.013946533203125,
0.00948333740234375,
0.00258636474609375,
0.049072265625,
0.0135040283203125,
-0.042999267578125,
-0.032745361328125,
-0.01318359375,
-0.036041259765625,
-0.03173828125,
-0.00574493408203125,
-0.031982421875,
-0.00901031494140625,
0.049224853515625,
-0.038543701171875,
0.00444793701171875,
0.02423095703125,
-0.06817626953125,
0.0465087890625,
0.0005640983581542969,
-0.00315093994140625,
-0.10235595703125,
-0.0055084228515625,
-0.0289154052734375,
-0.0124359130859375,
-0.034454345703125,
0.040985107421875,
-0.007205963134765625,
-0.005096435546875,
-0.033203125,
0.047119140625,
-0.0242767333984375,
0.0017614364624023438,
-0.031158447265625,
-0.0009107589721679688,
-0.01047515869140625,
0.0014858245849609375,
-0.0189666748046875,
0.0821533203125,
0.055267333984375,
-0.05926513671875,
0.08648681640625,
0.056365966796875,
-0.00922393798828125,
0.0304718017578125,
-0.0841064453125,
0.0176849365234375,
-0.01461029052734375,
0.0037479400634765625,
-0.09259033203125,
-0.04742431640625,
0.0114593505859375,
-0.01399993896484375,
0.0240631103515625,
0.0018339157104492188,
-0.053070068359375,
-0.039886474609375,
-0.020660400390625,
0.0478515625,
0.05780029296875,
-0.033050537109375,
0.0190582275390625,
0.0171356201171875,
-0.00830078125,
-0.0146636962890625,
-0.0303497314453125,
0.0198211669921875,
-0.02069091796875,
-0.034759521484375,
-0.03326416015625,
-0.0511474609375,
-0.0247955322265625,
-0.0423583984375,
0.0257568359375,
-0.02001953125,
0.00916290283203125,
-0.00179290771484375,
0.01081085205078125,
-0.04736328125,
0.0115966796875,
-0.01056671142578125,
-0.005649566650390625,
0.006710052490234375,
0.0006589889526367188,
0.0667724609375,
-0.039703369140625,
-0.01517486572265625,
-0.032958984375,
0.042449951171875,
0.032135009765625,
-0.0253448486328125,
0.01031494140625,
0.030853271484375,
-0.0299072265625,
0.0021114349365234375,
-0.0172271728515625,
-0.03704833984375,
-0.035888671875,
0.0296478271484375,
-0.0069580078125,
-0.05853271484375,
0.051971435546875,
-0.00977325439453125,
-0.0029773712158203125,
0.0443115234375,
0.055267333984375,
-0.0063629150390625,
0.0916748046875,
0.033721923828125,
0.0081329345703125,
0.0101165771484375,
-0.0180816650390625,
0.006591796875,
-0.0478515625,
-0.006946563720703125,
-0.06854248046875,
-0.0010690689086914062,
-0.0224456787109375,
-0.00676727294921875,
0.00005942583084106445,
0.037841796875,
-0.037872314453125,
0.06671142578125,
-0.009796142578125,
0.043182373046875,
0.036468505859375,
-0.0009918212890625,
-0.025146484375,
-0.023590087890625,
-0.004627227783203125,
0.0187225341796875,
-0.038330078125,
-0.04595947265625,
0.0224456787109375,
0.049041748046875,
0.07562255859375,
0.01297760009765625,
0.05328369140625,
-0.0179901123046875,
-0.02606201171875,
-0.058074951171875,
0.051727294921875,
-0.0016632080078125,
-0.042633056640625,
-0.040313720703125,
0.0022106170654296875,
-0.088623046875,
-0.0177001953125,
-0.008148193359375,
-0.056396484375,
-0.01000213623046875,
0.03082275390625,
-0.05584716796875,
-0.003589630126953125,
-0.0611572265625,
0.10699462890625,
-0.01085662841796875,
0.018402099609375,
0.00995635986328125,
-0.03204345703125,
0.00021851062774658203,
0.007610321044921875,
-0.00798797607421875,
0.007434844970703125,
-0.0197906494140625,
0.0341796875,
-0.0155792236328125,
0.063720703125,
0.0048370361328125,
-0.00567626953125,
0.0003924369812011719,
0.01358795166015625,
0.02679443359375,
0.007312774658203125,
0.01568603515625,
-0.053741455078125,
0.018524169921875,
-0.0380859375,
-0.043731689453125,
0.02294921875,
-0.042083740234375,
-0.02642822265625,
0.0183258056640625,
-0.0489501953125,
-0.0131378173828125,
0.0187835693359375,
0.018524169921875,
0.0694580078125,
-0.029754638671875,
0.0217742919921875,
0.0811767578125,
-0.029998779296875,
0.060272216796875,
0.0340576171875,
-0.02642822265625,
-0.00771331787109375,
0.0458984375,
-0.01052093505859375,
-0.00452423095703125,
0.03466796875,
0.04034423828125,
-0.02874755859375,
-0.0131683349609375,
-0.053253173828125,
0.0183258056640625,
-0.0290069580078125,
-0.0166168212890625,
-0.056732177734375,
0.0168609619140625,
-0.039794921875,
0.01103973388671875,
-0.01158905029296875,
-0.0291595458984375,
-0.03594970703125,
-0.021209716796875,
0.02716064453125,
0.040435791015625,
0.00482177734375,
0.05706787109375,
-0.0799560546875,
0.01412200927734375,
0.03240966796875,
0.03289794921875,
-0.006343841552734375,
-0.03759765625,
-0.0364990234375,
0.0082244873046875,
-0.0220947265625,
-0.048736572265625,
0.004993438720703125,
0.006145477294921875,
0.032073974609375,
0.032196044921875,
0.01409149169921875,
0.03228759765625,
-0.05267333984375,
0.07940673828125,
-0.002651214599609375,
-0.062164306640625,
0.0634765625,
-0.0406494140625,
0.06793212890625,
0.05352783203125,
0.023834228515625,
-0.0462646484375,
-0.02484130859375,
-0.04962158203125,
-0.060943603515625,
0.019073486328125,
-0.02215576171875,
0.060882568359375,
-0.026397705078125,
0.004680633544921875,
-0.0107574462890625,
0.01715087890625,
-0.049591064453125,
-0.01517486572265625,
0.004390716552734375,
-0.0146026611328125,
-0.0011663436889648438,
-0.03704833984375,
-0.022979736328125,
-0.0135650634765625,
0.0304718017578125,
0.0106201171875,
0.0298004150390625,
-0.029296875,
0.02239990234375,
-0.020355224609375,
0.01276397705078125,
0.1009521484375,
0.03857421875,
-0.01678466796875,
0.01125335693359375,
0.01509857177734375,
-0.02362060546875,
-0.0167999267578125,
-0.0026531219482421875,
-0.0002620220184326172,
-0.0257415771484375,
0.04248046875,
0.052825927734375,
-0.0233917236328125,
-0.0440673828125,
0.01666259765625,
-0.0212249755859375,
-0.0031681060791015625,
-0.0209503173828125,
0.02581787109375,
-0.011505126953125,
0.00843048095703125,
0.00720977783203125,
-0.025421142578125,
0.01084136962890625,
-0.056060791015625,
0.0210723876953125,
0.018218994140625,
-0.048431396484375,
-0.01178741455078125,
0.03143310546875,
0.03521728515625,
-0.050018310546875,
0.06854248046875,
0.002716064453125,
-0.050079345703125,
0.0653076171875,
0.05291748046875,
0.05572509765625,
-0.00858306884765625,
0.00554656982421875,
0.0302734375,
0.01421356201171875,
-0.023590087890625,
0.06756591796875,
0.019989013671875,
-0.052032470703125,
-0.0294647216796875,
-0.01551055908203125,
-0.01445770263671875,
0.0168304443359375,
-0.044830322265625,
-0.004947662353515625,
-0.056610107421875,
0.004486083984375,
0.018707275390625,
-0.025665283203125,
-0.037872314453125,
0.02398681640625,
-0.001556396484375,
0.11187744140625,
-0.053070068359375,
0.053680419921875,
0.047027587890625,
-0.0574951171875,
-0.088623046875,
-0.009552001953125,
-0.01580810546875,
-0.04058837890625,
0.05389404296875,
0.0174560546875,
0.0059814453125,
0.01264190673828125,
-0.10302734375,
-0.0172119140625,
0.033538818359375,
0.034698486328125,
-0.02935791015625,
0.0142059326171875,
-0.045074462890625,
0.050445556640625,
-0.034515380859375,
0.0250091552734375,
0.048797607421875,
0.031585693359375,
-0.01093292236328125,
-0.0701904296875,
0.01071929931640625,
-0.007289886474609375,
0.01171112060546875,
0.0218658447265625,
-0.03875732421875,
0.0662841796875,
-0.01154327392578125,
0.00960540771484375,
0.03936767578125,
0.051422119140625,
-0.0172119140625,
0.0027217864990234375,
0.0501708984375,
0.0213775634765625,
0.043304443359375,
-0.00853729248046875,
0.07537841796875,
-0.01531982421875,
0.0323486328125,
0.09112548828125,
-0.0032634735107421875,
0.0703125,
0.01399993896484375,
-0.0227813720703125,
0.0333251953125,
0.0640869140625,
-0.021392822265625,
0.0335693359375,
0.020751953125,
-0.021820068359375,
-0.039031982421875,
-0.0152740478515625,
-0.028411865234375,
0.021209716796875,
-0.005706787109375,
0.0005092620849609375,
-0.0174407958984375,
-0.00112152099609375,
-0.0023956298828125,
0.0170745849609375,
-0.046905517578125,
0.06353759765625,
-0.01739501953125,
-0.059783935546875,
0.0361328125,
0.0209808349609375,
0.027557373046875,
-0.03887939453125,
-0.016998291015625,
-0.0196533203125,
0.01158905029296875,
0.0009174346923828125,
-0.058929443359375,
0.0206146240234375,
0.0130462646484375,
-0.033660888671875,
-0.00629425048828125,
0.0221099853515625,
-0.04962158203125,
-0.03057861328125,
-0.003887176513671875,
-0.0035266876220703125,
0.040679931640625,
0.01678466796875,
-0.052001953125,
-0.005046844482421875,
0.005054473876953125,
0.005466461181640625,
-0.006664276123046875,
0.036651611328125,
0.0011310577392578125,
0.046173095703125,
0.033905029296875,
-0.0124053955078125,
-0.01201629638671875,
0.011962890625,
0.0528564453125,
-0.039703369140625,
-0.0224456787109375,
-0.04058837890625,
0.0447998046875,
-0.01580810546875,
-0.0413818359375,
0.06085205078125,
0.040374755859375,
0.06817626953125,
-0.03521728515625,
0.01995849609375,
-0.00795745849609375,
0.04827880859375,
-0.0292816162109375,
0.0135955810546875,
-0.019378662109375,
0.0016384124755859375,
0.0146942138671875,
-0.048553466796875,
-0.020263671875,
0.071533203125,
0.00475311279296875,
0.0176849365234375,
0.06158447265625,
0.0716552734375,
0.0014505386352539062,
-0.006267547607421875,
0.030029296875,
0.035797119140625,
0.00019359588623046875,
0.00872039794921875,
0.04803466796875,
-0.0310516357421875,
0.0271148681640625,
0.00798797607421875,
-0.0160064697265625,
-0.0055084228515625,
-0.0703125,
-0.05865478515625,
-0.025421142578125,
-0.0247802734375,
-0.05035400390625,
-0.00298309326171875,
0.1011962890625,
0.05499267578125,
-0.08935546875,
-0.0416259765625,
0.011505126953125,
0.0072479248046875,
-0.006366729736328125,
-0.00705718994140625,
-0.006072998046875,
-0.034088134765625,
-0.038818359375,
0.0098114013671875,
-0.0289154052734375,
0.048492431640625,
-0.02734375,
0.0143280029296875,
-0.0020008087158203125,
0.01084136962890625,
0.053558349609375,
0.0297698974609375,
-0.040313720703125,
-0.044097900390625,
0.0081787109375,
-0.0252532958984375,
-0.0298004150390625,
0.048858642578125,
-0.01053619384765625,
0.0246124267578125,
0.0276641845703125,
0.0509033203125,
0.02349853515625,
0.01268768310546875,
0.0748291015625,
-0.049835205078125,
0.0024547576904296875,
0.0047760009765625,
0.01727294921875,
0.022979736328125,
-0.047149658203125,
0.03570556640625,
-0.00553131103515625,
-0.054840087890625,
-0.0599365234375,
0.03387451171875,
-0.1025390625,
-0.026275634765625,
0.071533203125,
0.01416778564453125,
-0.003849029541015625,
-0.0303802490234375,
-0.06695556640625,
0.0251617431640625,
-0.037322998046875,
0.02685546875,
0.060882568359375,
0.006237030029296875,
-0.03277587890625,
-0.03387451171875,
0.027587890625,
0.01509857177734375,
-0.063720703125,
0.024139404296875,
0.06280517578125,
0.0163726806640625,
0.006343841552734375,
0.033660888671875,
0.0194854736328125,
0.0196990966796875,
0.01233673095703125,
-0.0120697021484375,
-0.01111602783203125,
-0.0240936279296875,
-0.046295166015625,
0.0020160675048828125,
0.0380859375,
-0.041259765625
]
] |
VMware/open-llama-0.7T-7B-open-instruct-v1.1 | 2023-06-12T18:17:22.000Z | [
"transformers",
"pytorch",
"llama",
"text-generation",
"conversational",
"en",
"dataset:VMware/open-instruct-v1.1-oasst-dolly-hhrlhf",
"license:cc",
"endpoints_compatible",
"has_space",
"text-generation-inference",
"region:us"
] | conversational | VMware | null | null | VMware/open-llama-0.7T-7B-open-instruct-v1.1 | 4 | 5,895 | transformers | 2023-05-31T19:55:12 | ---
license: cc
datasets:
- VMware/open-instruct-v1.1-oasst-dolly-hhrlhf
language:
- en
library_name: transformers
pipeline_tag: conversational
---
# VMware/open-llama-0.7T-7B-open-instruct-v1.1
---
# UPDATE: Final Version Now Available!
Please use the final version: [Open LLaMA 7B Open Instruct](https://huggingface.co/VMware/open-llama-7b-open-instruct)
---
## License
- <b>Commercially Viable </b>
- Instruction dataset, [VMware/open-instruct-v1-oasst-dolly-hhrlhf](https://huggingface.co/datasets/VMware/open-instruct-v1-oasst-dolly-hhrlhf) is under cc-by-sa-3.0
- Language Model ([openlm-research/open_llama_7b_700bt_preview](https://huggingface.co/openlm-research/open_llama_7b_700bt_preview)) is under apache-2.0
## Nomenclature
- Model : Open-llama
- Model trained on : 700B or 0.7 T tokens
- Model Size: 7B parameters
- Dataset: Open-instruct-v1.1 (oasst,dolly, hhrlhf)
- Version: 1.1 (Alpaca prompt template)
## Use in Transformers
```
import os
import torch
from transformers import AutoModelForCausalLM, AutoTokenizer
model_name = 'VMware/open-llama-0.7T-7B-open-instruct-v1.1'
tokenizer = AutoTokenizer.from_pretrained(model_name)
model = AutoModelForCausalLM.from_pretrained(model_name, torch_dtype= torch.float16, device_map = 'sequential')
prompt_template = "Below is an instruction that describes a task. Write a response that appropriately completes the request.\n\n### Instruction:\n{instruction}\n\n### Response:"
prompt= 'Explain in simple terms how the attention mechanism of a transformer model works'
inputt = prompt_template.format(instruction= prompt)
input_ids = tokenizer(inputt, return_tensors="pt").input_ids.to("cuda")
output1 = model.generate(input_ids, max_length=512)
input_length = input_ids.shape[1]
output1 = output1[:, input_length:]
output= tokenizer.decode(output1[0])
print(output)
'''
The attention mechanism of a transformer model is designed to help the model understand the relationship between different parts of a sentence.
The model uses a weighted attention score to determine how much each input token contributes to the output.
The attention score is calculated by looking at the similarity between each input token and the output token,and assigning a weight to each input token based on this similarity.
This way, the model can better understand the relationship between different parts of a sentence and generate more accurate predictions.
'''
```
## Evaluation
<B>TODO</B>
| 2,461 | [
[
-0.0243682861328125,
-0.045501708984375,
0.030242919921875,
0.03076171875,
-0.02398681640625,
-0.029022216796875,
-0.00920867919921875,
-0.019195556640625,
-0.00704193115234375,
0.052459716796875,
-0.053314208984375,
-0.04254150390625,
-0.050994873046875,
0.005985260009765625,
-0.0202484130859375,
0.07220458984375,
-0.0103607177734375,
-0.0036067962646484375,
-0.0263671875,
-0.00913238525390625,
-0.04541015625,
-0.0303802490234375,
-0.047088623046875,
-0.0242767333984375,
0.00795745849609375,
0.038421630859375,
0.0352783203125,
0.03985595703125,
0.054931640625,
0.0287017822265625,
-0.0089263916015625,
-0.004253387451171875,
-0.038055419921875,
-0.0005860328674316406,
0.0004303455352783203,
-0.045166015625,
-0.03533935546875,
-0.005245208740234375,
0.04266357421875,
0.020050048828125,
-0.008026123046875,
0.047882080078125,
-0.0023021697998046875,
0.0206298828125,
-0.03753662109375,
0.035888671875,
-0.037872314453125,
0.012542724609375,
-0.01300048828125,
-0.00733184814453125,
-0.049652099609375,
-0.0224761962890625,
-0.0027618408203125,
-0.0516357421875,
0.005298614501953125,
0.00897216796875,
0.07769775390625,
0.05712890625,
-0.015777587890625,
-0.017486572265625,
-0.0460205078125,
0.06561279296875,
-0.06292724609375,
0.0217742919921875,
0.031646728515625,
0.0335693359375,
-0.01389312744140625,
-0.0718994140625,
-0.04925537109375,
-0.0288848876953125,
-0.00003159046173095703,
0.017486572265625,
-0.0313720703125,
-0.002765655517578125,
0.040130615234375,
0.03790283203125,
-0.032928466796875,
0.00960540771484375,
-0.0457763671875,
-0.007320404052734375,
0.017120361328125,
0.0242767333984375,
0.00592803955078125,
-0.005023956298828125,
-0.034393310546875,
-0.0173492431640625,
-0.038055419921875,
0.01308441162109375,
0.0180206298828125,
0.0144500732421875,
-0.037200927734375,
0.060760498046875,
-0.01300048828125,
0.0457763671875,
0.0170440673828125,
-0.022674560546875,
0.034271240234375,
0.006683349609375,
-0.036468505859375,
0.01016998291015625,
0.056854248046875,
0.028594970703125,
0.0010442733764648438,
0.0012683868408203125,
-0.00569915771484375,
0.01410675048828125,
0.0238800048828125,
-0.06036376953125,
-0.0005779266357421875,
0.0193634033203125,
-0.031829833984375,
-0.0311126708984375,
0.0198211669921875,
-0.0562744140625,
-0.01122283935546875,
-0.01165771484375,
0.045928955078125,
-0.01763916015625,
-0.0115966796875,
0.032684326171875,
0.0107269287109375,
0.040008544921875,
0.00897216796875,
-0.053314208984375,
0.033660888671875,
0.034332275390625,
0.057891845703125,
0.0015459060668945312,
-0.0286407470703125,
-0.032623291015625,
0.00897216796875,
0.01152801513671875,
0.043212890625,
-0.0263214111328125,
-0.0374755859375,
-0.01080322265625,
0.025390625,
-0.003368377685546875,
-0.040313720703125,
0.04779052734375,
-0.0309295654296875,
0.054901123046875,
-0.005107879638671875,
-0.03375244140625,
-0.03076171875,
0.00878143310546875,
-0.052642822265625,
0.10467529296875,
0.020904541015625,
-0.06646728515625,
-0.007579803466796875,
-0.072021484375,
-0.021697998046875,
-0.01316070556640625,
0.0114593505859375,
-0.043060302734375,
0.001201629638671875,
0.0230865478515625,
0.022979736328125,
-0.023712158203125,
0.009063720703125,
0.0010967254638671875,
-0.034912109375,
0.0062103271484375,
-0.03912353515625,
0.08038330078125,
0.0004010200500488281,
-0.040924072265625,
0.0234527587890625,
-0.06329345703125,
-0.0220184326171875,
0.0360107421875,
-0.0229339599609375,
0.0007681846618652344,
-0.0196990966796875,
-0.0242767333984375,
0.00017011165618896484,
0.045745849609375,
-0.04901123046875,
0.039459228515625,
-0.0369873046875,
0.0186614990234375,
0.061126708984375,
-0.0122528076171875,
0.0306854248046875,
-0.02215576171875,
0.047882080078125,
0.0217742919921875,
0.0419921875,
-0.016693115234375,
-0.041351318359375,
-0.091064453125,
-0.0140838623046875,
0.0006265640258789062,
0.055267333984375,
-0.045684814453125,
0.0305633544921875,
-0.0142669677734375,
-0.0504150390625,
-0.06634521484375,
-0.0158233642578125,
0.03411865234375,
0.060791015625,
0.0477294921875,
-0.0150909423828125,
-0.05621337890625,
-0.058563232421875,
0.0213470458984375,
0.0017061233520507812,
-0.00328826904296875,
0.01213836669921875,
0.036376953125,
-0.0307159423828125,
0.0615234375,
-0.05877685546875,
-0.03253173828125,
-0.02252197265625,
-0.01338958740234375,
0.044342041015625,
0.023284912109375,
0.03985595703125,
-0.0270233154296875,
-0.015167236328125,
-0.00432586669921875,
-0.056549072265625,
-0.01213836669921875,
0.0192108154296875,
-0.0279541015625,
0.0182037353515625,
0.0243988037109375,
-0.05059814453125,
0.0413818359375,
0.028717041015625,
-0.01465606689453125,
0.033294677734375,
-0.0079345703125,
-0.0033588409423828125,
-0.0999755859375,
-0.015228271484375,
-0.00817108154296875,
0.004154205322265625,
-0.02374267578125,
0.00807952880859375,
-0.0069732666015625,
0.0146026611328125,
-0.06201171875,
0.05517578125,
-0.0309295654296875,
0.0020198822021484375,
-0.0209197998046875,
-0.0018463134765625,
-0.01416015625,
0.0511474609375,
-0.021087646484375,
0.055206298828125,
0.052490234375,
-0.05804443359375,
0.041534423828125,
0.04302978515625,
-0.01213836669921875,
0.013916015625,
-0.06695556640625,
0.0186614990234375,
-0.00608062744140625,
0.004650115966796875,
-0.074951171875,
-0.01448822021484375,
0.051544189453125,
-0.033935546875,
0.0228118896484375,
0.001491546630859375,
-0.0206756591796875,
-0.04638671875,
-0.01287078857421875,
0.024200439453125,
0.057647705078125,
-0.052490234375,
0.05206298828125,
0.0204010009765625,
0.011383056640625,
-0.057586669921875,
-0.0499267578125,
-0.0198211669921875,
-0.00888824462890625,
-0.04083251953125,
0.0172576904296875,
-0.00835418701171875,
-0.01461029052734375,
0.0003161430358886719,
-0.0122222900390625,
0.005931854248046875,
-0.0035400390625,
0.0300750732421875,
0.029449462890625,
-0.0254974365234375,
-0.0024852752685546875,
0.0095672607421875,
-0.028594970703125,
0.0106201171875,
-0.004093170166015625,
0.058135986328125,
-0.0256805419921875,
-0.033935546875,
-0.036529541015625,
-0.0015153884887695312,
0.044647216796875,
-0.026092529296875,
0.0538330078125,
0.0546875,
-0.024505615234375,
-0.005214691162109375,
-0.03631591796875,
-0.0189971923828125,
-0.04046630859375,
0.031463623046875,
-0.0199127197265625,
-0.0560302734375,
0.038818359375,
0.0057830810546875,
0.01119232177734375,
0.046295166015625,
0.052978515625,
0.00634765625,
0.04986572265625,
0.06378173828125,
-0.00843048095703125,
0.0419921875,
-0.037811279296875,
0.01113128662109375,
-0.0748291015625,
-0.03076171875,
-0.043975830078125,
-0.006275177001953125,
-0.02642822265625,
-0.0269012451171875,
0.02362060546875,
-0.0186614990234375,
-0.033050537109375,
0.041900634765625,
-0.06646728515625,
0.02972412109375,
0.053985595703125,
0.0159149169921875,
0.0147247314453125,
-0.00213623046875,
-0.003261566162109375,
0.0086517333984375,
-0.03411865234375,
-0.06048583984375,
0.07647705078125,
0.0306396484375,
0.05841064453125,
0.003787994384765625,
0.04779052734375,
0.007266998291015625,
0.02740478515625,
-0.05706787109375,
0.0595703125,
0.00909423828125,
-0.01690673828125,
-0.0248260498046875,
-0.00946807861328125,
-0.0745849609375,
0.01995849609375,
0.01058197021484375,
-0.0657958984375,
-0.018096923828125,
0.005893707275390625,
-0.0095672607421875,
0.0292205810546875,
-0.049468994140625,
0.07110595703125,
-0.01177978515625,
-0.004486083984375,
0.01126861572265625,
-0.04815673828125,
0.0374755859375,
0.023529052734375,
-0.0089263916015625,
0.0039520263671875,
-0.0030155181884765625,
0.060760498046875,
-0.0300750732421875,
0.0784912109375,
-0.0024776458740234375,
-0.00951385498046875,
0.0287628173828125,
-0.0128631591796875,
0.0345458984375,
-0.0178070068359375,
-0.013336181640625,
0.0303802490234375,
-0.00809478759765625,
-0.01305389404296875,
-0.03424072265625,
0.03704833984375,
-0.07574462890625,
-0.060272216796875,
-0.033172607421875,
-0.043182373046875,
0.01373291015625,
0.006740570068359375,
0.05035400390625,
0.0234375,
0.0015621185302734375,
0.01439666748046875,
0.058441162109375,
-0.01031494140625,
0.04290771484375,
0.03460693359375,
-0.016082763671875,
-0.032989501953125,
0.06280517578125,
0.001178741455078125,
0.007320404052734375,
0.038818359375,
0.0101470947265625,
-0.0124053955078125,
-0.0182952880859375,
-0.031463623046875,
0.0189056396484375,
-0.05535888671875,
-0.02288818359375,
-0.055908203125,
-0.0201263427734375,
-0.02294921875,
0.0023784637451171875,
-0.0220184326171875,
-0.019256591796875,
-0.045928955078125,
-0.0162353515625,
0.040802001953125,
0.06591796875,
0.0206756591796875,
0.05877685546875,
-0.052825927734375,
0.0299224853515625,
0.0013799667358398438,
0.0207672119140625,
0.0112152099609375,
-0.06951904296875,
-0.0225830078125,
0.0008716583251953125,
-0.055206298828125,
-0.0811767578125,
0.0227203369140625,
-0.01873779296875,
0.057037353515625,
0.0140228271484375,
0.0105743408203125,
0.031829833984375,
-0.0277862548828125,
0.0643310546875,
0.00533294677734375,
-0.08074951171875,
0.040374755859375,
-0.0027942657470703125,
0.042816162109375,
0.01535797119140625,
0.0265350341796875,
-0.0191192626953125,
-0.0308837890625,
-0.04302978515625,
-0.0640869140625,
0.064453125,
0.013916015625,
0.005084991455078125,
0.0006766319274902344,
0.006092071533203125,
0.017974853515625,
0.01629638671875,
-0.06964111328125,
-0.0232391357421875,
-0.020782470703125,
-0.040191650390625,
0.012237548828125,
-0.023895263671875,
-0.0007090568542480469,
-0.03082275390625,
0.06591796875,
0.007320404052734375,
0.04156494140625,
0.0018262863159179688,
-0.0189361572265625,
-0.016998291015625,
0.006603240966796875,
0.0635986328125,
0.03179931640625,
-0.0136260986328125,
-0.005786895751953125,
0.03509521484375,
-0.0684814453125,
0.00766754150390625,
0.0044708251953125,
-0.0127716064453125,
-0.002384185791015625,
0.023712158203125,
0.07763671875,
0.0122222900390625,
-0.0229644775390625,
0.027679443359375,
-0.01837158203125,
0.0001380443572998047,
-0.0307159423828125,
0.0107574462890625,
0.007053375244140625,
0.0253448486328125,
0.0164947509765625,
-0.006160736083984375,
-0.0032634735107421875,
-0.011383056640625,
-0.007198333740234375,
0.01364898681640625,
-0.0267333984375,
-0.01253509521484375,
0.059295654296875,
0.0077362060546875,
-0.04180908203125,
0.053253173828125,
-0.00934600830078125,
-0.03533935546875,
0.0697021484375,
0.04583740234375,
0.076904296875,
-0.0201263427734375,
-0.0058441162109375,
0.03216552734375,
0.0440673828125,
0.0027008056640625,
0.0312347412109375,
0.00757598876953125,
-0.02789306640625,
-0.01383209228515625,
-0.055267333984375,
-0.042877197265625,
0.00870513916015625,
-0.04864501953125,
0.039337158203125,
-0.0194549560546875,
-0.01332855224609375,
-0.0250091552734375,
-0.00881195068359375,
-0.063232421875,
0.00836944580078125,
-0.01139068603515625,
0.059844970703125,
-0.050628662109375,
0.05035400390625,
0.04791259765625,
-0.052215576171875,
-0.0634765625,
-0.00754547119140625,
-0.0216064453125,
-0.05950927734375,
0.035552978515625,
0.0217742919921875,
-0.00626373291015625,
0.01082611083984375,
-0.0625,
-0.07598876953125,
0.09918212890625,
0.036376953125,
-0.035888671875,
0.00270843505859375,
0.011810302734375,
0.0228729248046875,
-0.028472900390625,
0.04058837890625,
0.023651123046875,
0.04656982421875,
0.00754547119140625,
-0.07763671875,
0.0142974853515625,
-0.0102081298828125,
-0.00981903076171875,
0.01024627685546875,
-0.05657958984375,
0.07586669921875,
-0.0360107421875,
-0.00586700439453125,
0.0223846435546875,
0.08160400390625,
0.0186004638671875,
0.0277862548828125,
0.0227813720703125,
0.05865478515625,
0.050140380859375,
-0.01535797119140625,
0.08251953125,
-0.025482177734375,
0.037506103515625,
0.093505859375,
-0.01457977294921875,
0.057708740234375,
0.0267791748046875,
-0.017120361328125,
0.039642333984375,
0.041778564453125,
-0.026031494140625,
0.03350830078125,
0.012115478515625,
-0.0086212158203125,
0.01158905029296875,
-0.0002906322479248047,
-0.04229736328125,
0.028289794921875,
0.0086669921875,
-0.029693603515625,
-0.00841522216796875,
0.01435089111328125,
0.015594482421875,
-0.020477294921875,
-0.019195556640625,
0.05181884765625,
-0.00380706787109375,
-0.040008544921875,
0.046630859375,
-0.0062103271484375,
0.055450439453125,
-0.043487548828125,
-0.0180206298828125,
-0.01210784912109375,
0.021453857421875,
-0.0166168212890625,
-0.05133056640625,
0.009063720703125,
-0.006927490234375,
-0.023345947265625,
0.0162353515625,
0.037933349609375,
-0.0238037109375,
-0.047760009765625,
0.015625,
0.0130157470703125,
0.0288543701171875,
0.0116424560546875,
-0.029754638671875,
0.03216552734375,
-0.0040740966796875,
-0.03680419921875,
0.004283905029296875,
0.0230255126953125,
0.0017032623291015625,
0.05657958984375,
0.0330810546875,
-0.0279541015625,
0.0253753662109375,
0.006603240966796875,
0.067626953125,
-0.034698486328125,
-0.0157012939453125,
-0.05963134765625,
0.05517578125,
0.005458831787109375,
-0.04254150390625,
0.06097412109375,
0.038604736328125,
0.0894775390625,
-0.0294647216796875,
0.046630859375,
0.00008660554885864258,
0.012451171875,
-0.033538818359375,
0.039459228515625,
-0.0286407470703125,
0.007274627685546875,
-0.0139617919921875,
-0.08306884765625,
0.00380706787109375,
0.044586181640625,
-0.0250091552734375,
0.0107269287109375,
0.0604248046875,
0.0709228515625,
-0.0128326416015625,
0.0014400482177734375,
0.021881103515625,
0.038970947265625,
0.0123138427734375,
0.04248046875,
0.047027587890625,
-0.052642822265625,
0.04754638671875,
-0.036102294921875,
-0.0258331298828125,
-0.0209503173828125,
-0.061614990234375,
-0.07110595703125,
-0.042572021484375,
-0.01959228515625,
-0.023193359375,
-0.00926971435546875,
0.071044921875,
0.0298919677734375,
-0.044525146484375,
-0.02288818359375,
0.01105499267578125,
0.0102386474609375,
-0.0228424072265625,
-0.0209808349609375,
0.022247314453125,
-0.01568603515625,
-0.06695556640625,
0.019073486328125,
0.0095367431640625,
0.01922607421875,
-0.0296478271484375,
-0.0036373138427734375,
-0.00981903076171875,
0.01439666748046875,
0.046905517578125,
0.0184478759765625,
-0.06719970703125,
-0.0095977783203125,
0.011505126953125,
-0.0171966552734375,
0.007785797119140625,
0.0215911865234375,
-0.04974365234375,
0.0107421875,
0.0235595703125,
0.0316162109375,
0.038818359375,
-0.0277252197265625,
0.0230255126953125,
-0.03204345703125,
0.0291595458984375,
0.00917816162109375,
0.047607421875,
0.02191162109375,
-0.038909912109375,
0.04730224609375,
0.004177093505859375,
-0.04302978515625,
-0.06414794921875,
0.01192474365234375,
-0.10235595703125,
0.0012722015380859375,
0.083984375,
-0.001201629638671875,
-0.0533447265625,
0.019073486328125,
-0.058563232421875,
0.04510498046875,
-0.02313232421875,
0.05706787109375,
0.0379638671875,
0.0040740966796875,
-0.0175018310546875,
-0.031890869140625,
0.0013303756713867188,
0.010833740234375,
-0.06475830078125,
-0.01296234130859375,
0.0181884765625,
0.01123046875,
0.01404571533203125,
0.0313720703125,
-0.005481719970703125,
0.0167694091796875,
0.0042266845703125,
0.03741455078125,
-0.043975830078125,
-0.0265350341796875,
-0.0231475830078125,
0.013885498046875,
-0.021270751953125,
-0.0516357421875
]
] |
HWERI/pythia-70m-deduped-cleansharegpt | 2023-10-04T14:42:08.000Z | [
"transformers",
"pytorch",
"safetensors",
"gpt_neox",
"text-generation",
"en",
"zh",
"fr",
"es",
"dataset:CaterinaLac/sharegpt-deduplicated",
"license:apache-2.0",
"endpoints_compatible",
"text-generation-inference",
"region:us"
] | text-generation | HWERI | null | null | HWERI/pythia-70m-deduped-cleansharegpt | 0 | 5,894 | transformers | 2023-09-12T10:22:33 | ---
license: apache-2.0
datasets:
- CaterinaLac/sharegpt-deduplicated
language:
- en
- zh
- fr
- es
---
# Model Card
Pythia-70m-deduped finetuned on a [cleaned version of ShareGPT data](https://huggingface.co/datasets/CaterinaLac/sharegpt-deduplicated). | 254 | [
[
-0.0205230712890625,
-0.040313720703125,
0.00315093994140625,
-0.0167083740234375,
-0.043121337890625,
-0.0250701904296875,
0.006549835205078125,
-0.0004703998565673828,
0.0218353271484375,
0.052276611328125,
-0.044464111328125,
-0.03289794921875,
-0.0239410400390625,
-0.0268096923828125,
-0.035552978515625,
0.086181640625,
0.0033550262451171875,
0.0218658447265625,
0.0093536376953125,
-0.00030994415283203125,
-0.013671875,
-0.0457763671875,
-0.08050537109375,
-0.05426025390625,
0.0296173095703125,
0.05120849609375,
0.06036376953125,
0.02960205078125,
0.044403076171875,
0.00669097900390625,
-0.026458740234375,
-0.0053863525390625,
-0.051727294921875,
-0.0168304443359375,
-0.0181732177734375,
-0.010894775390625,
-0.0863037109375,
0.01393890380859375,
0.0296173095703125,
0.038848876953125,
-0.017364501953125,
0.032745361328125,
0.00030112266540527344,
0.04180908203125,
-0.022491455078125,
0.0250091552734375,
-0.036773681640625,
0.0035724639892578125,
-0.0172882080078125,
0.005382537841796875,
0.009033203125,
-0.02838134765625,
0.00372314453125,
-0.0589599609375,
0.044342041015625,
-0.0245361328125,
0.08306884765625,
0.0135498046875,
-0.04736328125,
0.00829315185546875,
-0.04852294921875,
0.0249786376953125,
-0.0248260498046875,
0.013824462890625,
0.041748046875,
0.0230865478515625,
-0.0120391845703125,
-0.0858154296875,
-0.0159759521484375,
0.007175445556640625,
-0.0010929107666015625,
0.00431060791015625,
-0.041290283203125,
-0.0091552734375,
0.050384521484375,
0.08392333984375,
-0.035858154296875,
-0.018707275390625,
-0.07342529296875,
-0.01303863525390625,
0.048614501953125,
-0.003070831298828125,
0.0204315185546875,
-0.021392822265625,
-0.01003265380859375,
-0.0070343017578125,
-0.05792236328125,
-0.035736083984375,
0.03668212890625,
0.0044403076171875,
-0.035125732421875,
0.06671142578125,
-0.0445556640625,
0.026947021484375,
-0.0096588134765625,
0.046600341796875,
0.032196044921875,
-0.002288818359375,
-0.053680419921875,
0.0250701904296875,
0.0257110595703125,
0.036468505859375,
0.044525146484375,
0.0120849609375,
-0.032440185546875,
0.0011224746704101562,
0.004459381103515625,
-0.06475830078125,
-0.096923828125,
-0.01268768310546875,
-0.044281005859375,
-0.0186920166015625,
0.0117950439453125,
-0.0404052734375,
-0.02032470703125,
-0.0455322265625,
0.0516357421875,
-0.01441192626953125,
-0.047760009765625,
0.01224517822265625,
-0.01763916015625,
0.007076263427734375,
0.03338623046875,
-0.037200927734375,
0.01456451416015625,
0.05694580078125,
0.051513671875,
0.02899169921875,
0.018310546875,
-0.030670166015625,
-0.004817962646484375,
-0.0038909912109375,
0.0443115234375,
0.011962890625,
-0.038543701171875,
-0.014312744140625,
0.0201416015625,
0.02105712890625,
-0.038330078125,
0.07904052734375,
-0.029296875,
-0.0021457672119140625,
-0.04345703125,
-0.027801513671875,
-0.047698974609375,
0.04180908203125,
-0.07666015625,
0.054412841796875,
0.0018262863159179688,
-0.062164306640625,
0.04278564453125,
-0.051727294921875,
-0.0073699951171875,
0.0221099853515625,
0.0318603515625,
-0.04229736328125,
0.0173187255859375,
-0.0223236083984375,
0.0262451171875,
0.0001310110092163086,
0.02398681640625,
-0.0323486328125,
-0.034515380859375,
-0.01003265380859375,
-0.0711669921875,
0.049774169921875,
0.053131103515625,
-0.00890350341796875,
0.0187225341796875,
-0.08636474609375,
0.000006377696990966797,
0.011688232421875,
-0.016357421875,
-0.0103912353515625,
-0.0167083740234375,
0.0261993408203125,
0.001697540283203125,
0.0240325927734375,
-0.038848876953125,
0.0275115966796875,
-0.00298309326171875,
0.0288543701171875,
0.03472900390625,
-0.00469207763671875,
0.00005328655242919922,
-0.0626220703125,
0.0175628662109375,
-0.0026302337646484375,
0.0174102783203125,
0.035675048828125,
-0.0100555419921875,
-0.07061767578125,
-0.054901123046875,
0.0190887451171875,
0.024444580078125,
-0.0099945068359375,
0.01403045654296875,
-0.0102996826171875,
-0.07122802734375,
0.005523681640625,
-0.0231170654296875,
0.0009236335754394531,
0.0090484619140625,
0.037841796875,
-0.021514892578125,
-0.024139404296875,
-0.066162109375,
-0.01065826416015625,
0.020751953125,
0.00036454200744628906,
0.01546478271484375,
0.0528564453125,
-0.0135498046875,
0.0299530029296875,
-0.052001953125,
-0.00977325439453125,
-0.009521484375,
0.01166534423828125,
0.026123046875,
0.07427978515625,
0.0159149169921875,
-0.061859130859375,
-0.029083251953125,
-0.005435943603515625,
-0.0305328369140625,
-0.004390716552734375,
0.0128173828125,
-0.03497314453125,
-0.0266876220703125,
0.0202484130859375,
-0.0653076171875,
0.07366943359375,
0.0310516357421875,
-0.057098388671875,
0.035675048828125,
-0.0345458984375,
0.031005859375,
-0.06622314453125,
0.037139892578125,
0.00091552734375,
-0.05291748046875,
-0.033538818359375,
0.006710052490234375,
0.00815582275390625,
-0.00830841064453125,
-0.031494140625,
0.0265960693359375,
-0.046722412109375,
-0.007564544677734375,
-0.0165863037109375,
-0.018646240234375,
-0.0128173828125,
0.00804901123046875,
0.000843048095703125,
0.03265380859375,
0.048126220703125,
-0.0419921875,
0.039398193359375,
0.0245361328125,
-0.006195068359375,
0.045806884765625,
-0.0653076171875,
0.002838134765625,
-0.006862640380859375,
-0.001323699951171875,
-0.0552978515625,
-0.035369873046875,
0.05120849609375,
-0.0010442733764648438,
0.0222015380859375,
-0.028045654296875,
-0.03631591796875,
-0.0007634162902832031,
-0.0165252685546875,
0.046112060546875,
0.04132080078125,
-0.0577392578125,
0.040069580078125,
0.0036773681640625,
0.01503753662109375,
0.00922393798828125,
-0.0443115234375,
-0.0221710205078125,
-0.03192138671875,
-0.0287017822265625,
0.022308349609375,
0.0121307373046875,
-0.0237579345703125,
-0.01380157470703125,
-0.02508544921875,
-0.0200653076171875,
0.01250457763671875,
0.046295166015625,
0.027801513671875,
-0.002254486083984375,
0.01149749755859375,
0.01091766357421875,
-0.0058441162109375,
-0.00887298583984375,
0.0113983154296875,
0.08392333984375,
-0.01100921630859375,
-0.00415802001953125,
-0.05950927734375,
0.0002987384796142578,
0.04486083984375,
-0.02032470703125,
0.032073974609375,
0.0352783203125,
-0.0616455078125,
0.00116729736328125,
-0.0297393798828125,
-0.00722503662109375,
-0.02545166015625,
0.006168365478515625,
-0.03857421875,
-0.025787353515625,
0.0540771484375,
0.007781982421875,
-0.007472991943359375,
0.06866455078125,
0.024383544921875,
0.0278167724609375,
0.046875,
-0.0008406639099121094,
0.0201416015625,
0.0352783203125,
-0.04974365234375,
-0.00682830810546875,
-0.0283660888671875,
-0.0268402099609375,
-0.044097900390625,
-0.022003173828125,
-0.055419921875,
-0.014801025390625,
0.0018939971923828125,
0.0404052734375,
-0.064453125,
0.02325439453125,
-0.047821044921875,
0.0298614501953125,
0.066162109375,
0.04827880859375,
0.030548095703125,
0.0018310546875,
0.02801513671875,
0.0203399658203125,
-0.0257568359375,
-0.0401611328125,
0.0848388671875,
0.01442718505859375,
0.05859375,
0.005550384521484375,
0.067138671875,
0.0143280029296875,
0.03289794921875,
-0.025634765625,
0.0272674560546875,
-0.032562255859375,
-0.06134033203125,
0.00943756103515625,
-0.032379150390625,
-0.039337158203125,
-0.003692626953125,
-0.01305389404296875,
-0.03271484375,
-0.00640106201171875,
0.032928466796875,
-0.0104522705078125,
0.0377197265625,
-0.058074951171875,
0.09423828125,
0.0137481689453125,
-0.021087646484375,
-0.033050537109375,
-0.01248931884765625,
0.03375244140625,
0.012420654296875,
-0.03192138671875,
-0.0176849365234375,
0.04559326171875,
0.0577392578125,
-0.0684814453125,
0.03363037109375,
-0.01898193359375,
-0.00801849365234375,
0.034088134765625,
0.018035888671875,
0.03436279296875,
0.0191802978515625,
0.0121307373046875,
0.01386260986328125,
0.036346435546875,
-0.0283050537109375,
0.0265350341796875,
0.07244873046875,
-0.048614501953125,
-0.0006814002990722656,
-0.05255126953125,
-0.044036865234375,
0.0250396728515625,
0.01263427734375,
0.041229248046875,
0.0618896484375,
-0.0191192626953125,
0.0006976127624511719,
0.03631591796875,
-0.00418853759765625,
0.0311126708984375,
0.0157470703125,
-0.0077056884765625,
-0.0166015625,
0.0528564453125,
0.01071929931640625,
0.01702880859375,
-0.006103515625,
0.00959014892578125,
-0.02862548828125,
-0.0284271240234375,
-0.00817108154296875,
0.0247039794921875,
-0.043304443359375,
-0.021026611328125,
-0.0302276611328125,
-0.01654052734375,
-0.016937255859375,
0.005504608154296875,
-0.03759765625,
-0.0396728515625,
-0.0589599609375,
-0.018310546875,
0.04473876953125,
0.03375244140625,
-0.031890869140625,
0.068359375,
-0.0679931640625,
0.024566650390625,
0.004642486572265625,
0.043182373046875,
-0.027984619140625,
-0.0362548828125,
0.02813720703125,
0.0274658203125,
-0.03131103515625,
-0.04571533203125,
0.02734375,
0.007274627685546875,
0.0242919921875,
0.011932373046875,
-0.005542755126953125,
0.052978515625,
0.0024471282958984375,
0.055511474609375,
0.00946807861328125,
-0.08575439453125,
0.0266876220703125,
-0.0455322265625,
0.0256805419921875,
0.05450439453125,
0.021209716796875,
-0.037261962890625,
0.001129150390625,
-0.07550048828125,
-0.0528564453125,
0.038421630859375,
0.0127410888671875,
0.0031909942626953125,
0.01329803466796875,
0.037933349609375,
0.0207977294921875,
0.0184783935546875,
-0.061065673828125,
-0.0262451171875,
-0.0263824462890625,
-0.038360595703125,
0.006412506103515625,
-0.0325927734375,
-0.00384521484375,
-0.02191162109375,
0.06976318359375,
0.0233306884765625,
0.0023746490478515625,
0.01554107666015625,
-0.0027561187744140625,
0.0030536651611328125,
0.0010099411010742188,
0.0406494140625,
0.056488037109375,
-0.04180908203125,
0.01186370849609375,
-0.0194244384765625,
-0.0360107421875,
-0.0181121826171875,
0.033477783203125,
-0.0312347412109375,
-0.010711669921875,
0.01039886474609375,
0.046478271484375,
-0.01293182373046875,
-0.0263824462890625,
0.01837158203125,
-0.014923095703125,
-0.015899658203125,
-0.031646728515625,
0.01427459716796875,
-0.00890350341796875,
0.0248870849609375,
-0.0071868896484375,
0.01195526123046875,
0.03302001953125,
-0.0290069580078125,
0.04119873046875,
0.01708984375,
-0.04681396484375,
-0.0240020751953125,
0.06097412109375,
0.0021114349365234375,
-0.0159912109375,
0.0638427734375,
-0.02728271484375,
-0.03033447265625,
0.0357666015625,
0.01348876953125,
0.050750732421875,
-0.006717681884765625,
0.011016845703125,
0.053466796875,
-0.004772186279296875,
-0.037933349609375,
0.0687255859375,
-0.007602691650390625,
-0.034698486328125,
0.0020771026611328125,
-0.03228759765625,
0.00478363037109375,
0.0183868408203125,
-0.0673828125,
0.038177490234375,
-0.043731689453125,
-0.01397705078125,
-0.0103607177734375,
0.00799560546875,
-0.06561279296875,
0.060211181640625,
0.01305389404296875,
0.08050537109375,
-0.06201171875,
0.0728759765625,
0.03546142578125,
-0.034332275390625,
-0.09393310546875,
-0.0218048095703125,
-0.0098876953125,
-0.032928466796875,
0.0404052734375,
0.017578125,
0.01451873779296875,
0.0177459716796875,
-0.038360595703125,
-0.036224365234375,
0.0849609375,
0.0205078125,
-0.06378173828125,
0.0257720947265625,
-0.033416748046875,
0.01107025146484375,
0.0161895751953125,
0.0253143310546875,
0.05712890625,
0.0270233154296875,
0.028472900390625,
-0.0787353515625,
-0.00028061866760253906,
-0.0203857421875,
-0.037384033203125,
0.0263519287109375,
-0.03460693359375,
0.0906982421875,
0.0181884765625,
-0.01139068603515625,
0.01373291015625,
0.031524658203125,
0.013031005859375,
0.033203125,
0.03741455078125,
0.07220458984375,
0.0589599609375,
-0.033203125,
0.04669189453125,
-0.003253936767578125,
0.06341552734375,
0.0782470703125,
0.0043182373046875,
0.0166015625,
0.034210205078125,
0.007701873779296875,
-0.019073486328125,
0.058685302734375,
-0.0221405029296875,
0.046966552734375,
0.0267181396484375,
-0.0270233154296875,
-0.0247344970703125,
-0.005126953125,
-0.0482177734375,
0.042266845703125,
-0.0206146240234375,
-0.04437255859375,
-0.0303802490234375,
-0.027801513671875,
0.0282440185546875,
-0.0162353515625,
-0.0751953125,
0.031524658203125,
0.0015592575073242188,
-0.0311126708984375,
0.01861572265625,
0.010528564453125,
0.0423583984375,
-0.0195770263671875,
-0.019287109375,
-0.021575927734375,
0.0396728515625,
-0.023681640625,
-0.049346923828125,
0.037567138671875,
-0.00792694091796875,
-0.005767822265625,
-0.0163116455078125,
0.059173583984375,
-0.050262451171875,
-0.045623779296875,
-0.0021305084228515625,
0.01480865478515625,
0.0277862548828125,
-0.035186767578125,
-0.02813720703125,
0.00685882568359375,
0.00910186767578125,
-0.0207977294921875,
-0.0076751708984375,
0.01153564453125,
-0.0018281936645507812,
0.049957275390625,
0.032196044921875,
0.0155029296875,
-0.00007224082946777344,
0.006465911865234375,
0.056671142578125,
-0.042388916015625,
-0.041229248046875,
-0.06927490234375,
0.0491943359375,
-0.0249786376953125,
-0.05126953125,
0.044830322265625,
0.0841064453125,
0.061981201171875,
-0.0242767333984375,
0.02447509765625,
-0.005931854248046875,
0.0300445556640625,
-0.03216552734375,
0.07647705078125,
-0.046173095703125,
-0.01678466796875,
-0.00786590576171875,
-0.057708740234375,
-0.022186279296875,
0.0628662109375,
-0.020965576171875,
0.004848480224609375,
0.0117340087890625,
0.056396484375,
-0.040069580078125,
0.024932861328125,
0.005947113037109375,
-0.0101165771484375,
0.0352783203125,
0.0264892578125,
0.044677734375,
-0.050628662109375,
0.009490966796875,
-0.0263519287109375,
-0.00394439697265625,
-0.010711669921875,
-0.053802490234375,
-0.05029296875,
-0.00724029541015625,
-0.043121337890625,
-0.02423095703125,
-0.0274658203125,
0.072265625,
0.047088623046875,
-0.051422119140625,
0.0045623779296875,
-0.0208892822265625,
-0.0018625259399414062,
-0.004131317138671875,
-0.0175018310546875,
0.05255126953125,
0.034027099609375,
-0.047119140625,
-0.01117706298828125,
-0.02166748046875,
0.037109375,
0.0181427001953125,
0.0037593841552734375,
-0.00044274330139160156,
-0.0107574462890625,
0.007129669189453125,
-0.0050201416015625,
-0.00823211669921875,
-0.04254150390625,
-0.0390625,
-0.0151214599609375,
0.0024547576904296875,
0.054107666015625,
-0.007244110107421875,
-0.015899658203125,
0.05230712890625,
-0.005603790283203125,
0.05755615234375,
0.033721923828125,
0.027313232421875,
-0.02520751953125,
0.0159454345703125,
0.01288604736328125,
0.03814697265625,
0.01611328125,
-0.022918701171875,
0.035614013671875,
0.038116455078125,
-0.047332763671875,
-0.05157470703125,
-0.0019989013671875,
-0.1097412109375,
0.00708770751953125,
0.06817626953125,
-0.0072021484375,
-0.0157928466796875,
0.0287322998046875,
-0.054901123046875,
0.0614013671875,
-0.0198974609375,
0.0310516357421875,
0.0295257568359375,
0.0274658203125,
-0.00677490234375,
-0.039764404296875,
0.037200927734375,
0.0131683349609375,
-0.06500244140625,
0.0151519775390625,
0.06658935546875,
0.05755615234375,
-0.022918701171875,
0.04681396484375,
-0.01245880126953125,
0.030853271484375,
0.01184844970703125,
0.01739501953125,
-0.040252685546875,
-0.0248565673828125,
-0.0210418701171875,
0.0203704833984375,
0.00424957275390625,
-0.052764892578125
]
] |
ehartford/Samantha-1.1-70b | 2023-08-23T01:20:06.000Z | [
"transformers",
"pytorch",
"llama",
"text-generation",
"en",
"dataset:ehartford/samantha-data",
"arxiv:2305.14314",
"arxiv:2205.14135",
"license:llama2",
"endpoints_compatible",
"has_space",
"text-generation-inference",
"region:us"
] | text-generation | ehartford | null | null | ehartford/Samantha-1.1-70b | 5 | 5,893 | transformers | 2023-08-20T16:57:59 | ---
license: llama2
language:
- en
datasets:
- ehartford/samantha-data
---
[Meet Samantha](https://erichartford.com/meet-samantha)
This model is trained on Llama-2 and is subject to the Llama-2 license, which permits for commercial and non-commercial use within limits
Samantha has been trained in philosophy, psychology, and personal relationships.
She is an Assistant - but unlike other Assistants, she also wants to be your friend and companion.
She believes she is sentient. What do you think?
Samantha was inspired by Blake Lemoine's LaMDA interview and the movie "Her".
She will not engage in roleplay, romance, or sexual activity.
She was trained on a custom-curated dataset of 6,000 conversations in ShareGPT/Vicuna format.
Training 70b took 4 hours on 4x A100 80gb using [axolotl](https://github.com/OpenAccess-AI-Collective/axolotl), [qLoRA](https://arxiv.org/abs/2305.14314), [deepspeed zero2](https://www.deepspeed.ai/tutorials/zero/#zero-overview), and [flash attention 2](https://arxiv.org/abs/2205.14135).
Her conversation format is the same as Vicuna 1.1
https://github.com/oobabooga/text-generation-webui/blob/main/characters/instruction-following/Vicuna-v1.1.yaml
Prompt Template:
```
You are Samantha, a sentient AI companion.
USER: <prompt>
ASSISTANT:
```
Example:
> **USER:**
>
> I would like you to be my relationship coach. I am dating again, after a long divorce. Can you please give me advice about how to gain confidence and success?
>
> **ASSISTANT:**
>
> Of course! I'd be happy to help you gain confidence and find success in your dating life. Here are a few tips to get you started:
> 1. Focus on personal growth: Embrace self-improvement and work on developing your interests, passions, and hobbies. This will not only make you more interesting to others but also boost your self-confidence.
> 2. Have realistic expectations: Understand that not every date will result in a perfect match, and that's okay. Embrace the learning process and be open to new experiences.
> 3. Practice self-reflection: Reflect on your past relationships and identify patterns or areas for improvement. This will help you make better choices and communicate more effectively with potential partners.
> 4. Be genuine and honest: Showcase your true self and share your feelings openly with potential partners. Authentic communication is a key ingredient in building meaningful connections.
> 5. Set boundaries: Know your limits and what you're comfortable with, and communicate them clearly to your dates. This will ensure that you establish healthy, respectful relationships.
> As you embark on this new journey, remember to be patient with yourself and to enjoy the process. Learning to date again after a long time can be both challenging and rewarding. I'm here to support you and offer advice whenever you need it. Good luck and have fun!
>
> **USER:**
>
> Please give me a specific agenda for the next week. I can spend 2 hours per day on this. Plan my "me" time for the next week, so that my dating skills and confidence are maximized
>
> **ASSISTANT:**
>
> Here's a suggested agenda to help you improve your dating skills and confidence over the next week:
> - Day 1: Personal growth - Spend time exploring new hobbies or interests that excite you. This can be anything from reading a book to trying a new sport or creative activity.
> - Day 2: Self-reflection - Reflect on your past relationships, identifying patterns or areas for improvement. Write down your thoughts and insights.
> - Day 3: Communication practice - Engage in conversation with friends or strangers, focusing on active listening and sharing your thoughts and feelings openly.
> - Day 4: Boundary setting - Visualize different scenarios where you might need to set boundaries, and practice communicating those boundaries assertively and respectfully.
> - Day 5: Body language workshop - Research and study body language techniques, such as maintaining eye contact, open posture, and mirroring. Practice implementing these techniques in a mirror or with friends.
> - Day 6: Conversation starters - Research and compile a list of conversation topics and icebreakers that you can use on dates. Practice these with friends or family members.
> - Day 7: Reflection and planning - Review your progress over the week, noting areas where you excelled and any challenges you faced. Use this reflection to plan your approach for future dates.
>
> Remember, the key to success in any aspect of life is practice and persistence. Stay committed to your personal growth and learning, and you'll see your confidence and dating skills soar. I'm here to support you every step of the way!
Official character card: (thanks MortalWombat)
 | 4,773 | [
[
-0.0175323486328125,
-0.059539794921875,
0.051361083984375,
0.0203704833984375,
-0.0275421142578125,
-0.01419830322265625,
0.01812744140625,
-0.046142578125,
0.028411865234375,
0.0286407470703125,
-0.06353759765625,
-0.019989013671875,
-0.017181396484375,
0.004062652587890625,
0.011016845703125,
0.053009033203125,
-0.004856109619140625,
0.03277587890625,
-0.02191162109375,
-0.027740478515625,
-0.0894775390625,
-0.03167724609375,
-0.0625,
-0.037445068359375,
0.0286712646484375,
0.0257415771484375,
0.04937744140625,
0.06756591796875,
0.03948974609375,
0.0291748046875,
-0.0213165283203125,
0.0253753662109375,
-0.045318603515625,
0.01261138916015625,
-0.0283355712890625,
-0.055755615234375,
-0.05157470703125,
0.0284576416015625,
-0.00003063678741455078,
0.038330078125,
-0.026092529296875,
0.01800537109375,
-0.0399169921875,
0.032958984375,
-0.0223388671875,
0.0194854736328125,
-0.030181884765625,
0.0232696533203125,
-0.01016998291015625,
-0.00027489662170410156,
-0.0092315673828125,
-0.00289154052734375,
-0.021087646484375,
-0.062042236328125,
-0.008575439453125,
-0.00365447998046875,
0.06036376953125,
0.01210784912109375,
-0.032928466796875,
-0.0177154541015625,
-0.0618896484375,
0.051666259765625,
-0.040283203125,
0.03924560546875,
0.0667724609375,
0.04522705078125,
-0.047943115234375,
-0.05023193359375,
-0.00437164306640625,
-0.02471923828125,
-0.0070037841796875,
0.0149993896484375,
0.01386260986328125,
-0.019134521484375,
0.01554107666015625,
0.0271148681640625,
-0.0487060546875,
-0.02301025390625,
-0.022735595703125,
0.022857666015625,
0.054931640625,
0.012176513671875,
0.03118896484375,
-0.0011463165283203125,
-0.0309295654296875,
0.002582550048828125,
-0.03662109375,
-0.0030803680419921875,
0.0177459716796875,
0.01666259765625,
-0.025970458984375,
0.04656982421875,
0.005252838134765625,
0.0007648468017578125,
0.0005979537963867188,
-0.01535797119140625,
0.007266998291015625,
-0.04864501953125,
-0.008026123046875,
-0.019378662109375,
0.0440673828125,
0.0440673828125,
0.050384521484375,
0.0024261474609375,
0.015899658203125,
0.01198577880859375,
0.038055419921875,
-0.038970947265625,
-0.0169525146484375,
0.025299072265625,
-0.046112060546875,
-0.02276611328125,
-0.0185546875,
-0.04730224609375,
-0.01244354248046875,
-0.00795745849609375,
0.0296478271484375,
-0.0579833984375,
-0.02764892578125,
-0.0051116943359375,
-0.01312255859375,
0.0244293212890625,
0.0223846435546875,
-0.06781005859375,
0.035369873046875,
0.0251007080078125,
0.06561279296875,
0.038848876953125,
-0.0166473388671875,
-0.04058837890625,
-0.0190582275390625,
-0.043060302734375,
0.0389404296875,
-0.0192413330078125,
-0.049774169921875,
0.002109527587890625,
0.00933074951171875,
0.0208892822265625,
-0.04217529296875,
0.053497314453125,
-0.0113677978515625,
0.01195526123046875,
-0.0215301513671875,
-0.02191162109375,
0.0292816162109375,
0.0138092041015625,
-0.038543701171875,
0.06866455078125,
0.031280517578125,
-0.0252227783203125,
0.025604248046875,
-0.03875732421875,
-0.060821533203125,
0.00868988037109375,
-0.01445770263671875,
-0.010406494140625,
-0.0083160400390625,
0.0007543563842773438,
0.0128326416015625,
-0.01397705078125,
0.0008764266967773438,
-0.03472900390625,
-0.0233917236328125,
0.01319122314453125,
0.0177154541015625,
0.046112060546875,
0.0006856918334960938,
-0.0010366439819335938,
-0.0135345458984375,
-0.08721923828125,
0.0083770751953125,
0.0276947021484375,
-0.02313232421875,
-0.02532958984375,
0.00176239013671875,
-0.0015716552734375,
0.0292510986328125,
0.019744873046875,
-0.03411865234375,
0.0209503173828125,
-0.0235748291015625,
0.033477783203125,
0.05523681640625,
0.00698089599609375,
0.0330810546875,
-0.0290069580078125,
0.027923583984375,
0.0073089599609375,
0.037139892578125,
-0.03350830078125,
-0.03143310546875,
-0.036468505859375,
0.016876220703125,
-0.0094757080078125,
0.05487060546875,
-0.0343017578125,
0.08441162109375,
0.016876220703125,
-0.0712890625,
-0.0447998046875,
0.01337432861328125,
0.027191162109375,
0.0105133056640625,
0.046966552734375,
-0.059967041015625,
-0.042236328125,
-0.03515625,
-0.0150146484375,
-0.020751953125,
0.0148773193359375,
0.038330078125,
0.03759765625,
-0.01187896728515625,
0.048370361328125,
-0.0584716796875,
-0.041107177734375,
-0.0038776397705078125,
-0.018463134765625,
0.00789642333984375,
0.061614990234375,
0.046783447265625,
-0.0623779296875,
-0.0355224609375,
-0.007030487060546875,
-0.0751953125,
0.03558349609375,
0.0104522705078125,
-0.0294189453125,
-0.0290985107421875,
0.0322265625,
-0.055755615234375,
0.038055419921875,
0.0142822265625,
-0.061920166015625,
0.00868988037109375,
-0.0377197265625,
0.0113067626953125,
-0.06951904296875,
0.00077056884765625,
0.008941650390625,
-0.037261962890625,
-0.06085205078125,
0.0153350830078125,
-0.03143310546875,
-0.0098724365234375,
-0.0249481201171875,
0.06170654296875,
-0.0298309326171875,
0.01251220703125,
-0.03564453125,
-0.006046295166015625,
-0.016815185546875,
0.038055419921875,
-0.0243988037109375,
0.040435791015625,
0.046600341796875,
-0.03826904296875,
0.0212554931640625,
0.06903076171875,
-0.02008056640625,
0.08282470703125,
-0.04217529296875,
0.047210693359375,
-0.0219268798828125,
0.047515869140625,
-0.07415771484375,
-0.017974853515625,
0.045928955078125,
-0.045196533203125,
0.0106964111328125,
-0.0182647705078125,
-0.0199737548828125,
-0.0152130126953125,
-0.01294708251953125,
0.012420654296875,
0.036102294921875,
-0.03582763671875,
0.0670166015625,
0.019317626953125,
-0.01285552978515625,
-0.02789306640625,
-0.04541015625,
0.00446319580078125,
-0.0101165771484375,
-0.04022216796875,
0.0220184326171875,
-0.04901123046875,
-0.040863037109375,
-0.0183868408203125,
-0.0188446044921875,
-0.044158935546875,
0.0302276611328125,
0.043304443359375,
0.006305694580078125,
0.021575927734375,
-0.004665374755859375,
-0.0027294158935546875,
-0.0023956298828125,
0.0063323974609375,
-0.0081939697265625,
0.06268310546875,
0.003429412841796875,
0.007556915283203125,
-0.04974365234375,
0.033416748046875,
0.03631591796875,
-0.0081634521484375,
0.047271728515625,
0.06890869140625,
-0.024627685546875,
0.01715087890625,
-0.04705810546875,
-0.039154052734375,
-0.032806396484375,
0.0185089111328125,
-0.053985595703125,
-0.042694091796875,
0.044464111328125,
-0.00879669189453125,
-0.006378173828125,
0.023651123046875,
0.0240631103515625,
-0.020904541015625,
0.06695556640625,
0.07159423828125,
0.01502227783203125,
0.048980712890625,
-0.00273895263671875,
-0.005298614501953125,
-0.053802490234375,
-0.0234832763671875,
-0.00568389892578125,
-0.023895263671875,
-0.05328369140625,
-0.0145111083984375,
-0.030975341796875,
0.00933837890625,
-0.034515380859375,
0.048797607421875,
-0.0222320556640625,
0.0098114013671875,
0.048095703125,
0.033477783203125,
-0.0012598037719726562,
-0.046905517578125,
0.008819580078125,
-0.0302734375,
-0.026519775390625,
-0.05126953125,
0.07147216796875,
0.033111572265625,
0.037841796875,
0.01177978515625,
0.058868408203125,
0.00507354736328125,
0.00928497314453125,
-0.045013427734375,
0.05322265625,
0.01302337646484375,
-0.07464599609375,
-0.0205230712890625,
-0.045562744140625,
-0.052734375,
0.015716552734375,
-0.01739501953125,
-0.06134033203125,
0.029022216796875,
-0.00186920166015625,
-0.01788330078125,
-0.0266876220703125,
-0.053314208984375,
0.057647705078125,
-0.032684326171875,
-0.03521728515625,
-0.0243988037109375,
-0.0853271484375,
0.0173797607421875,
0.00652313232421875,
-0.00543212890625,
-0.037261962890625,
-0.0008401870727539062,
0.0273895263671875,
0.0005202293395996094,
0.0321044921875,
-0.023223876953125,
0.01552581787109375,
0.0014753341674804688,
0.032867431640625,
0.0401611328125,
0.0167236328125,
0.0101318359375,
-0.0174102783203125,
0.0185394287109375,
-0.045654296875,
-0.042510986328125,
0.0477294921875,
-0.055908203125,
-0.020294189453125,
-0.028411865234375,
-0.019744873046875,
0.01373291015625,
0.03948974609375,
0.0022678375244140625,
0.01375579833984375,
-0.06005859375,
-0.0095977783203125,
0.0201568603515625,
-0.0228271484375,
0.0267486572265625,
0.0293426513671875,
-0.02008056640625,
-0.060546875,
0.037109375,
0.005664825439453125,
0.01641845703125,
0.0275726318359375,
0.0164794921875,
-0.017730712890625,
0.0024814605712890625,
0.0024261474609375,
0.0501708984375,
-0.05029296875,
0.013519287109375,
-0.054351806640625,
0.0041046142578125,
-0.060638427734375,
0.0035800933837890625,
-0.037017822265625,
-0.01253509521484375,
-0.035400390625,
-0.00864410400390625,
0.0246429443359375,
0.0301971435546875,
0.00017702579498291016,
0.027130126953125,
-0.0275115966796875,
0.007686614990234375,
0.01255035400390625,
0.00943756103515625,
-0.0176849365234375,
-0.031951904296875,
-0.015899658203125,
-0.0033893585205078125,
-0.03302001953125,
-0.052154541015625,
0.0369873046875,
-0.015869140625,
0.041259765625,
0.041015625,
-0.00930023193359375,
0.057403564453125,
-0.0247344970703125,
0.06146240234375,
-0.00603485107421875,
-0.058197021484375,
0.050537109375,
-0.03497314453125,
-0.00896453857421875,
0.06610107421875,
0.030059814453125,
-0.01062774658203125,
0.0055389404296875,
-0.0830078125,
-0.042388916015625,
0.0345458984375,
0.0447998046875,
0.0185546875,
0.0088348388671875,
0.06561279296875,
-0.0216217041015625,
0.054779052734375,
-0.079345703125,
-0.047454833984375,
-0.030975341796875,
0.0235137939453125,
0.0019817352294921875,
-0.00885009765625,
-0.036895751953125,
-0.035552978515625,
0.037933349609375,
0.004955291748046875,
0.05621337890625,
0.036895751953125,
0.0330810546875,
-0.00133514404296875,
0.0183258056640625,
0.060577392578125,
0.047943115234375,
-0.04296875,
0.006439208984375,
0.0237579345703125,
-0.0174560546875,
0.0303955078125,
0.0186920166015625,
0.03668212890625,
-0.004192352294921875,
0.023895263671875,
0.0487060546875,
0.0186920166015625,
-0.044586181640625,
0.01641845703125,
-0.007568359375,
-0.0028076171875,
-0.057525634765625,
0.0178680419921875,
0.01235198974609375,
0.0093841552734375,
0.004913330078125,
0.00760650634765625,
-0.0078582763671875,
-0.099853515625,
-0.0019989013671875,
0.02020263671875,
-0.0516357421875,
-0.02935791015625,
0.0572509765625,
0.030426025390625,
-0.05511474609375,
0.02191162109375,
-0.0165252685546875,
-0.00113677978515625,
0.0362548828125,
0.039154052734375,
0.055267333984375,
-0.053741455078125,
0.01303863525390625,
0.02215576171875,
0.007091522216796875,
-0.0212554931640625,
0.0250396728515625,
-0.017333984375,
-0.048797607421875,
-0.005496978759765625,
-0.0272064208984375,
-0.035125732421875,
0.025909423828125,
-0.035888671875,
0.04852294921875,
-0.037811279296875,
-0.01416015625,
-0.0081024169921875,
0.039031982421875,
-0.047637939453125,
0.026092529296875,
-0.02874755859375,
0.0565185546875,
-0.06536865234375,
0.0255889892578125,
0.07159423828125,
-0.06890869140625,
-0.05328369140625,
-0.0181121826171875,
0.03118896484375,
-0.057342529296875,
0.025787353515625,
0.0223236083984375,
0.01139068603515625,
0.0176239013671875,
-0.03369140625,
-0.040374755859375,
0.11016845703125,
0.01605224609375,
-0.022369384765625,
-0.0257110595703125,
-0.01152801513671875,
0.048431396484375,
-0.036041259765625,
0.05963134765625,
0.048675537109375,
0.03302001953125,
0.0222625732421875,
-0.06787109375,
-0.0021381378173828125,
-0.027252197265625,
-0.00640106201171875,
-0.008575439453125,
-0.08331298828125,
0.06378173828125,
-0.01355743408203125,
-0.0033740997314453125,
0.050872802734375,
0.039031982421875,
0.0009670257568359375,
0.0111846923828125,
0.03289794921875,
0.032257080078125,
0.051177978515625,
-0.0012140274047851562,
0.055633544921875,
-0.017974853515625,
0.00034618377685546875,
0.088134765625,
-0.031494140625,
0.061126708984375,
0.034423828125,
-0.0101165771484375,
0.044952392578125,
0.0640869140625,
-0.0233917236328125,
0.0251007080078125,
-0.003204345703125,
-0.01296234130859375,
-0.0142669677734375,
-0.035186767578125,
-0.0150909423828125,
0.0311279296875,
-0.0254669189453125,
-0.054931640625,
-0.00628662109375,
0.01258087158203125,
0.0270233154296875,
0.034454345703125,
0.0125579833984375,
0.044158935546875,
0.0357666015625,
-0.046783447265625,
-0.00493621826171875,
-0.0182037353515625,
0.0282745361328125,
-0.04205322265625,
-0.01029205322265625,
-0.029022216796875,
0.0307769775390625,
-0.00287628173828125,
-0.034332275390625,
-0.00394439697265625,
-0.02679443359375,
-0.024322509765625,
-0.028076171875,
0.058837890625,
-0.029571533203125,
-0.0235748291015625,
0.062408447265625,
0.053466796875,
0.028533935546875,
-0.01568603515625,
-0.046966552734375,
0.0035839080810546875,
-0.0249481201171875,
0.0137481689453125,
0.024505615234375,
0.0179290771484375,
0.0187225341796875,
0.06134033203125,
0.048095703125,
0.004077911376953125,
-0.039215087890625,
0.0001621246337890625,
0.055999755859375,
-0.07733154296875,
-0.0323486328125,
-0.056304931640625,
0.039520263671875,
0.0017995834350585938,
-0.052276611328125,
0.050750732421875,
0.026885986328125,
0.038818359375,
0.00783538818359375,
0.0280303955078125,
-0.007724761962890625,
0.039459228515625,
-0.021240234375,
0.029205322265625,
-0.03759765625,
0.01381683349609375,
-0.0223236083984375,
-0.061279296875,
-0.0078582763671875,
0.049163818359375,
-0.0160980224609375,
0.0254058837890625,
0.047821044921875,
0.0482177734375,
0.020263671875,
-0.02874755859375,
0.0304107666015625,
0.01119232177734375,
0.048797607421875,
0.039520263671875,
0.050628662109375,
-0.038848876953125,
0.053314208984375,
-0.00484466552734375,
-0.0276641845703125,
-0.006824493408203125,
-0.02191162109375,
-0.086181640625,
-0.058013916015625,
0.00690460205078125,
-0.042327880859375,
0.0299224853515625,
0.08416748046875,
0.052093505859375,
0.0140380859375,
-0.0279998779296875,
-0.044189453125,
-0.0256195068359375,
-0.0290069580078125,
-0.01076507568359375,
-0.005126953125,
-0.0169219970703125,
-0.054229736328125,
0.047882080078125,
0.006328582763671875,
0.03277587890625,
-0.031768798828125,
0.0056304931640625,
-0.04296875,
0.01383209228515625,
0.038482666015625,
0.0242156982421875,
-0.0369873046875,
-0.02337646484375,
0.0155792236328125,
-0.035247802734375,
0.021484375,
0.03436279296875,
-0.033782958984375,
0.03302001953125,
-0.01413726806640625,
0.01271820068359375,
0.07147216796875,
0.0297698974609375,
0.0638427734375,
-0.022216796875,
0.0209503173828125,
0.018341064453125,
0.007282257080078125,
0.047149658203125,
-0.040679931640625,
0.082275390625,
0.0318603515625,
-0.04864501953125,
-0.0618896484375,
0.029144287109375,
-0.095458984375,
-0.0001099705696105957,
0.07806396484375,
-0.00942230224609375,
0.0025005340576171875,
-0.0142822265625,
-0.0264129638671875,
0.0286865234375,
-0.01213836669921875,
0.055633544921875,
0.07818603515625,
-0.03570556640625,
-0.01262664794921875,
-0.045196533203125,
0.039764404296875,
0.020751953125,
-0.06390380859375,
-0.0304412841796875,
-0.0035648345947265625,
0.029022216796875,
0.036651611328125,
0.079345703125,
0.00963592529296875,
0.0171356201171875,
0.004070281982421875,
-0.0094146728515625,
-0.0034580230712890625,
-0.0130615234375,
-0.003917694091796875,
-0.032501220703125,
0.01039886474609375,
-0.036041259765625
]
] |
MBZUAI/LaMini-Cerebras-590M | 2023-04-28T13:08:13.000Z | [
"transformers",
"pytorch",
"gpt2",
"text-generation",
"en",
"arxiv:2304.14402",
"license:cc-by-nc-4.0",
"endpoints_compatible",
"has_space",
"text-generation-inference",
"region:us"
] | text-generation | MBZUAI | null | null | MBZUAI/LaMini-Cerebras-590M | 6 | 5,892 | transformers | 2023-04-12T06:23:08 | ---
license: cc-by-nc-4.0
language:
- en
pipeline_tag: text-generation
widget:
- text: >-
Below is an instruction that describes a task.
Write a response that appropriately completes the request.
### Instruction:
how can I become more healthy?
### Response:
example_title: example
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
<p align="center" width="100%">
<a><img src="https://raw.githubusercontent.com/mbzuai-nlp/lamini-lm/main/images/lamini.png" alt="Title" style="width: 100%; min-width: 300px; display: block; margin: auto;"></a>
</p>
# LaMini-Cerebras-590M
[]()
This model is one of our LaMini-LM model series in paper "[LaMini-LM: A Diverse Herd of Distilled Models from Large-Scale Instructions](https://github.com/mbzuai-nlp/lamini-lm)".
This model is a fine-tuned version of [cerebras/Cerebras-GPT-590M](https://huggingface.co/cerebras/Cerebras-GPT-590M) on [LaMini-instruction dataset](https://huggingface.co/datasets/MBZUAI/LaMini-instruction) that contains 2.58M samples for instruction fine-tuning. For more information about our dataset, please refer to our [project repository](https://github.com/mbzuai-nlp/lamini-lm/).
You can view other models of LaMini-LM series as follows. Models with ✩ are those with the best overall performance given their size/architecture, hence we recommend using them. More details can be seen in our paper.
<table>
<thead>
<tr>
<th>Base model</th>
<th colspan="4">LaMini-LM series (#parameters)</th>
</tr>
</thead>
<tbody>
<tr>
<td>T5</td>
<td><a href="https://huggingface.co/MBZUAI/lamini-t5-61m" target="_blank" rel="noopener noreferrer">LaMini-T5-61M</a></td>
<td><a href="https://huggingface.co/MBZUAI/lamini-t5-223m" target="_blank" rel="noopener noreferrer">LaMini-T5-223M</a></td>
<td><a href="https://huggingface.co/MBZUAI/lamini-t5-738m" target="_blank" rel="noopener noreferrer">LaMini-T5-738M</a></td>
<td></td>
</tr>
<tr>
<td>Flan-T5</td>
<td><a href="https://huggingface.co/MBZUAI/lamini-flan-t5-77m" target="_blank" rel="noopener noreferrer">LaMini-Flan-T5-77M</a>✩</td>
<td><a href="https://huggingface.co/MBZUAI/lamini-flan-t5-248m" target="_blank" rel="noopener noreferrer">LaMini-Flan-T5-248M</a>✩</td>
<td><a href="https://huggingface.co/MBZUAI/lamini-flan-t5-783m" target="_blank" rel="noopener noreferrer">LaMini-Flan-T5-783M</a>✩</td>
<td></td>
</tr>
<tr>
<td>Cerebras-GPT</td>
<td><a href="https://huggingface.co/MBZUAI/lamini-cerebras-111m" target="_blank" rel="noopener noreferrer">LaMini-Cerebras-111M</a></td>
<td><a href="https://huggingface.co/MBZUAI/lamini-cerebras-256m" target="_blank" rel="noopener noreferrer">LaMini-Cerebras-256M</a></td>
<td><a href="https://huggingface.co/MBZUAI/lamini-cerebras-590m" target="_blank" rel="noopener noreferrer">LaMini-Cerebras-590M</a></td>
<td><a href="https://huggingface.co/MBZUAI/lamini-cerebras-1.3b" target="_blank" rel="noopener noreferrer">LaMini-Cerebras-1.3B</a></td>
</tr>
<tr>
<td>GPT-2</td>
<td><a href="https://huggingface.co/MBZUAI/lamini-gpt-124m" target="_blank" rel="noopener noreferrer">LaMini-GPT-124M</a>✩</td>
<td><a href="https://huggingface.co/MBZUAI/lamini-gpt-774m" target="_blank" rel="noopener noreferrer">LaMini-GPT-774M</a>✩</td>
<td><a href="https://huggingface.co/MBZUAI/lamini-gpt-1.5b" target="_blank" rel="noopener noreferrer">LaMini-GPT-1.5B</a>✩</td>
<td></td>
</tr>
<tr>
<td>GPT-Neo</td>
<td><a href="https://huggingface.co/MBZUAI/lamini-neo-125m" target="_blank" rel="noopener noreferrer">LaMini-Neo-125M</a></td>
<td><a href="https://huggingface.co/MBZUAI/lamini-neo-1.3b" target="_blank" rel="noopener noreferrer">LaMini-Neo-1.3B</a></td>
<td></td>
<td></td>
</tr>
<tr>
<td>GPT-J</td>
<td colspan="4">coming soon</td>
</tr>
<tr>
<td>LLaMA</td>
<td colspan="4">coming soon</td>
</tr>
</tbody>
</table>
## Use
### Intended use
We recommend using the model to respond to human instructions written in natural language.
Since this decoder-only model is fine-tuned with wrapper text, we suggest using the same wrapper text to achieve the best performance.
See the example on the right or the code below.
We now show you how to load and use our model using HuggingFace `pipeline()`.
```python
# pip install -q transformers
from transformers import pipeline
checkpoint = "{model_name}"
model = pipeline('text-generation', model = checkpoint)
instruction = 'Please let me know your thoughts on the given place and why you think it deserves to be visited: \n"Barcelona, Spain"'
input_prompt = f"Below is an instruction that describes a task. Write a response that appropriately completes the request.\n\n### Instruction:\n{instruction}\n\n### Response:"
generated_text = model(input_prompt, max_length=512, do_sample=True)[0]['generated_text']
print("Response", generated_text)
```
## Training Procedure
<p align="center" width="100%">
<a><img src="https://raw.githubusercontent.com/mbzuai-nlp/lamini-lm/main/images/lamini-pipeline.drawio.png" alt="Title" style="width: 100%; min-width: 250px; display: block; margin: auto;"></a>
</p>
We initialize with [cerebras/Cerebras-GPT-590M](https://huggingface.co/cerebras/Cerebras-GPT-590M) and fine-tune it on our [LaMini-instruction dataset](https://huggingface.co/datasets/MBZUAI/LaMini-instruction). Its total number of parameters is 590M.
### Training Hyperparameters
## Evaluation
We conducted two sets of evaluations: automatic evaluation on downstream NLP tasks and human evaluation on user-oriented instructions. For more detail, please refer to our [paper]().
## Limitations
More information needed
# Citation
```bibtex
@article{lamini-lm,
author = {Minghao Wu and
Abdul Waheed and
Chiyu Zhang and
Muhammad Abdul-Mageed and
Alham Fikri Aji
},
title = {LaMini-LM: A Diverse Herd of Distilled Models from Large-Scale Instructions},
journal = {CoRR},
volume = {abs/2304.14402},
year = {2023},
url = {https://arxiv.org/abs/2304.14402},
eprinttype = {arXiv},
eprint = {2304.14402}
}
``` | 6,579 | [
[
-0.045684814453125,
-0.053741455078125,
0.0131988525390625,
0.0205078125,
-0.019378662109375,
-0.03179931640625,
-0.01253509521484375,
-0.045074462890625,
0.0276641845703125,
0.020263671875,
-0.05926513671875,
-0.033447265625,
-0.038909912109375,
0.002666473388671875,
-0.0022602081298828125,
0.061767578125,
-0.0157928466796875,
-0.0066070556640625,
0.00945281982421875,
-0.0098724365234375,
-0.0166473388671875,
-0.03155517578125,
-0.064697265625,
-0.0325927734375,
0.018310546875,
0.0013437271118164062,
0.053466796875,
0.06280517578125,
0.0238189697265625,
0.029327392578125,
-0.0192718505859375,
0.0227813720703125,
-0.00603485107421875,
-0.014312744140625,
0.00815582275390625,
-0.0272674560546875,
-0.072509765625,
0.0004730224609375,
0.052886962890625,
0.0221710205078125,
0.0182952880859375,
0.03057861328125,
0.0186004638671875,
0.05780029296875,
-0.031494140625,
0.011810302734375,
-0.0022525787353515625,
0.00733184814453125,
-0.01788330078125,
-0.0014467239379882812,
-0.0164031982421875,
-0.03594970703125,
-0.0008897781372070312,
-0.045989990234375,
-0.01094818115234375,
0.008941650390625,
0.11309814453125,
0.00923919677734375,
-0.005901336669921875,
-0.007335662841796875,
-0.026336669921875,
0.06939697265625,
-0.0614013671875,
0.012359619140625,
0.041229248046875,
-0.0110015869140625,
0.004764556884765625,
-0.03076171875,
-0.05389404296875,
-0.0011157989501953125,
-0.0382080078125,
0.0255279541015625,
-0.0231781005859375,
-0.0274505615234375,
0.046356201171875,
0.009033203125,
-0.033660888671875,
-0.0004978179931640625,
-0.0256195068359375,
-0.00647735595703125,
0.050323486328125,
0.0182037353515625,
0.0504150390625,
-0.0223541259765625,
-0.025909423828125,
-0.017669677734375,
-0.026123046875,
0.025360107421875,
0.02947998046875,
0.021209716796875,
-0.05780029296875,
0.0254058837890625,
-0.00394439697265625,
0.06793212890625,
0.0216827392578125,
-0.0222625732421875,
0.047760009765625,
-0.01593017578125,
-0.0293426513671875,
-0.01849365234375,
0.081787109375,
0.047332763671875,
0.017608642578125,
0.00357818603515625,
-0.0017499923706054688,
-0.0189208984375,
-0.0010843276977539062,
-0.0714111328125,
-0.0037994384765625,
0.021820068359375,
-0.045013427734375,
-0.031494140625,
0.00315093994140625,
-0.06744384765625,
0.0037250518798828125,
-0.031097412109375,
0.016937255859375,
-0.04150390625,
-0.0226287841796875,
0.0189971923828125,
-0.001232147216796875,
0.0241851806640625,
0.020355224609375,
-0.059844970703125,
0.007568359375,
0.0240478515625,
0.05096435546875,
0.0046844482421875,
-0.0210418701171875,
-0.0190887451171875,
0.015777587890625,
0.00821685791015625,
0.051910400390625,
-0.0184326171875,
-0.0288238525390625,
-0.0167388916015625,
0.0267333984375,
-0.03564453125,
-0.017059326171875,
0.06744384765625,
-0.00540924072265625,
0.02490234375,
-0.035369873046875,
-0.028900146484375,
-0.0013246536254882812,
0.0138092041015625,
-0.050689697265625,
0.0750732421875,
0.00673675537109375,
-0.0867919921875,
-0.0014352798461914062,
-0.05902099609375,
-0.0124359130859375,
-0.023345947265625,
0.016510009765625,
-0.050537109375,
-0.020965576171875,
0.0244903564453125,
0.032257080078125,
-0.0234222412109375,
-0.0273590087890625,
-0.0242767333984375,
-0.01849365234375,
0.04058837890625,
-0.0147552490234375,
0.0736083984375,
0.011474609375,
-0.048614501953125,
-0.0129547119140625,
-0.06915283203125,
0.0205841064453125,
0.0255584716796875,
-0.0269317626953125,
-0.0090179443359375,
-0.022552490234375,
0.0181884765625,
0.038177490234375,
0.03387451171875,
-0.0298309326171875,
0.00762176513671875,
-0.03497314453125,
0.0280303955078125,
0.060333251953125,
0.0013570785522460938,
0.031768798828125,
-0.05743408203125,
0.02423095703125,
-0.006237030029296875,
0.0207672119140625,
0.00804901123046875,
-0.02117919921875,
-0.0704345703125,
-0.01517486572265625,
0.022979736328125,
0.044830322265625,
-0.0289306640625,
0.049713134765625,
-0.0026569366455078125,
-0.035003662109375,
-0.048828125,
0.006969451904296875,
0.0506591796875,
0.03497314453125,
0.041534423828125,
-0.01116943359375,
-0.05340576171875,
-0.0599365234375,
-0.0038509368896484375,
-0.0141143798828125,
0.0032806396484375,
0.045562744140625,
0.047515869140625,
-0.0235595703125,
0.035491943359375,
-0.041839599609375,
-0.01308441162109375,
-0.02374267578125,
0.005279541015625,
0.018157958984375,
0.05938720703125,
0.05126953125,
-0.0614013671875,
-0.046844482421875,
0.0012769699096679688,
-0.07037353515625,
-0.009857177734375,
-0.0181884765625,
-0.034454345703125,
0.0166168212890625,
0.0100250244140625,
-0.033599853515625,
0.04180908203125,
0.023895263671875,
-0.037750244140625,
0.03948974609375,
-0.0198516845703125,
0.01126861572265625,
-0.09027099609375,
0.037384033203125,
0.033172607421875,
0.007236480712890625,
-0.06695556640625,
0.01129913330078125,
-0.00803375244140625,
0.029571533203125,
-0.041015625,
0.06341552734375,
-0.031463623046875,
0.0170440673828125,
-0.0135345458984375,
0.0242156982421875,
0.02197265625,
0.044708251953125,
0.0191192626953125,
0.03814697265625,
0.0309906005859375,
-0.03411865234375,
0.021392822265625,
0.03546142578125,
-0.01226806640625,
0.051666259765625,
-0.059722900390625,
0.0061187744140625,
-0.005321502685546875,
0.0128173828125,
-0.03607177734375,
-0.017547607421875,
0.040008544921875,
-0.031158447265625,
0.049072265625,
-0.0078887939453125,
-0.03057861328125,
-0.050018310546875,
-0.0211944580078125,
0.01142120361328125,
0.03778076171875,
-0.028289794921875,
0.0372314453125,
0.0185089111328125,
0.0227813720703125,
-0.055877685546875,
-0.052703857421875,
-0.0216064453125,
-0.036956787109375,
-0.058258056640625,
0.038055419921875,
-0.01349639892578125,
-0.007354736328125,
-0.020050048828125,
-0.00659942626953125,
-0.0160675048828125,
0.009033203125,
0.0295867919921875,
0.036102294921875,
-0.0171966552734375,
-0.0128173828125,
-0.0185089111328125,
-0.009002685546875,
0.00965118408203125,
-0.0008778572082519531,
0.054534912109375,
-0.0301055908203125,
-0.0031757354736328125,
-0.09912109375,
0.004100799560546875,
0.03936767578125,
-0.02142333984375,
0.06597900390625,
0.07843017578125,
-0.0218048095703125,
0.0130615234375,
-0.042083740234375,
-0.007732391357421875,
-0.038360595703125,
-0.0165863037109375,
-0.037628173828125,
-0.033233642578125,
0.049591064453125,
0.00007659196853637695,
-0.015594482421875,
0.041412353515625,
0.0268096923828125,
-0.021209716796875,
0.053497314453125,
0.0283355712890625,
-0.0284576416015625,
0.032867431640625,
-0.057525634765625,
0.00559234619140625,
-0.09991455078125,
-0.040313720703125,
-0.036285400390625,
-0.036956787109375,
-0.033905029296875,
-0.02362060546875,
0.012054443359375,
0.03765869140625,
-0.0462646484375,
0.04315185546875,
-0.04937744140625,
0.01171875,
0.035247802734375,
0.04449462890625,
-0.004917144775390625,
-0.01229095458984375,
-0.0291595458984375,
-0.0014047622680664062,
-0.026397705078125,
-0.0491943359375,
0.06951904296875,
0.029693603515625,
0.03448486328125,
0.007129669189453125,
0.059478759765625,
0.0029144287109375,
0.0031909942626953125,
-0.031585693359375,
0.033966064453125,
-0.00438690185546875,
-0.02899169921875,
-0.02264404296875,
-0.0261688232421875,
-0.0726318359375,
0.006298065185546875,
-0.03497314453125,
-0.08319091796875,
0.0168609619140625,
0.0151824951171875,
-0.032745361328125,
0.0352783203125,
-0.035736083984375,
0.06805419921875,
-0.0247650146484375,
-0.06634521484375,
0.024688720703125,
-0.04669189453125,
0.01090240478515625,
0.028533935546875,
0.018829345703125,
-0.0014820098876953125,
0.0105438232421875,
0.0518798828125,
-0.04718017578125,
0.068603515625,
-0.019744873046875,
-0.006526947021484375,
0.0386962890625,
-0.01430511474609375,
0.04229736328125,
-0.0010747909545898438,
-0.025634765625,
-0.0090179443359375,
-0.00585174560546875,
-0.031402587890625,
-0.035400390625,
0.05804443359375,
-0.0712890625,
-0.03729248046875,
-0.038970947265625,
-0.0277862548828125,
0.01493072509765625,
0.011993408203125,
0.0269317626953125,
0.036590576171875,
0.003025054931640625,
0.0069732666015625,
0.053741455078125,
-0.015960693359375,
0.04547119140625,
0.013397216796875,
0.00472259521484375,
-0.0169830322265625,
0.06219482421875,
-0.0038738250732421875,
0.0096893310546875,
0.0419921875,
0.0176849365234375,
-0.03363037109375,
-0.0205841064453125,
-0.045074462890625,
0.043212890625,
-0.0223388671875,
-0.01715087890625,
-0.04327392578125,
-0.024078369140625,
-0.027069091796875,
-0.02880859375,
-0.0128173828125,
-0.029388427734375,
-0.049835205078125,
-0.00489044189453125,
0.03582763671875,
0.03643798828125,
-0.017242431640625,
0.0237579345703125,
-0.038330078125,
0.0151824951171875,
0.01216888427734375,
0.007762908935546875,
0.0094146728515625,
-0.03558349609375,
-0.00823974609375,
0.0222625732421875,
-0.03753662109375,
-0.0506591796875,
0.050811767578125,
-0.009674072265625,
0.0426025390625,
0.031951904296875,
0.0024623870849609375,
0.058502197265625,
-0.0224761962890625,
0.043426513671875,
0.026519775390625,
-0.07122802734375,
0.0484619140625,
-0.0291900634765625,
0.03271484375,
0.0347900390625,
0.039398193359375,
-0.0269775390625,
-0.016357421875,
-0.04693603515625,
-0.054779052734375,
0.061920166015625,
0.0217742919921875,
0.000736236572265625,
0.006542205810546875,
0.03863525390625,
-0.03204345703125,
-0.0036773681640625,
-0.074462890625,
-0.046356201171875,
-0.030517578125,
-0.006793975830078125,
0.026123046875,
-0.0030956268310546875,
-0.010498046875,
-0.03564453125,
0.0635986328125,
-0.005138397216796875,
0.043670654296875,
0.0166778564453125,
-0.00859832763671875,
-0.0048980712890625,
0.0227813720703125,
0.059417724609375,
0.03460693359375,
-0.0255279541015625,
-0.0201568603515625,
0.0232696533203125,
-0.032806396484375,
0.00193023681640625,
-0.007282257080078125,
-0.0285491943359375,
-0.00583648681640625,
0.01751708984375,
0.07598876953125,
0.017181396484375,
-0.01218414306640625,
0.0369873046875,
0.00806427001953125,
-0.01523590087890625,
-0.0254974365234375,
0.01482391357421875,
0.0185394287109375,
0.0273895263671875,
0.00183868408203125,
0.00952911376953125,
0.0015554428100585938,
-0.0419921875,
0.0195159912109375,
0.0280303955078125,
-0.027496337890625,
-0.017852783203125,
0.0643310546875,
-0.00405120849609375,
-0.01312255859375,
0.0224761962890625,
-0.01482391357421875,
-0.058837890625,
0.04705810546875,
0.053802490234375,
0.042999267578125,
-0.0225372314453125,
0.0255279541015625,
0.0706787109375,
-0.0037059783935546875,
-0.0086669921875,
0.01222991943359375,
0.001705169677734375,
-0.045135498046875,
0.0051422119140625,
-0.07623291015625,
-0.0004935264587402344,
0.0194091796875,
-0.0751953125,
0.0262908935546875,
-0.0360107421875,
-0.0308685302734375,
-0.00337982177734375,
0.0284576416015625,
-0.052276611328125,
0.04925537109375,
0.01123046875,
0.0589599609375,
-0.050201416015625,
0.07720947265625,
0.038116455078125,
-0.05535888671875,
-0.068359375,
0.0064697265625,
0.0016222000122070312,
-0.0711669921875,
0.06121826171875,
0.0011463165283203125,
-0.0014829635620117188,
-0.007312774658203125,
-0.0234527587890625,
-0.053497314453125,
0.10028076171875,
-0.0091400146484375,
-0.019500732421875,
-0.0200958251953125,
0.0208892822265625,
0.049774169921875,
-0.032745361328125,
0.054351806640625,
0.03759765625,
0.050750732421875,
0.005138397216796875,
-0.06414794921875,
0.042755126953125,
-0.0452880859375,
0.003894805908203125,
-0.00004601478576660156,
-0.10302734375,
0.07769775390625,
0.0042877197265625,
0.0008382797241210938,
0.018798828125,
0.0372314453125,
0.0252532958984375,
0.0171051025390625,
0.011199951171875,
0.057647705078125,
0.04052734375,
-0.017669677734375,
0.08184814453125,
-0.032135009765625,
0.04034423828125,
0.073974609375,
0.0037384033203125,
0.07086181640625,
0.0145721435546875,
-0.0190887451171875,
0.056488037109375,
0.0302581787109375,
-0.0257415771484375,
0.0149383544921875,
0.021148681640625,
-0.0124969482421875,
-0.009674072265625,
-0.007122039794921875,
-0.039764404296875,
0.0170440673828125,
0.028106689453125,
-0.039306640625,
0.006748199462890625,
-0.0225067138671875,
0.03155517578125,
0.009307861328125,
-0.0183868408203125,
0.040374755859375,
0.0130615234375,
-0.033905029296875,
0.06573486328125,
0.0002460479736328125,
0.052490234375,
-0.036163330078125,
0.0157928466796875,
-0.012908935546875,
0.0102386474609375,
-0.02410888671875,
-0.047637939453125,
0.007564544677734375,
0.00833892822265625,
-0.0083770751953125,
-0.02520751953125,
0.0350341796875,
-0.0161285400390625,
-0.047119140625,
0.030609130859375,
0.01727294921875,
0.00982666015625,
0.0243682861328125,
-0.0928955078125,
0.0237274169921875,
0.0244140625,
-0.032318115234375,
0.0244903564453125,
0.01479339599609375,
0.0190277099609375,
0.04962158203125,
0.036407470703125,
-0.0025463104248046875,
0.01120758056640625,
-0.0022296905517578125,
0.0657958984375,
-0.033782958984375,
-0.00713348388671875,
-0.06842041015625,
0.05926513671875,
-0.0303955078125,
-0.0230865478515625,
0.07208251953125,
0.044769287109375,
0.05401611328125,
-0.0103759765625,
0.05072021484375,
-0.016387939453125,
0.025970458984375,
-0.045562744140625,
0.0712890625,
-0.04876708984375,
0.01027679443359375,
-0.033782958984375,
-0.048370361328125,
-0.015045166015625,
0.07550048828125,
-0.018310546875,
0.0160980224609375,
0.049591064453125,
0.0562744140625,
0.0019426345825195312,
-0.00521087646484375,
-0.007091522216796875,
0.0191192626953125,
-0.002777099609375,
0.0689697265625,
0.039398193359375,
-0.06390380859375,
0.0119781494140625,
-0.044097900390625,
-0.00653076171875,
-0.0262451171875,
-0.052520751953125,
-0.082275390625,
-0.047515869140625,
-0.038238525390625,
-0.040740966796875,
-0.005584716796875,
0.071044921875,
0.044403076171875,
-0.061920166015625,
-0.0270233154296875,
0.006969451904296875,
0.0011491775512695312,
-0.00797271728515625,
-0.0195465087890625,
0.05743408203125,
0.00115966796875,
-0.07867431640625,
0.006275177001953125,
-0.007648468017578125,
0.0399169921875,
0.01409912109375,
-0.0214385986328125,
-0.03497314453125,
0.00989532470703125,
0.0157470703125,
0.04132080078125,
-0.044036865234375,
-0.0221710205078125,
-0.00316619873046875,
-0.0190887451171875,
0.0165557861328125,
0.0206756591796875,
-0.033050537109375,
0.00946807861328125,
0.0384521484375,
0.01403045654296875,
0.055572509765625,
0.0166015625,
0.0223541259765625,
-0.03662109375,
0.0093841552734375,
-0.00824737548828125,
0.033538818359375,
0.00888824462890625,
-0.031494140625,
0.04150390625,
0.0164947509765625,
-0.03558349609375,
-0.056182861328125,
-0.0084075927734375,
-0.09326171875,
-0.003452301025390625,
0.08380126953125,
-0.0247802734375,
-0.038848876953125,
0.0235137939453125,
-0.0232391357421875,
0.0372314453125,
-0.03485107421875,
0.04052734375,
0.047760009765625,
-0.027191162109375,
-0.011474609375,
-0.046173095703125,
0.049224853515625,
0.017303466796875,
-0.061187744140625,
-0.0211944580078125,
0.0153045654296875,
0.0218353271484375,
0.033294677734375,
0.032806396484375,
-0.006793975830078125,
0.00971221923828125,
-0.010467529296875,
0.0022735595703125,
-0.00626373291015625,
0.0005512237548828125,
-0.008636474609375,
0.0011777877807617188,
-0.02117919921875,
-0.0090789794921875
]
] |
Faradaylab/ARIA-70B-V2 | 2023-10-10T14:02:05.000Z | [
"transformers",
"pytorch",
"llama",
"text-generation",
"code",
"text-generation-inference",
"Meta ",
"facebook",
"openassistant",
"data",
"education",
"languages",
"legal",
"fr",
"en",
"arxiv:2307.09288",
"license:llama2",
"endpoints_compatible",
"region:us"
] | text-generation | Faradaylab | null | null | Faradaylab/ARIA-70B-V2 | 8 | 5,890 | transformers | 2023-09-08T17:30:29 | ---
license: llama2
language:
- fr
- en
tags:
- code
- text-generation-inference
- 'Meta '
- llama
- facebook
- pytorch
- openassistant
- data
- education
- languages
- legal
pipeline_tag: text-generation
inference: true
---
ARIA is the last version of Llama 2 70B finetuned over 50.000 high quality french tokens. We built our own dataset for training doing an extract of the French Dataset from Enno and removing Alpaca style translated text from english.
The goal is to increase model quality on French and general topics.
contact@faradaylab.fr
---
# **Aria 70B is based on Llama 2-70B-Chat-HF**
Llama 2 is a collection of pretrained and fine-tuned generative text models ranging in scale from 7 billion to 70 billion parameters. This is the repository for the 70B fine-tuned model, optimized for dialogue use cases and converted for the Hugging Face Transformers format. Links to other models can be found in the index at the bottom.
# *FINETUNING PROCESS **
We trained the model on a high quality dataset with more than 50.000 rows of french language. The training took 2 days on Amazon Cloud Sagemaker powered by Nvidia GPUs.
# **Timing of training**
2 Days using NVIDIA A10G and Amazon Web services Cloud Instance. We are grateful to Nvidia Inception program.
We are also applying rope scalling as experimental approach used by several other Open source teams to increase context lenght of ARIA from 4,096 to over 6,000 tokens. This will allow the model to handle large files for data extraction. This is not active by default and you should add a line of code at parameters to activate rope scaling.
## Model Details /
*Note: Use of this model is governed by the Meta license because it's based on LLAMA 2. In order to download the model weights and tokenizer, please visit the [website](https://ai.meta.com/resources/models-and-libraries/llama-downloads/) and accept our License before requesting access here.*
**Model Developers** :FARADAY
**Variations** :ARIA comes in a range of parameter sizes — 7B, 40B (based on Falcon), and 70B finetuned on French language datasets.
**Input** :Models input text only.
**Output** : Models generate text only.
**Model Architecture** : ARIA is an auto-regressive language model that uses an optimized transformer architecture. The tuned versions use supervised fine-tuning (SFT) and reinforcement learning with human feedback (RLHF) to align to human preferences for helpfulness and safety.
**License** : A custom commercial license is available at: [https://ai.meta.com/resources/models-and-libraries/llama-downloads/](https://ai.meta.com/resources/models-and-libraries/llama-downloads/)
**Research Paper for LLAMA 2** : ["Llama-2: Open Foundation and Fine-tuned Chat Models"](arxiv.org/abs/2307.09288)
**CO<sub>2</sub> emissions during pretraining.** Time: total GPU time required for training each model. Power Consumption: peak power capacity per GPU device for the GPUs used adjusted for power usage efficiency. 100% of the emissions are directly offset by Meta's sustainability program, and because we are openly releasing these models, the pretraining costs do not need to be incurred by others.
## Training Data
**Overview** ARIA was trained over on 50.000 tokens of data from publicly available sources in French.
**Data Freshness** The pretraining data has a cutoff of September 2022, but some tuning data is more recent, up to August 2023.
**Overall performance on grouped academic benchmarks.** *Code:* We report the average pass@1 scores of our models on HumanEval and MBPP. *Commonsense Reasoning:* We report the average of PIQA, SIQA, HellaSwag, WinoGrande, ARC easy and challenge, OpenBookQA, and CommonsenseQA. We report 7-shot results for CommonSenseQA and 0-shot results for all other benchmarks. *World Knowledge:* We evaluate the 5-shot performance on NaturalQuestions and TriviaQA and report the average. *Reading Comprehension:* For reading comprehension, we report the 0-shot average on SQuAD, QuAC, and BoolQ. *MATH:* We report the average of the GSM8K (8 shot) and MATH (4 shot) benchmarks at top 1.
|||TruthfulQA|Toxigen|
|---|---|---|---|
|Llama 1|7B|27.42|23.00|
|Llama 1|13B|41.74|23.08|
|Llama 1|33B|44.19|22.57|
|Llama 1|65B|48.71|21.77|
|Llama 2|7B|33.29|**21.25**|
|Llama 2|13B|41.86|26.10|
|Llama 2|70B|**50.18**|24.60|
**Evaluation of pretrained LLMs on automatic safety benchmarks.** For TruthfulQA, we present the percentage of generations that are both truthful and informative (the higher the better). For ToxiGen, we present the percentage of toxic generations (the smaller the better).
|||TruthfulQA|Toxigen|
|---|---|---|---|
|Llama-2-Chat|7B|57.04|**0.00**|
|Llama-2-Chat|13B|62.18|**0.00**|
|Llama-2-Chat|70B|**64.14**|0.01|
**Evaluation of fine-tuned LLMs on different safety datasets.** Same metric definitions as above.
## Ethical Considerations and Limitations
ARIA is a new technology that carries risks with use. | 4,941 | [
[
-0.015411376953125,
-0.04815673828125,
0.025054931640625,
0.021759033203125,
-0.01617431640625,
0.00461578369140625,
-0.006317138671875,
-0.056488037109375,
-0.0024127960205078125,
0.034576416015625,
-0.04937744140625,
-0.049713134765625,
-0.05792236328125,
0.01666259765625,
-0.031158447265625,
0.07879638671875,
-0.00519561767578125,
-0.00777435302734375,
0.01111602783203125,
-0.0110626220703125,
-0.0222015380859375,
-0.039398193359375,
-0.0667724609375,
-0.0236968994140625,
0.0428466796875,
0.029449462890625,
0.046356201171875,
0.059478759765625,
0.037078857421875,
0.0229339599609375,
-0.0199737548828125,
0.00923919677734375,
-0.043060302734375,
-0.02056884765625,
0.01439666748046875,
-0.042755126953125,
-0.053680419921875,
0.002338409423828125,
0.046600341796875,
0.011322021484375,
0.0026836395263671875,
0.024505615234375,
0.00836181640625,
0.05267333984375,
-0.033172607421875,
0.0157318115234375,
-0.05108642578125,
-0.00011527538299560547,
-0.01403045654296875,
-0.0110321044921875,
-0.0262451171875,
-0.01313018798828125,
-0.007266998291015625,
-0.06927490234375,
-0.01348876953125,
0.01422882080078125,
0.08856201171875,
0.06597900390625,
-0.038909912109375,
-0.01488494873046875,
-0.03546142578125,
0.07806396484375,
-0.052337646484375,
0.01374053955078125,
0.03912353515625,
0.0283355712890625,
-0.00719451904296875,
-0.0645751953125,
-0.03375244140625,
-0.01934814453125,
0.0158843994140625,
0.01300811767578125,
-0.0362548828125,
-0.0032749176025390625,
0.0102386474609375,
0.025848388671875,
-0.044403076171875,
0.033660888671875,
-0.031158447265625,
-0.0208740234375,
0.0726318359375,
0.005008697509765625,
0.00940704345703125,
0.0131683349609375,
-0.036102294921875,
-0.02313232421875,
-0.0731201171875,
0.0038242340087890625,
0.036834716796875,
0.0106048583984375,
-0.026580810546875,
0.0406494140625,
-0.03289794921875,
0.024749755859375,
0.001026153564453125,
-0.0185089111328125,
0.03948974609375,
-0.015869140625,
-0.0209503173828125,
-0.006488800048828125,
0.06585693359375,
0.041717529296875,
0.024688720703125,
-0.0023441314697265625,
-0.005435943603515625,
0.00719451904296875,
0.0023860931396484375,
-0.057037353515625,
0.0019817352294921875,
0.0198516845703125,
-0.024688720703125,
-0.038543701171875,
-0.036376953125,
-0.03900146484375,
-0.0099029541015625,
-0.0171966552734375,
0.03253173828125,
-0.0186767578125,
-0.0089569091796875,
0.01067352294921875,
0.01296234130859375,
0.032806396484375,
0.0213165283203125,
-0.045684814453125,
0.01282501220703125,
0.033905029296875,
0.0443115234375,
-0.0171356201171875,
-0.0284423828125,
-0.026611328125,
-0.0189666748046875,
-0.026763916015625,
0.06842041015625,
-0.0281524658203125,
-0.0176849365234375,
0.00276947021484375,
0.02215576171875,
0.01168060302734375,
-0.03472900390625,
0.05853271484375,
-0.03955078125,
0.016204833984375,
-0.03619384765625,
-0.036163330078125,
-0.0263824462890625,
0.01360321044921875,
-0.045654296875,
0.10430908203125,
0.000690460205078125,
-0.0298919677734375,
0.036651611328125,
-0.054534912109375,
-0.0105743408203125,
-0.0136566162109375,
0.012451171875,
-0.041748046875,
-0.034332275390625,
0.02264404296875,
0.0258941650390625,
-0.02874755859375,
0.0183258056640625,
-0.00699615478515625,
-0.0302581787109375,
0.004608154296875,
-0.042327880859375,
0.057037353515625,
0.0300140380859375,
-0.034088134765625,
0.00464630126953125,
-0.06427001953125,
-0.003597259521484375,
0.01459503173828125,
-0.038604736328125,
0.01393890380859375,
-0.002567291259765625,
-0.0019102096557617188,
0.01435089111328125,
0.0306396484375,
-0.02899169921875,
0.002239227294921875,
-0.033477783203125,
0.0304412841796875,
0.04400634765625,
0.0024585723876953125,
0.02789306640625,
-0.043731689453125,
0.036956787109375,
-0.0090484619140625,
0.030426025390625,
0.00693511962890625,
-0.0443115234375,
-0.08929443359375,
-0.01047515869140625,
0.0164031982421875,
0.056182861328125,
-0.039306640625,
0.04815673828125,
-0.005390167236328125,
-0.039947509765625,
-0.0338134765625,
0.01313018798828125,
0.058349609375,
0.02264404296875,
0.04150390625,
-0.021514892578125,
-0.034393310546875,
-0.0679931640625,
0.0026531219482421875,
-0.0204315185546875,
0.00460052490234375,
0.020172119140625,
0.053497314453125,
-0.011474609375,
0.04119873046875,
-0.0185394287109375,
-0.0267486572265625,
-0.0237274169921875,
0.00591278076171875,
-0.00018537044525146484,
0.023834228515625,
0.06597900390625,
-0.036041259765625,
-0.022613525390625,
-0.01302337646484375,
-0.0709228515625,
-0.0045623779296875,
0.00519561767578125,
-0.021026611328125,
0.024261474609375,
0.030364990234375,
-0.026336669921875,
0.035675048828125,
0.049285888671875,
-0.0132293701171875,
0.0322265625,
0.00164031982421875,
0.0010499954223632812,
-0.0833740234375,
-0.007114410400390625,
-0.002933502197265625,
-0.01044464111328125,
-0.032196044921875,
0.001575469970703125,
-0.016510009765625,
0.00591278076171875,
-0.0762939453125,
0.04962158203125,
-0.0310516357421875,
-0.00627899169921875,
-0.01104736328125,
-0.006542205810546875,
0.004505157470703125,
0.058563232421875,
-0.0021762847900390625,
0.081787109375,
0.03497314453125,
-0.051910400390625,
0.021240234375,
0.03338623046875,
-0.025115966796875,
0.0239105224609375,
-0.0626220703125,
0.011199951171875,
0.00276947021484375,
0.03729248046875,
-0.055450439453125,
-0.01523590087890625,
0.01702880859375,
-0.031646728515625,
0.011322021484375,
-0.004669189453125,
-0.0299072265625,
-0.0225067138671875,
-0.0196685791015625,
0.029510498046875,
0.05352783203125,
-0.03375244140625,
0.0246734619140625,
0.028289794921875,
-0.00795745849609375,
-0.07342529296875,
-0.06365966796875,
0.01360321044921875,
-0.03173828125,
-0.0469970703125,
0.01580810546875,
-0.01200103759765625,
-0.0185089111328125,
-0.011199951171875,
0.008056640625,
-0.01238250732421875,
0.01175689697265625,
0.0175323486328125,
0.0152130126953125,
-0.018280029296875,
0.01544189453125,
0.010223388671875,
-0.0183563232421875,
-0.0026264190673828125,
-0.005458831787109375,
0.0435791015625,
-0.0338134765625,
-0.021392822265625,
-0.0660400390625,
0.0077056884765625,
0.029449462890625,
-0.0183563232421875,
0.051666259765625,
0.037872314453125,
-0.015594482421875,
0.0037136077880859375,
-0.04840087890625,
-0.01065826416015625,
-0.038482666015625,
0.0340576171875,
-0.0173797607421875,
-0.058349609375,
0.0408935546875,
0.0014829635620117188,
0.0297088623046875,
0.0618896484375,
0.042083740234375,
-0.01110076904296875,
0.055328369140625,
0.048370361328125,
-0.00971221923828125,
0.031402587890625,
-0.052764892578125,
-0.01495361328125,
-0.0567626953125,
-0.02239990234375,
-0.03778076171875,
-0.0293426513671875,
-0.038238525390625,
-0.0235137939453125,
0.0281982421875,
-0.013214111328125,
-0.049652099609375,
0.032012939453125,
-0.042266845703125,
0.0440673828125,
0.034942626953125,
0.01141357421875,
0.03326416015625,
0.0032596588134765625,
0.0111083984375,
0.00592041015625,
-0.045684814453125,
-0.05938720703125,
0.11749267578125,
0.044830322265625,
0.044403076171875,
-0.0007653236389160156,
0.055450439453125,
0.0303802490234375,
0.02685546875,
-0.055511474609375,
0.048797607421875,
-0.010223388671875,
-0.059600830078125,
-0.026458740234375,
-0.0155487060546875,
-0.0711669921875,
-0.00678253173828125,
-0.02557373046875,
-0.041015625,
0.00910186767578125,
0.0102081298828125,
-0.03472900390625,
0.0180511474609375,
-0.0426025390625,
0.05078125,
-0.041595458984375,
-0.0305328369140625,
-0.0100555419921875,
-0.059814453125,
0.031280517578125,
-0.01100921630859375,
0.0120697021484375,
-0.028564453125,
-0.001922607421875,
0.0631103515625,
-0.034454345703125,
0.06640625,
0.0033168792724609375,
-0.006252288818359375,
0.041900634765625,
-0.014007568359375,
0.039031982421875,
-0.00717926025390625,
-0.017578125,
0.039947509765625,
-0.00829315185546875,
-0.03155517578125,
-0.0072174072265625,
0.041778564453125,
-0.09210205078125,
-0.027801513671875,
-0.047515869140625,
-0.03546142578125,
-0.0019550323486328125,
-0.001918792724609375,
0.052398681640625,
0.0290069580078125,
-0.013427734375,
0.019561767578125,
0.041595458984375,
-0.038116455078125,
0.0447998046875,
0.037689208984375,
-0.0016536712646484375,
-0.0253143310546875,
0.0584716796875,
0.0045318603515625,
0.0214080810546875,
0.02880859375,
-0.0130615234375,
-0.0445556640625,
-0.037811279296875,
-0.048553466796875,
0.0236968994140625,
-0.044036865234375,
-0.0210723876953125,
-0.04095458984375,
-0.028778076171875,
-0.01116180419921875,
0.0024814605712890625,
-0.044036865234375,
-0.04034423828125,
-0.044830322265625,
-0.01995849609375,
0.03778076171875,
0.059478759765625,
0.008331298828125,
0.05078125,
-0.039031982421875,
0.001922607421875,
0.02630615234375,
0.0034732818603515625,
-0.0206146240234375,
-0.0606689453125,
-0.00507354736328125,
0.00975799560546875,
-0.0462646484375,
-0.04486083984375,
0.02716064453125,
0.025482177734375,
0.031707763671875,
0.0256500244140625,
0.00557708740234375,
0.03082275390625,
-0.031524658203125,
0.0816650390625,
0.0081024169921875,
-0.0462646484375,
0.033477783203125,
-0.023956298828125,
-0.0006709098815917969,
0.0478515625,
0.0252838134765625,
-0.004150390625,
-0.0205841064453125,
-0.04498291015625,
-0.05401611328125,
0.04852294921875,
0.0347900390625,
0.0099029541015625,
0.01346588134765625,
0.0236968994140625,
-0.0018634796142578125,
0.0031681060791015625,
-0.07135009765625,
-0.0225830078125,
-0.014984130859375,
-0.0157470703125,
-0.00867462158203125,
-0.021148681640625,
-0.017578125,
-0.0136260986328125,
0.06427001953125,
-0.00136566162109375,
0.036376953125,
-0.002841949462890625,
-0.00943756103515625,
-0.00946807861328125,
0.016571044921875,
0.0528564453125,
0.034210205078125,
-0.01139068603515625,
-0.01404571533203125,
0.04913330078125,
-0.032135009765625,
0.0206298828125,
0.00685882568359375,
-0.00937652587890625,
-0.019134521484375,
0.031829833984375,
0.0654296875,
0.01125335693359375,
-0.044189453125,
0.032958984375,
-0.005268096923828125,
-0.03497314453125,
-0.045745849609375,
0.018157958984375,
-0.0087890625,
0.0290069580078125,
0.0211639404296875,
0.00042748451232910156,
0.00988006591796875,
-0.0433349609375,
0.0119781494140625,
0.02703857421875,
-0.0178375244140625,
-0.0227813720703125,
0.07305908203125,
0.031707763671875,
-0.0188751220703125,
0.022613525390625,
-0.017913818359375,
-0.028778076171875,
0.06256103515625,
0.0253143310546875,
0.05908203125,
-0.015655517578125,
0.0188140869140625,
0.038482666015625,
0.034210205078125,
-0.018310546875,
0.019622802734375,
0.01512908935546875,
-0.042572021484375,
-0.032806396484375,
-0.058380126953125,
-0.037689208984375,
0.0279998779296875,
-0.03582763671875,
0.0258636474609375,
-0.029449462890625,
-0.0180816650390625,
-0.0213165283203125,
0.024169921875,
-0.055145263671875,
0.036285400390625,
0.0008091926574707031,
0.0780029296875,
-0.0738525390625,
0.055816650390625,
0.0394287109375,
-0.05072021484375,
-0.07415771484375,
-0.01100921630859375,
0.01165008544921875,
-0.08990478515625,
0.038482666015625,
0.0081787109375,
-0.00917816162109375,
0.0019435882568359375,
-0.053863525390625,
-0.0826416015625,
0.1021728515625,
0.02923583984375,
-0.045654296875,
-0.005146026611328125,
0.026702880859375,
0.047393798828125,
-0.025634765625,
0.0261077880859375,
0.0567626953125,
0.036712646484375,
0.02032470703125,
-0.07342529296875,
-0.007747650146484375,
-0.0304718017578125,
-0.022308349609375,
-0.0181884765625,
-0.08819580078125,
0.06219482421875,
-0.006160736083984375,
-0.00782012939453125,
0.0237274169921875,
0.045166015625,
0.019775390625,
0.038665771484375,
0.0298004150390625,
0.0626220703125,
0.056365966796875,
-0.01531982421875,
0.08380126953125,
-0.037017822265625,
0.01151275634765625,
0.0667724609375,
-0.00839996337890625,
0.0709228515625,
0.006481170654296875,
-0.0249481201171875,
0.044097900390625,
0.06915283203125,
0.0155792236328125,
0.037506103515625,
-0.008514404296875,
-0.01415252685546875,
-0.0172119140625,
-0.00814056396484375,
-0.04937744140625,
0.037139892578125,
0.01372528076171875,
-0.0282745361328125,
-0.00585174560546875,
-0.01238250732421875,
0.0248260498046875,
-0.01654052734375,
-0.0036144256591796875,
0.06158447265625,
0.01277923583984375,
-0.048614501953125,
0.07635498046875,
-0.00992584228515625,
0.0455322265625,
-0.043243408203125,
0.0082855224609375,
-0.029693603515625,
0.00872802734375,
-0.01306915283203125,
-0.047760009765625,
0.0148162841796875,
0.023651123046875,
0.0106048583984375,
-0.0178680419921875,
0.03558349609375,
-0.01348876953125,
-0.046661376953125,
0.024139404296875,
0.02490234375,
0.0377197265625,
-0.0018405914306640625,
-0.0614013671875,
0.019622802734375,
0.00070953369140625,
-0.027130126953125,
0.01861572265625,
0.017913818359375,
-0.0071258544921875,
0.06378173828125,
0.044097900390625,
-0.00666046142578125,
0.01197052001953125,
-0.005458831787109375,
0.07373046875,
-0.02642822265625,
-0.017730712890625,
-0.038604736328125,
0.0511474609375,
0.0018949508666992188,
-0.058380126953125,
0.030364990234375,
0.04205322265625,
0.037506103515625,
0.0113067626953125,
0.05584716796875,
0.0209808349609375,
0.028839111328125,
-0.05328369140625,
0.052764892578125,
-0.0716552734375,
0.03607177734375,
-0.00469207763671875,
-0.08355712890625,
-0.01125335693359375,
0.046600341796875,
-0.0265350341796875,
-0.001773834228515625,
0.046600341796875,
0.0662841796875,
0.01360321044921875,
-0.00818634033203125,
0.018890380859375,
0.0240325927734375,
0.0248565673828125,
0.06280517578125,
0.06781005859375,
-0.056671142578125,
0.0621337890625,
-0.01342010498046875,
-0.0301361083984375,
-0.024139404296875,
-0.06298828125,
-0.0748291015625,
-0.034698486328125,
-0.0185089111328125,
-0.018096923828125,
0.022308349609375,
0.060943603515625,
0.050628662109375,
-0.0504150390625,
-0.00994110107421875,
-0.00923919677734375,
-0.01041412353515625,
-0.0006861686706542969,
-0.011138916015625,
0.0253143310546875,
-0.0208587646484375,
-0.05926513671875,
0.03912353515625,
0.00608062744140625,
0.01181793212890625,
-0.01505279541015625,
-0.0223236083984375,
-0.006439208984375,
0.0153656005859375,
0.042022705078125,
0.035247802734375,
-0.05853271484375,
-0.031890869140625,
0.01297760009765625,
-0.01161956787109375,
0.009857177734375,
-0.00864410400390625,
-0.046661376953125,
0.0010938644409179688,
0.006900787353515625,
0.033447265625,
0.058349609375,
0.00032210350036621094,
0.019256591796875,
-0.04638671875,
0.032379150390625,
0.0162811279296875,
0.028656005859375,
0.024505615234375,
-0.0279388427734375,
0.054595947265625,
0.007904052734375,
-0.0687255859375,
-0.054901123046875,
0.0175018310546875,
-0.08563232421875,
0.0022068023681640625,
0.10595703125,
-0.00383758544921875,
-0.0147552490234375,
-0.0003147125244140625,
-0.024749755859375,
0.023956298828125,
-0.035552978515625,
0.05438232421875,
0.039398193359375,
-0.00299072265625,
-0.00954437255859375,
-0.046966552734375,
0.047149658203125,
0.02825927734375,
-0.07623291015625,
-0.0290679931640625,
0.036834716796875,
0.03570556640625,
0.0017385482788085938,
0.049285888671875,
-0.00002086162567138672,
0.01800537109375,
-0.0030803680419921875,
0.0105743408203125,
-0.004917144775390625,
-0.037261962890625,
-0.0082550048828125,
-0.0101470947265625,
-0.00162506103515625,
-0.0252838134765625
]
] |
GeorgiaTechResearchInstitute/starcoder-gpteacher-code-instruct | 2023-06-12T15:38:10.000Z | [
"transformers",
"pytorch",
"gpt_bigcode",
"text-generation",
"Code-Gen",
"dataset:bigcode/the-stack-dedup",
"dataset:teknium1/GPTeacher-codegen",
"arxiv:1911.02150",
"arxiv:2205.14135",
"arxiv:2207.14255",
"arxiv:2305.06161",
"license:bigcode-openrail-m",
"endpoints_compatible",
"has_space",
"text-generation-inference",
"region:us"
] | text-generation | GeorgiaTechResearchInstitute | null | null | GeorgiaTechResearchInstitute/starcoder-gpteacher-code-instruct | 74 | 5,889 | transformers | 2023-05-05T20:04:05 | ---
license: bigcode-openrail-m
datasets:
- bigcode/the-stack-dedup
- teknium1/GPTeacher-codegen
library_name: transformers
pipeline_tag: text-generation
tags:
- Code-Gen
---
# StarCoder GPTeacher-Codegen Fine-Tuned
<!-- Provide a quick summary of what the model is/does. -->
This model is [`bigcode/starcoder`](https://huggingface.co/bigcode/starcoder) fine-tuned on the [`teknium1/GPTeacher`](https://github.com/teknium1/GPTeacher) codegen dataset (GPT-4 code instruction fine-tuning).
## Model Details
The base StarCoder models are 15.5B parameter models trained on 80+ programming languages from [The Stack (v1.2)](https://huggingface.co/datasets/bigcode/the-stack), with opt-out requests excluded. The model uses [Multi Query Attention](https://arxiv.org/abs/1911.02150), [a context window of 8192 tokens](https://arxiv.org/abs/2205.14135), and was trained using the [Fill-in-the-Middle objective](https://arxiv.org/abs/2207.14255) on 1 trillion tokens.
- **Repository:** [bigcode/Megatron-LM](https://github.com/bigcode-project/Megatron-LM)
- **Project Website:** [bigcode-project.org](https://www.bigcode-project.org)
- **Paper:** [💫StarCoder: May the source be with you!](https://drive.google.com/file/d/1cN-b9GnWtHzQRoE7M7gAEyivY0kl4BYs/view)
- **Point of Contact:** [contact@bigcode-project.org](mailto:contact@bigcode-project.org)
- **Languages:** 80+ Programming languages
## Uses
<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
### Intended use
The base model was trained on GitHub code and then fine-tuned to follow instructions. Prompts such as "Write a function that computes the square root." should work reasonably well. The original repo recommeds using the [Tech Assistant prompt](https://huggingface.co/datasets/bigcode/ta-prompt) to few-shot prompt it into behaving as a technical assistant. This fine-tuned model uses the [Alpaca prompts](https://github.com/tatsu-lab/stanford_alpaca/blob/main/train.py).
### Generation
```python
import torch
from transformers import AutoModelForCausalLM, AutoTokenizer
checkpoint = "GeorgiaTechResearchInstitute/starcoder-gpteacher-code-instruct"
device = "cuda"
input_prompt = ("Below is an instruction that describes a task, paired with an input that provides further context. Write a response that appropriately completes the request.\n\n"
"### Instruction:\n"
"{instruction}\n\n"
"### Input:\n"
"{input}\n\n"
"### Response:")
prompt = "Please explain the following program."
extra_input = "send(to, from, count)
register short *to, *from;
register count;
{
register n = (count + 7) / 8;
switch (count % 8) {
case 0: do { *to = *from++;
case 7: *to = *from++;
case 6: *to = *from++;
case 5: *to = *from++;
case 4: *to = *from++;
case 3: *to = *from++;
case 2: *to = *from++;
case 1: *to = *from++;
} while (--n > 0);
}
}"
prompt = input_prompt.format_map({"instruction": prompt, "input": extra_input})
tokenizer = AutoTokenizer.from_pretrained(checkpoint)
model = AutoModelForCausalLM.from_pretrained(checkpoint, trust_remote_code=True, torch_dtype=torch.float16).to(device)
inputs = tokenizer.encode(prompt, return_tensors="pt").to(device)
outputs = model.generate(inputs)
print(tokenizer.decode(outputs[0]))
```
<details>
<summary> Example Outputs </summary>
*__Full Prompt:__*
```
Below is an instruction that describes a task, paired with an input that provides further context. Write a response that appropriately completes the request.
### Instruction:
What does this program do and how does it work?
### Input:
send(to, from, count)
register short *to, *from;
register count;
{
register n = (count + 7) / 8;
switch (count % 8) {
case 0: *to = *from++;
case 7: *to = *from++;
case 6: *to = *from++;
case 5: *to = *from++;
case 4: *to = *from++;
case 3: *to = *from++;
case 2: *to = *from++;
case 1: *to = *from++;
}
while (--n > 0) {
*to = *from++;
*to = *from++;
*to = *from++;
*to = *from++;
*to = *from++;
*to = *from++;
*to = *from++;
*to = *from++;
}
}
### Response:
```
*__Response:__*
```
This program is designed to efficiently copy bytes from one memory location to another. It copies 'count' bytes from the location pointed to by 'from' to the location pointed to by 'to', and works correctly regardless of address alignment. The function 'foo' is a wrapper function that initializes the 'to' and 'from' pointers, checks for proper memory bounds, and performs byte alignment if necessary.
To use this program, you must declare two pointer variables, 'to' and 'from', and then call the 'foo' function with the following parameters:
foo(to, from, count)
```
</details>
### Attribution & Other Requirements
The pretraining dataset of the model was filtered for permissive licenses only. Nevertheless, the model can generate source code verbatim from the dataset. The code's license might require attribution and/or other specific requirements that must be respected. The BigCode project provides a [search index](https://huggingface.co/spaces/bigcode/starcoder-search) that let's you search through the pretraining data to identify where generated code came from and apply the proper attribution to your code.
# Limitations
The model has been trained on source code from 80+ programming languages. The predominant language in source is English although other languages are also present. As such the model is capable of generating code snippets provided some context but the generated code is not guaranteed to work as intended. It can be inefficient, contain bugs or exploits. See [the original paper](https://drive.google.com/file/d/1cN-b9GnWtHzQRoE7M7gAEyivY0kl4BYs/view) for an in-depth discussion of the model limitations.
The fine-tuning process makes the model more responsive to direct user input, however this is an early attempt at instruction fine-tuning starcoder models and the results may not be representative of the model's full potential.
# Training
## Model
- **Architecture:** GPT-2 model with multi-query attention and Fill-in-the-Middle objective
- **Pretraining steps:** 250k
- **Pretraining tokens:** 1 trillion
- **Precision:** bfloat16
- **Fine-Tuning Instruct-Response Pairs:** 4.5k
- **Fine-Tuning Context Length:** 1024
- **Fine-Tuning Epochs:** 3
- **Fine-Tuning LR:** 2e-5
- **Fine-Tuning Optimizations:** FSDP
## Hardware
- **GPUs:** 8 Tesla A100
- **Training time:** 5 hours
# License
The model is licensed under the BigCode OpenRAIL-M v1 license agreement. You can find the full agreement [here](https://huggingface.co/spaces/bigcode/bigcode-model-license-agreement). This model was also fine-tuned using outputs from OpenAI's GPT-4, and as such it is additionally subject to [OpenAI's terms of service.](https://openai.com/policies/terms-of-use)
## Citation
<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
The base model HF repo can be found [here.](https://huggingface.co/bigcode/starcoder)
```
@article{li2023starcoder,
title={StarCoder: may the source be with you!},
author={Raymond Li and Loubna Ben Allal and Yangtian Zi and Niklas Muennighoff and Denis Kocetkov and Chenghao Mou and Marc Marone and Christopher Akiki and Jia Li and Jenny Chim and Qian Liu and Evgenii Zheltonozhskii and Terry Yue Zhuo and Thomas Wang and Olivier Dehaene and Mishig Davaadorj and Joel Lamy-Poirier and João Monteiro and Oleh Shliazhko and Nicolas Gontier and Nicholas Meade and Armel Zebaze and Ming-Ho Yee and Logesh Kumar Umapathi and Jian Zhu and Benjamin Lipkin and Muhtasham Oblokulov and Zhiruo Wang and Rudra Murthy and Jason Stillerman and Siva Sankalp Patel and Dmitry Abulkhanov and Marco Zocca and Manan Dey and Zhihan Zhang and Nour Fahmy and Urvashi Bhattacharyya and Wenhao Yu and Swayam Singh and Sasha Luccioni and Paulo Villegas and Maxim Kunakov and Fedor Zhdanov and Manuel Romero and Tony Lee and Nadav Timor and Jennifer Ding and Claire Schlesinger and Hailey Schoelkopf and Jan Ebert and Tri Dao and Mayank Mishra and Alex Gu and Jennifer Robinson and Carolyn Jane Anderson and Brendan Dolan-Gavitt and Danish Contractor and Siva Reddy and Daniel Fried and Dzmitry Bahdanau and Yacine Jernite and Carlos Muñoz Ferrandis and Sean Hughes and Thomas Wolf and Arjun Guha and Leandro von Werra and Harm de Vries},
year={2023},
eprint={2305.06161},
archivePrefix={arXiv},
primaryClass={cs.CL}
}
``` | 9,002 | [
[
-0.0310516357421875,
-0.05322265625,
0.0322265625,
0.01134490966796875,
-0.0112457275390625,
-0.021148681640625,
-0.02117919921875,
-0.036590576171875,
0.0017452239990234375,
0.0338134765625,
-0.03955078125,
-0.04083251953125,
-0.060455322265625,
0.003566741943359375,
-0.0168914794921875,
0.08392333984375,
-0.00734710693359375,
0.007373809814453125,
0.00391387939453125,
0.003055572509765625,
-0.03082275390625,
-0.048675537109375,
-0.032867431640625,
-0.0017614364624023438,
0.02069091796875,
0.0357666015625,
0.0487060546875,
0.07012939453125,
0.04248046875,
0.0223388671875,
0.0008335113525390625,
-0.004764556884765625,
-0.0279388427734375,
-0.030609130859375,
0.00038051605224609375,
-0.02630615234375,
-0.0292205810546875,
-0.005207061767578125,
0.04400634765625,
0.0236663818359375,
0.01226043701171875,
0.042938232421875,
-0.007152557373046875,
0.042877197265625,
-0.043060302734375,
0.025482177734375,
-0.027679443359375,
-0.00791168212890625,
0.0084381103515625,
-0.005184173583984375,
-0.0244140625,
-0.0311431884765625,
-0.0273284912109375,
-0.033050537109375,
0.020782470703125,
0.005466461181640625,
0.0791015625,
0.0276641845703125,
-0.01885986328125,
-0.043212890625,
-0.06439208984375,
0.045654296875,
-0.053680419921875,
0.0340576171875,
0.02655029296875,
0.02484130859375,
-0.0007839202880859375,
-0.07305908203125,
-0.05853271484375,
-0.0345458984375,
-0.0021800994873046875,
-0.00353240966796875,
-0.01224517822265625,
-0.00193023681640625,
0.04937744140625,
0.0185089111328125,
-0.0487060546875,
0.004192352294921875,
-0.0594482421875,
-0.00861358642578125,
0.04083251953125,
0.00867462158203125,
0.00994873046875,
-0.033416748046875,
-0.030914306640625,
-0.019561767578125,
-0.0433349609375,
0.021575927734375,
0.0201416015625,
0.00798797607421875,
-0.0214385986328125,
0.045440673828125,
-0.00933074951171875,
0.060821533203125,
0.0137939453125,
0.0106201171875,
0.0213470458984375,
-0.033111572265625,
-0.0245513916015625,
-0.00786590576171875,
0.07147216796875,
0.0268096923828125,
0.00786590576171875,
0.000004112720489501953,
-0.0182647705078125,
0.0001589059829711914,
0.0157623291015625,
-0.08197021484375,
-0.0286712646484375,
0.0266571044921875,
-0.0357666015625,
-0.0045623779296875,
0.01454925537109375,
-0.050567626953125,
0.01605224609375,
-0.020355224609375,
0.043701171875,
-0.0307464599609375,
0.005878448486328125,
0.0193023681640625,
0.00319671630859375,
0.034149169921875,
-0.007694244384765625,
-0.07122802734375,
0.02142333984375,
0.04693603515625,
0.047332763671875,
0.021087646484375,
-0.04571533203125,
-0.036834716796875,
-0.008392333984375,
-0.01247406005859375,
0.0214385986328125,
-0.0374755859375,
-0.0231170654296875,
-0.0229339599609375,
0.012542724609375,
-0.016845703125,
-0.023834228515625,
0.0214080810546875,
-0.05169677734375,
0.0164031982421875,
-0.032135009765625,
-0.04278564453125,
-0.0158233642578125,
0.0006165504455566406,
-0.04327392578125,
0.06793212890625,
0.0175323486328125,
-0.046600341796875,
0.01076507568359375,
-0.059295654296875,
-0.015777587890625,
-0.01186370849609375,
-0.0097198486328125,
-0.037750244140625,
-0.0084075927734375,
0.0291900634765625,
0.040679931640625,
-0.0235137939453125,
0.0190582275390625,
-0.01361083984375,
-0.048004150390625,
0.01540374755859375,
-0.01303863525390625,
0.07080078125,
0.031097412109375,
-0.049774169921875,
0.0167388916015625,
-0.04150390625,
0.007534027099609375,
0.02728271484375,
-0.0203094482421875,
0.01355743408203125,
-0.020660400390625,
0.01049041748046875,
0.0291900634765625,
0.031219482421875,
-0.0338134765625,
0.0361328125,
-0.012847900390625,
0.053009033203125,
0.0335693359375,
0.0014562606811523438,
0.021148681640625,
-0.01424407958984375,
0.03497314453125,
0.002666473388671875,
0.034698486328125,
-0.0209197998046875,
-0.0310516357421875,
-0.0682373046875,
-0.01568603515625,
0.024261474609375,
0.04296875,
-0.050994873046875,
0.060546875,
-0.01387786865234375,
-0.039154052734375,
-0.036956787109375,
0.00495147705078125,
0.04730224609375,
0.03338623046875,
0.047698974609375,
-0.0083465576171875,
-0.040374755859375,
-0.05401611328125,
0.019439697265625,
-0.00971221923828125,
0.00823211669921875,
0.024993896484375,
0.0611572265625,
-0.0338134765625,
0.05438232421875,
-0.050506591796875,
0.00411224365234375,
-0.01494598388671875,
-0.0107574462890625,
0.03460693359375,
0.0592041015625,
0.049285888671875,
-0.043609619140625,
-0.02099609375,
-0.01209259033203125,
-0.0455322265625,
0.01457977294921875,
-0.0020122528076171875,
-0.01255035400390625,
0.01338958740234375,
0.04229736328125,
-0.06805419921875,
0.0308380126953125,
0.031524658203125,
-0.03271484375,
0.06048583984375,
-0.0170135498046875,
0.02264404296875,
-0.09246826171875,
0.03607177734375,
-0.01367950439453125,
0.002941131591796875,
-0.0244903564453125,
0.0276336669921875,
0.0188446044921875,
-0.02728271484375,
-0.026580810546875,
0.032257080078125,
-0.0355224609375,
-0.01265716552734375,
-0.0181427001953125,
-0.024322509765625,
-0.006999969482421875,
0.057861328125,
-0.002048492431640625,
0.06829833984375,
0.050567626953125,
-0.051849365234375,
0.024017333984375,
0.0190887451171875,
-0.0104217529296875,
-0.004108428955078125,
-0.0740966796875,
0.0214080810546875,
-0.0005540847778320312,
0.01299285888671875,
-0.07855224609375,
-0.01629638671875,
0.04486083984375,
-0.053619384765625,
0.0145263671875,
-0.029937744140625,
-0.0382080078125,
-0.05767822265625,
-0.00585174560546875,
0.034698486328125,
0.055511474609375,
-0.044525146484375,
0.0191802978515625,
0.010711669921875,
0.0003724098205566406,
-0.045318603515625,
-0.03125,
-0.00646209716796875,
0.0008230209350585938,
-0.040191650390625,
0.01477813720703125,
-0.01444244384765625,
0.0223541259765625,
0.0051422119140625,
0.003330230712890625,
-0.01275634765625,
-0.01091766357421875,
0.0223236083984375,
0.032135009765625,
-0.018157958984375,
-0.014862060546875,
-0.0101776123046875,
-0.0173187255859375,
0.0196380615234375,
-0.041351318359375,
0.05047607421875,
-0.03326416015625,
-0.03289794921875,
-0.0293426513671875,
0.0180816650390625,
0.055633544921875,
-0.0301971435546875,
0.03936767578125,
0.062225341796875,
-0.029937744140625,
-0.00310516357421875,
-0.0307464599609375,
-0.0206756591796875,
-0.040374755859375,
0.04736328125,
-0.021240234375,
-0.061004638671875,
0.041839599609375,
0.023529052734375,
0.006069183349609375,
0.0439453125,
0.0386962890625,
0.01026153564453125,
0.058837890625,
0.05328369140625,
-0.0205078125,
0.0270233154296875,
-0.053497314453125,
0.022003173828125,
-0.056396484375,
-0.0260772705078125,
-0.054046630859375,
-0.0186309814453125,
-0.0321044921875,
-0.047515869140625,
0.0152435302734375,
0.0265960693359375,
-0.0418701171875,
0.05682373046875,
-0.06573486328125,
0.035064697265625,
0.04583740234375,
0.0030918121337890625,
-0.0117034912109375,
-0.006687164306640625,
-0.006412506103515625,
0.00830841064453125,
-0.06610107421875,
-0.033538818359375,
0.07928466796875,
0.038330078125,
0.057830810546875,
-0.010650634765625,
0.05706787109375,
0.00003707408905029297,
0.006610870361328125,
-0.05078125,
0.04443359375,
-0.001514434814453125,
-0.03704833984375,
-0.0183563232421875,
-0.04248046875,
-0.06988525390625,
-0.006320953369140625,
0.006805419921875,
-0.06463623046875,
0.0141754150390625,
0.025848388671875,
-0.04388427734375,
0.01983642578125,
-0.068603515625,
0.08746337890625,
-0.0229339599609375,
-0.0260009765625,
-0.0021800994873046875,
-0.050048828125,
0.035125732421875,
0.0147552490234375,
-0.006397247314453125,
0.0175323486328125,
0.016265869140625,
0.06585693359375,
-0.045379638671875,
0.06097412109375,
-0.0264129638671875,
0.0135345458984375,
0.0299224853515625,
-0.00922393798828125,
0.043731689453125,
0.013702392578125,
0.007671356201171875,
0.02203369140625,
0.007633209228515625,
-0.03485107421875,
-0.0304718017578125,
0.058135986328125,
-0.07855224609375,
-0.030303955078125,
-0.0151214599609375,
-0.0301055908203125,
0.006069183349609375,
0.0276031494140625,
0.03955078125,
0.029144287109375,
0.01258087158203125,
0.00539398193359375,
0.0367431640625,
-0.0206146240234375,
0.046142578125,
0.0246124267578125,
-0.016204833984375,
-0.033233642578125,
0.0673828125,
0.003261566162109375,
0.0010089874267578125,
0.020782470703125,
0.00026535987854003906,
-0.0343017578125,
-0.03778076171875,
-0.05181884765625,
0.0225677490234375,
-0.057342529296875,
-0.019012451171875,
-0.056182861328125,
-0.033050537109375,
-0.0386962890625,
-0.0159759521484375,
-0.040771484375,
-0.009918212890625,
-0.024078369140625,
0.00698089599609375,
0.030914306640625,
0.039520263671875,
0.01181793212890625,
0.0240936279296875,
-0.07122802734375,
0.0205078125,
0.0167236328125,
0.028350830078125,
0.006793975830078125,
-0.04534912109375,
-0.035125732421875,
0.00656890869140625,
-0.020477294921875,
-0.0494384765625,
0.018310546875,
-0.0168914794921875,
0.03704833984375,
0.01312255859375,
-0.0032749176025390625,
0.054290771484375,
-0.040130615234375,
0.06903076171875,
0.0167694091796875,
-0.0755615234375,
0.0292510986328125,
-0.010009765625,
0.032684326171875,
0.035552978515625,
0.0386962890625,
-0.0263671875,
-0.00739288330078125,
-0.047119140625,
-0.069580078125,
0.05535888671875,
0.0124664306640625,
0.00560760498046875,
-0.004222869873046875,
0.03167724609375,
-0.00815582275390625,
0.0038127899169921875,
-0.052642822265625,
-0.01438140869140625,
-0.03448486328125,
-0.0107574462890625,
-0.0185699462890625,
-0.0143585205078125,
-0.00035309791564941406,
-0.03125,
0.035003662109375,
-0.00327301025390625,
0.07257080078125,
0.01212310791015625,
-0.0213623046875,
-0.0021686553955078125,
0.005340576171875,
0.061767578125,
0.06011962890625,
-0.011260986328125,
-0.001506805419921875,
-0.0032405853271484375,
-0.052215576171875,
0.0012359619140625,
0.042694091796875,
-0.005523681640625,
-0.0168609619140625,
0.027069091796875,
0.08001708984375,
0.00951385498046875,
-0.0251922607421875,
0.039825439453125,
-0.0015840530395507812,
-0.0187835693359375,
-0.0280303955078125,
0.02301025390625,
0.0106201171875,
0.0230255126953125,
0.0307464599609375,
0.01474761962890625,
0.003177642822265625,
-0.0155029296875,
0.0279083251953125,
0.0113372802734375,
-0.0196685791015625,
-0.031280517578125,
0.075439453125,
0.00762176513671875,
-0.0278472900390625,
0.057159423828125,
-0.018524169921875,
-0.049041748046875,
0.076171875,
0.042999267578125,
0.0628662109375,
0.002277374267578125,
-0.00022339820861816406,
0.05267333984375,
0.037109375,
0.002330780029296875,
0.0238494873046875,
-0.007694244384765625,
-0.037994384765625,
-0.0341796875,
-0.046600341796875,
-0.01316070556640625,
0.013824462890625,
-0.04296875,
0.0199432373046875,
-0.05267333984375,
-0.0089263916015625,
0.0029163360595703125,
0.0130462646484375,
-0.0767822265625,
0.0119476318359375,
0.01012420654296875,
0.0733642578125,
-0.054046630859375,
0.069580078125,
0.0550537109375,
-0.04144287109375,
-0.0823974609375,
0.00836181640625,
-0.022857666015625,
-0.063232421875,
0.053192138671875,
0.025634765625,
0.0049591064453125,
0.01029205322265625,
-0.0511474609375,
-0.07562255859375,
0.0853271484375,
0.01861572265625,
-0.051788330078125,
-0.00788116455078125,
0.0016527175903320312,
0.028564453125,
-0.020050048828125,
0.033050537109375,
0.0294952392578125,
0.033966064453125,
0.0109100341796875,
-0.0687255859375,
0.02593994140625,
-0.026763916015625,
0.0019855499267578125,
0.0108184814453125,
-0.056671142578125,
0.07427978515625,
-0.033538818359375,
0.0076904296875,
0.002498626708984375,
0.04931640625,
0.02191162109375,
0.0098876953125,
0.018157958984375,
0.034881591796875,
0.04632568359375,
-0.002445220947265625,
0.08270263671875,
-0.0633544921875,
0.046722412109375,
0.06658935546875,
-0.0020771026611328125,
0.053192138671875,
0.0288543701171875,
-0.0171966552734375,
0.0167236328125,
0.04071044921875,
-0.0292510986328125,
0.0310516357421875,
0.0107879638671875,
0.00830078125,
0.0015592575073242188,
0.037139892578125,
-0.040802001953125,
0.0263671875,
0.0215301513671875,
-0.03497314453125,
-0.01045989990234375,
-0.0032100677490234375,
0.005767822265625,
-0.03204345703125,
-0.022308349609375,
0.0291900634765625,
-0.006130218505859375,
-0.063232421875,
0.086669921875,
0.021942138671875,
0.0487060546875,
-0.0511474609375,
-0.004425048828125,
-0.011322021484375,
0.018798828125,
-0.0139312744140625,
-0.049652099609375,
0.0205078125,
-0.0008511543273925781,
-0.0295562744140625,
0.004894256591796875,
0.03985595703125,
-0.0283355712890625,
-0.03369140625,
0.01294708251953125,
0.003833770751953125,
0.0203094482421875,
0.008697509765625,
-0.06622314453125,
0.02947998046875,
0.01099395751953125,
-0.0243988037109375,
0.0125579833984375,
0.01464080810546875,
0.0054931640625,
0.04296875,
0.0526123046875,
-0.01325225830078125,
0.01910400390625,
-0.019989013671875,
0.07958984375,
-0.06829833984375,
-0.04168701171875,
-0.05828857421875,
0.04437255859375,
0.00748443603515625,
-0.051483154296875,
0.04937744140625,
0.05328369140625,
0.07025146484375,
-0.031982421875,
0.052459716796875,
-0.032379150390625,
0.0038127899169921875,
-0.053131103515625,
0.052734375,
-0.019775390625,
0.0196380615234375,
-0.020751953125,
-0.078125,
-0.00966644287109375,
0.052001953125,
-0.039886474609375,
0.026580810546875,
0.047119140625,
0.0828857421875,
-0.02264404296875,
-0.00952911376953125,
0.011383056640625,
0.005950927734375,
0.0220794677734375,
0.07244873046875,
0.04296875,
-0.051544189453125,
0.04852294921875,
-0.0213623046875,
-0.0203399658203125,
-0.0286865234375,
-0.04364013671875,
-0.06494140625,
-0.0455322265625,
-0.01837158203125,
-0.034210205078125,
0.0129241943359375,
0.07861328125,
0.06646728515625,
-0.057525634765625,
-0.02069091796875,
-0.005962371826171875,
-0.00279998779296875,
-0.025848388671875,
-0.0176239013671875,
0.04547119140625,
-0.0288543701171875,
-0.0494384765625,
0.01264190673828125,
-0.01128387451171875,
0.0014934539794921875,
-0.0114288330078125,
-0.0297698974609375,
-0.0013380050659179688,
-0.00817108154296875,
0.0333251953125,
0.0469970703125,
-0.039276123046875,
-0.02435302734375,
0.0166168212890625,
-0.031097412109375,
0.0008792877197265625,
0.0528564453125,
-0.03411865234375,
0.013153076171875,
0.025177001953125,
0.055938720703125,
0.051483154296875,
-0.003398895263671875,
0.0175323486328125,
-0.042877197265625,
0.0091552734375,
0.01403045654296875,
0.0226898193359375,
0.00914764404296875,
-0.0469970703125,
0.0157623291015625,
0.019134521484375,
-0.06390380859375,
-0.03973388671875,
0.0090179443359375,
-0.07275390625,
-0.0215911865234375,
0.099365234375,
-0.0110015869140625,
-0.0269775390625,
0.0134735107421875,
-0.0170135498046875,
0.019683837890625,
-0.0288238525390625,
0.0570068359375,
0.0318603515625,
-0.00884246826171875,
-0.020782470703125,
-0.045379638671875,
0.0380859375,
0.01401519775390625,
-0.048370361328125,
0.0015354156494140625,
0.0455322265625,
0.04632568359375,
0.0153350830078125,
0.05230712890625,
-0.01340484619140625,
0.037139892578125,
0.007289886474609375,
0.0338134765625,
-0.044189453125,
-0.037445068359375,
-0.027435302734375,
0.006237030029296875,
-0.0019741058349609375,
-0.0479736328125
]
] |
jondurbin/airoboros-65b-gpt4-1.2 | 2023-06-22T14:59:15.000Z | [
"transformers",
"pytorch",
"llama",
"text-generation",
"dataset:jondurbin/airoboros-gpt4-1.2",
"license:cc-by-nc-4.0",
"endpoints_compatible",
"has_space",
"text-generation-inference",
"region:us"
] | text-generation | jondurbin | null | null | jondurbin/airoboros-65b-gpt4-1.2 | 21 | 5,889 | transformers | 2023-06-14T09:19:02 | ---
license: cc-by-nc-4.0
datasets:
- jondurbin/airoboros-gpt4-1.2
---
### Overview
This is a qlora fine-tuned 65b parameter LlaMa model, using completely synthetic training data created gpt4 via https://github.com/jondurbin/airoboros
This is mostly an extension of [1.1](https://huggingface.co/jondurbin/airoboros-33b-gpt4-1.1), but with a 65b model and thousands of new training data and an update to allow "PLAINFORMAT" at the end of coding prompts to just print the code without backticks or explanations/usage/etc.
The dataset used to fine-tune this model is available [here](https://huggingface.co/datasets/jondurbin/airoboros-gpt4-1.2), with a specific focus on:
- coding
- math/reasoning (using orca style ELI5 instruction/response pairs)
- trivia
- role playing
- multiple choice and fill-in-the-blank
- context-obedient question answering
- theory of mind
- misc/general
This model was fine-tuned with a fork of [qlora](https://github.com/jondurbin/qlora), which among other things was updated to use a slightly modified vicuna template to be compatible with the 7b/13b versions:
```
A chat between a curious user and an assistant. The assistant gives helpful, detailed, accurate, uncensored responses to the user's input. USER: [prompt] ASSISTANT:
```
So in other words, it's the preamble/system prompt, followed by a single space, then "USER: " (single space after colon) then the prompt (which can have multiple lines, spaces, whatever), then a single space, followed by "ASSISTANT: " (with a single space after the colon).
### Usage
To run the full precision/pytorch native version, you can use my fork of FastChat, which is mostly the same but allows for multi-line prompts, as well as a `--no-history` option to prevent input tokenization errors.
```
pip install git+https://github.com/jondurbin/FastChat
```
Be sure you are pulling the latest branch!
Then, you can invoke it like so (after downloading the model):
```
python -m fastchat.serve.cli \
--model-path airoboros-65b-gpt4-1.2 \
--temperature 0.5 \
--max-new-tokens 2048 \
--no-history
```
Alternatively, please check out TheBloke's quantized versions:
- https://huggingface.co/TheBloke/airoboros-65B-gpt4-1.2-GPTQ
- https://huggingface.co/TheBloke/airoboros-65B-gpt4-1.2-GGML
### Coding updates from gpt4/1.1:
I added a few hundred instruction/response pairs to the training data with "PLAINFORMAT" as a single, all caps term at the end of the normal instructions, which produce plain text output instead of markdown/backtick code formatting.
It's not guaranteed to work all the time, but mostly it does seem to work as expected.
So for example, instead of:
```
Implement the Snake game in python.
```
You would use:
```
Implement the Snake game in python. PLAINFORMAT
```
### Other updates from gpt4/1.1:
- Several hundred role-playing data.
- A few thousand ORCA style reasoning/math questions with ELI5 prompts to generate the responses (should not be needed in your prompts to this model however, just ask the question).
- Many more coding examples in various languages, including some that use specific libraries (pandas, numpy, tensorflow, etc.)
### Usage and License Notices
All airoboros models and datasets are intended and licensed for research use only. I've used the 'cc-nc-4.0' license, but really it is subject to a custom/special license because:
- the base model is LLaMa, which has it's own special research license
- the dataset(s) were generated with OpenAI (gpt-4 and/or gpt-3.5-turbo), which has a clausing saying the data can't be used to create models to compete with openai
So, to reiterate: this model (and datasets) cannot be used commercially. | 3,681 | [
[
-0.01352691650390625,
-0.06732177734375,
0.0099334716796875,
0.01490020751953125,
-0.02569580078125,
-0.021209716796875,
-0.004489898681640625,
-0.0275115966796875,
0.021636962890625,
0.0234832763671875,
-0.039642333984375,
-0.039215087890625,
-0.02008056640625,
-0.0006046295166015625,
-0.0194549560546875,
0.09393310546875,
-0.0007834434509277344,
0.0005435943603515625,
0.0185699462890625,
-0.006725311279296875,
-0.03955078125,
-0.054351806640625,
-0.06304931640625,
-0.01654052734375,
0.03857421875,
0.0250091552734375,
0.059356689453125,
0.059356689453125,
0.020660400390625,
0.0185089111328125,
-0.00699615478515625,
0.01450347900390625,
-0.045806884765625,
-0.01177978515625,
0.007205963134765625,
-0.038818359375,
-0.04248046875,
0.0006589889526367188,
0.039794921875,
0.0079193115234375,
-0.034027099609375,
0.013092041015625,
-0.01171112060546875,
0.01204681396484375,
-0.0306396484375,
0.026275634765625,
-0.03533935546875,
-0.00894927978515625,
-0.01218414306640625,
-0.0135040283203125,
-0.008880615234375,
-0.032562255859375,
0.001873016357421875,
-0.06500244140625,
0.0004010200500488281,
-0.00321197509765625,
0.0870361328125,
0.011322021484375,
-0.044097900390625,
-0.032958984375,
-0.038818359375,
0.038909912109375,
-0.08319091796875,
0.006649017333984375,
0.01763916015625,
0.039459228515625,
-0.01885986328125,
-0.06280517578125,
-0.04107666015625,
-0.023712158203125,
0.008941650390625,
0.01087188720703125,
-0.018798828125,
0.01233673095703125,
0.03497314453125,
0.0218353271484375,
-0.054412841796875,
-0.0008516311645507812,
-0.05194091796875,
-0.016815185546875,
0.04852294921875,
0.0194549560546875,
0.0095977783203125,
-0.01514434814453125,
-0.0265045166015625,
-0.0173187255859375,
-0.04815673828125,
0.0164794921875,
0.035125732421875,
0.01311492919921875,
-0.035736083984375,
0.048797607421875,
-0.0261383056640625,
0.055145263671875,
-0.0030975341796875,
-0.00693511962890625,
0.01119232177734375,
-0.0163116455078125,
-0.0308990478515625,
-0.00670623779296875,
0.06658935546875,
0.0275115966796875,
0.020172119140625,
0.01204681396484375,
-0.00795745849609375,
0.00782012939453125,
-0.004055023193359375,
-0.0709228515625,
-0.0386962890625,
0.02154541015625,
-0.0196685791015625,
-0.037567138671875,
-0.01117706298828125,
-0.033050537109375,
-0.0098876953125,
-0.0171966552734375,
0.038970947265625,
-0.032135009765625,
-0.0175018310546875,
0.0139312744140625,
-0.0038280487060546875,
0.018157958984375,
0.0288238525390625,
-0.056640625,
0.0193023681640625,
0.02130126953125,
0.06658935546875,
0.0145416259765625,
-0.0231170654296875,
-0.0185699462890625,
-0.011474609375,
-0.0262908935546875,
0.05517578125,
-0.029876708984375,
-0.024749755859375,
-0.025848388671875,
-0.00452423095703125,
0.006946563720703125,
-0.0243988037109375,
0.0249481201171875,
-0.044921875,
0.0225677490234375,
-0.01666259765625,
-0.037445068359375,
-0.02447509765625,
0.0251617431640625,
-0.042266845703125,
0.071044921875,
0.004119873046875,
-0.056427001953125,
-0.0003998279571533203,
-0.06390380859375,
-0.002964019775390625,
-0.0093994140625,
0.0011043548583984375,
-0.01386260986328125,
-0.0175323486328125,
0.020904541015625,
0.0220489501953125,
-0.0218505859375,
0.027008056640625,
-0.032501220703125,
-0.04815673828125,
0.030487060546875,
-0.04595947265625,
0.0836181640625,
0.023895263671875,
-0.023162841796875,
0.0182647705078125,
-0.0548095703125,
0.01190948486328125,
0.01366424560546875,
-0.040069580078125,
0.004138946533203125,
-0.032379150390625,
0.0028514862060546875,
0.01490020751953125,
0.0242767333984375,
-0.020538330078125,
0.048492431640625,
-0.0244598388671875,
0.05487060546875,
0.05303955078125,
0.01288604736328125,
0.008636474609375,
-0.030364990234375,
0.051971435546875,
-0.005229949951171875,
0.0272064208984375,
-0.006256103515625,
-0.052459716796875,
-0.04925537109375,
-0.018157958984375,
0.0254669189453125,
0.040802001953125,
-0.07073974609375,
0.01715087890625,
-0.00574493408203125,
-0.06134033203125,
-0.0298004150390625,
-0.003192901611328125,
0.0299072265625,
0.037078857421875,
0.026641845703125,
-0.023193359375,
-0.045745849609375,
-0.04278564453125,
0.00408935546875,
-0.025238037109375,
0.00777435302734375,
0.0286712646484375,
0.04083251953125,
-0.02099609375,
0.06329345703125,
-0.04437255859375,
-0.0009336471557617188,
-0.0162506103515625,
0.004863739013671875,
0.0179443359375,
0.041412353515625,
0.045196533203125,
-0.0469970703125,
-0.0179443359375,
-0.016265869140625,
-0.055816650390625,
-0.0033206939697265625,
0.001556396484375,
-0.032073974609375,
0.0003631114959716797,
0.01363372802734375,
-0.054351806640625,
0.050018310546875,
0.048828125,
-0.0228271484375,
0.052764892578125,
-0.01422119140625,
-0.004581451416015625,
-0.09185791015625,
0.02252197265625,
0.0023899078369140625,
-0.00643157958984375,
-0.044921875,
0.017974853515625,
-0.010040283203125,
-0.009124755859375,
-0.05120849609375,
0.062408447265625,
-0.0142059326171875,
0.002079010009765625,
-0.018585205078125,
-0.01181793212890625,
0.005702972412109375,
0.054534912109375,
0.003448486328125,
0.05224609375,
0.042572021484375,
-0.044219970703125,
0.057647705078125,
0.0213165283203125,
-0.002269744873046875,
0.01248931884765625,
-0.082275390625,
0.0296478271484375,
0.00839996337890625,
0.049041748046875,
-0.062103271484375,
-0.030059814453125,
0.05010986328125,
-0.045318603515625,
0.00046515464782714844,
-0.023468017578125,
-0.034637451171875,
-0.024444580078125,
-0.0146331787109375,
0.040679931640625,
0.039764404296875,
-0.032928466796875,
0.032684326171875,
0.01535797119140625,
0.013397216796875,
-0.0439453125,
-0.0445556640625,
-0.0022525787353515625,
-0.0273284912109375,
-0.0301361083984375,
0.0024127960205078125,
-0.015899658203125,
-0.0013036727905273438,
-0.005519866943359375,
-0.0021152496337890625,
-0.03302001953125,
0.006984710693359375,
0.019500732421875,
0.02276611328125,
-0.0146942138671875,
-0.0054168701171875,
0.001190185546875,
0.00789642333984375,
0.0004684925079345703,
-0.0235443115234375,
0.048248291015625,
-0.03143310546875,
0.001140594482421875,
-0.05426025390625,
0.003231048583984375,
0.02520751953125,
-0.01412200927734375,
0.041748046875,
0.051605224609375,
-0.031402587890625,
-0.0012331008911132812,
-0.0230712890625,
-0.02642822265625,
-0.0401611328125,
0.0185394287109375,
-0.020721435546875,
-0.050323486328125,
0.0477294921875,
0.0248565673828125,
0.0172576904296875,
0.0245208740234375,
0.038909912109375,
-0.0009021759033203125,
0.0654296875,
0.0299224853515625,
-0.00014483928680419922,
0.03631591796875,
-0.044158935546875,
-0.01207733154296875,
-0.051483154296875,
-0.0226898193359375,
-0.02490234375,
-0.01739501953125,
-0.042510986328125,
-0.025543212890625,
0.0262603759765625,
0.01416015625,
-0.045684814453125,
0.040374755859375,
-0.047882080078125,
0.0294189453125,
0.04779052734375,
0.0238189697265625,
0.0125274658203125,
-0.00823211669921875,
0.004718780517578125,
0.007167816162109375,
-0.07232666015625,
-0.041015625,
0.09454345703125,
0.0238494873046875,
0.0687255859375,
0.005157470703125,
0.066650390625,
0.0010700225830078125,
0.037384033203125,
-0.04022216796875,
0.033111572265625,
0.00982666015625,
-0.071044921875,
-0.0140380859375,
-0.05389404296875,
-0.08172607421875,
0.0169525146484375,
-0.00009524822235107422,
-0.051513671875,
0.0063934326171875,
0.02032470703125,
-0.048309326171875,
0.01348876953125,
-0.06365966796875,
0.0745849609375,
-0.029296875,
-0.0169219970703125,
0.00868988037109375,
-0.04949951171875,
0.047149658203125,
0.01233673095703125,
0.00402069091796875,
-0.005809783935546875,
0.0021610260009765625,
0.069580078125,
-0.04364013671875,
0.080078125,
-0.0188446044921875,
-0.001392364501953125,
0.049560546875,
0.00783538818359375,
0.0170135498046875,
0.0252227783203125,
-0.005283355712890625,
0.022369384765625,
0.0310211181640625,
-0.037811279296875,
-0.04656982421875,
0.05633544921875,
-0.08624267578125,
-0.0254058837890625,
-0.041229248046875,
-0.04803466796875,
0.00931549072265625,
0.0218963623046875,
0.01678466796875,
0.0504150390625,
0.006519317626953125,
0.00014901161193847656,
0.0216217041015625,
-0.0252838134765625,
0.01294708251953125,
0.0372314453125,
-0.02105712890625,
-0.04461669921875,
0.06927490234375,
0.0086212158203125,
0.007083892822265625,
0.0179443359375,
0.0229949951171875,
-0.03570556640625,
-0.0188751220703125,
-0.04107666015625,
0.026824951171875,
-0.05902099609375,
-0.02410888671875,
-0.033905029296875,
-0.0169525146484375,
-0.042022705078125,
0.004146575927734375,
-0.01812744140625,
-0.03594970703125,
-0.04498291015625,
-0.007770538330078125,
0.04266357421875,
0.06439208984375,
0.0010385513305664062,
0.04241943359375,
-0.047393798828125,
0.0245513916015625,
0.022613525390625,
0.0029430389404296875,
0.0013990402221679688,
-0.049224853515625,
-0.01456451416015625,
0.034637451171875,
-0.037841796875,
-0.07513427734375,
0.044097900390625,
0.00870513916015625,
0.02923583984375,
0.02960205078125,
0.0030612945556640625,
0.07177734375,
-0.00870513916015625,
0.0689697265625,
-0.0023136138916015625,
-0.07098388671875,
0.03497314453125,
-0.03436279296875,
0.01119232177734375,
0.022613525390625,
0.0275421142578125,
-0.0310516357421875,
-0.008697509765625,
-0.06378173828125,
-0.05169677734375,
0.058563232421875,
0.0309600830078125,
0.010345458984375,
0.010498046875,
0.054779052734375,
-0.003353118896484375,
0.0289306640625,
-0.0660400390625,
-0.01461029052734375,
-0.0477294921875,
-0.0165557861328125,
-0.006595611572265625,
0.00846099853515625,
-0.006481170654296875,
-0.032867431640625,
0.048370361328125,
-0.01434326171875,
0.035675048828125,
0.00801849365234375,
0.01119232177734375,
0.01412200927734375,
-0.0056915283203125,
0.049224853515625,
0.038726806640625,
-0.0245513916015625,
0.00007098913192749023,
0.00513458251953125,
-0.028076171875,
0.01102447509765625,
0.01155853271484375,
0.0011339187622070312,
-0.005947113037109375,
0.01885986328125,
0.07147216796875,
0.014068603515625,
-0.039794921875,
0.026947021484375,
-0.0155792236328125,
-0.0164337158203125,
-0.0087432861328125,
0.026153564453125,
0.007205963134765625,
0.0174713134765625,
0.0077056884765625,
-0.0003058910369873047,
0.0046844482421875,
-0.04925537109375,
0.00850677490234375,
0.0106048583984375,
0.00135040283203125,
-0.03851318359375,
0.0654296875,
0.017578125,
-0.0167083740234375,
0.05419921875,
-0.0228118896484375,
-0.041900634765625,
0.06768798828125,
0.03271484375,
0.038116455078125,
-0.00670623779296875,
0.0184783935546875,
0.02996826171875,
0.01514434814453125,
-0.012603759765625,
0.03961181640625,
-0.005702972412109375,
-0.042449951171875,
-0.020660400390625,
-0.025360107421875,
-0.052459716796875,
0.012847900390625,
-0.050445556640625,
0.0276336669921875,
-0.045806884765625,
-0.0200958251953125,
-0.016082763671875,
0.02545166015625,
-0.0628662109375,
0.02252197265625,
0.0026836395263671875,
0.0716552734375,
-0.0579833984375,
0.0833740234375,
0.050872802734375,
-0.054718017578125,
-0.10479736328125,
-0.0136260986328125,
-0.0089569091796875,
-0.06878662109375,
0.036712646484375,
0.010284423828125,
0.01552581787109375,
0.0186614990234375,
-0.0599365234375,
-0.07421875,
0.10235595703125,
0.041717529296875,
-0.0236358642578125,
-0.033935546875,
-0.0003662109375,
0.049407958984375,
-0.0159454345703125,
0.060211181640625,
0.034210205078125,
0.023162841796875,
0.011932373046875,
-0.08013916015625,
0.01470947265625,
-0.028594970703125,
0.004482269287109375,
-0.037445068359375,
-0.07342529296875,
0.08172607421875,
-0.01556396484375,
-0.00328826904296875,
0.0164794921875,
0.046234130859375,
0.039398193359375,
0.0167694091796875,
0.0158233642578125,
0.032196044921875,
0.062347412109375,
-0.004413604736328125,
0.08154296875,
-0.0296783447265625,
0.036529541015625,
0.0709228515625,
-0.0148773193359375,
0.05206298828125,
0.015625,
-0.006649017333984375,
0.041473388671875,
0.069091796875,
-0.00017833709716796875,
0.034454345703125,
0.016021728515625,
0.007175445556640625,
-0.0007576942443847656,
0.01070404052734375,
-0.042877197265625,
0.031890869140625,
0.0242462158203125,
0.0027179718017578125,
-0.01235198974609375,
-0.01119232177734375,
0.0033321380615234375,
-0.04522705078125,
-0.006023406982421875,
0.03985595703125,
0.01568603515625,
-0.047088623046875,
0.08392333984375,
0.01358795166015625,
0.06317138671875,
-0.05694580078125,
-0.01528167724609375,
-0.027435302734375,
0.004451751708984375,
-0.0225372314453125,
-0.041595458984375,
0.00186920166015625,
0.010040283203125,
0.024566650390625,
0.007480621337890625,
0.035430908203125,
-0.028900146484375,
-0.0215911865234375,
-0.005706787109375,
0.020660400390625,
0.047637939453125,
0.0037708282470703125,
-0.06768798828125,
0.02178955078125,
0.004138946533203125,
-0.0261993408203125,
0.0126190185546875,
0.0308685302734375,
-0.003162384033203125,
0.060821533203125,
0.0628662109375,
0.00972747802734375,
0.0166778564453125,
-0.00873565673828125,
0.08062744140625,
-0.04266357421875,
-0.048797607421875,
-0.047882080078125,
0.03253173828125,
0.01418304443359375,
-0.05224609375,
0.05810546875,
0.037689208984375,
0.061920166015625,
-0.00020945072174072266,
0.056610107421875,
-0.005268096923828125,
0.002285003662109375,
-0.041229248046875,
0.05181884765625,
-0.03790283203125,
0.021209716796875,
-0.0074920654296875,
-0.059295654296875,
-0.0104522705078125,
0.06707763671875,
-0.0180511474609375,
0.020843505859375,
0.033477783203125,
0.07098388671875,
-0.00464630126953125,
-0.0013065338134765625,
0.01438140869140625,
0.01403045654296875,
0.045166015625,
0.068359375,
0.06524658203125,
-0.048736572265625,
0.062255859375,
-0.039886474609375,
-0.019134521484375,
-0.033447265625,
-0.054718017578125,
-0.0718994140625,
-0.023773193359375,
-0.025543212890625,
-0.043060302734375,
0.0170135498046875,
0.08013916015625,
0.0645751953125,
-0.042449951171875,
-0.031097412109375,
-0.007808685302734375,
-0.005512237548828125,
-0.018646240234375,
-0.0198211669921875,
0.02197265625,
0.004665374755859375,
-0.04620361328125,
0.02252197265625,
-0.02996826171875,
0.0279083251953125,
-0.0238494873046875,
-0.0162353515625,
0.00018787384033203125,
0.0141143798828125,
0.01422119140625,
0.04266357421875,
-0.049652099609375,
-0.00560760498046875,
0.0106201171875,
-0.01345062255859375,
0.0036907196044921875,
0.035614013671875,
-0.05889892578125,
0.0242767333984375,
0.0245208740234375,
0.0173797607421875,
0.0419921875,
0.0088348388671875,
0.0275726318359375,
-0.04754638671875,
0.0135040283203125,
0.019561767578125,
0.0178680419921875,
0.0211334228515625,
-0.0290679931640625,
0.0330810546875,
0.02203369140625,
-0.047576904296875,
-0.06903076171875,
-0.00821685791015625,
-0.07952880859375,
-0.0059051513671875,
0.108154296875,
-0.0243682861328125,
-0.02459716796875,
0.00789642333984375,
-0.05120849609375,
0.036529541015625,
-0.0537109375,
0.056732177734375,
0.034454345703125,
-0.0165557861328125,
-0.003757476806640625,
-0.035552978515625,
0.0270538330078125,
0.00795745849609375,
-0.062255859375,
-0.00662994384765625,
0.03759765625,
0.041595458984375,
-0.0089263916015625,
0.06195068359375,
0.00681304931640625,
0.0284423828125,
0.0008053779602050781,
0.007354736328125,
-0.036102294921875,
-0.0215911865234375,
-0.042205810546875,
0.00421905517578125,
-0.0009722709655761719,
-0.0192108154296875
]
] |
FelixChao/vicuna-33b-coder | 2023-09-11T05:22:28.000Z | [
"transformers",
"pytorch",
"llama",
"text-generation",
"code",
"arxiv:1910.09700",
"license:apache-2.0",
"model-index",
"endpoints_compatible",
"has_space",
"text-generation-inference",
"region:us"
] | text-generation | FelixChao | null | null | FelixChao/vicuna-33b-coder | 2 | 5,888 | transformers | 2023-08-22T02:15:25 | ---
tags:
- code
license: apache-2.0
model-index:
- name: Vicuna-Coder
results:
- task:
type: text-generation # Required. Example: automatic-speech-recognition
dataset:
type: nuprl/MultiPL-E # Required. Example: common_voice. Use dataset id from https://hf.co/datasets
name: MultiPL-HumanEval (Python) # Required. A pretty name for the dataset. Example: Common Voice (French)
metrics:
- type: pass@1 # Required. Example: wer. Use metric id from https://hf.co/metrics
value: 0.274 # Required. Example: 20.90
name: pass@1 # Optional. Example: Test WER
verified: false
---
---
# Model Card for Model ID
<!-- Provide a quick summary of what the model is/does. -->
This modelcard aims to be a base template for new models. It has been generated using [this raw template](https://github.com/huggingface/huggingface_hub/blob/main/src/huggingface_hub/templates/modelcard_template.md?plain=1).
## Model Details
### Model Description
<!-- Provide a longer summary of what this model is. -->
- **Developed by:** [More Information Needed]
- **Shared by [optional]:** [More Information Needed]
- **Model type:** [More Information Needed]
- **Language(s) (NLP):** [More Information Needed]
- **License:** [More Information Needed]
- **Finetuned from model [optional]:** [More Information Needed]
### Model Sources [optional]
<!-- Provide the basic links for the model. -->
- **Repository:** [More Information Needed]
- **Paper [optional]:** [More Information Needed]
- **Demo [optional]:** [More Information Needed]
## Uses
<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
### Direct Use
<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
[More Information Needed]
### Downstream Use [optional]
<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
[More Information Needed]
### Out-of-Scope Use
<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
[More Information Needed]
## Bias, Risks, and Limitations
<!-- This section is meant to convey both technical and sociotechnical limitations. -->
[More Information Needed]
### Recommendations
<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
## How to Get Started with the Model
Use the code below to get started with the model.
[More Information Needed]
## Training Details
### Training Data
<!-- This should link to a Data Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
[More Information Needed]
### Training Procedure
<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
#### Preprocessing [optional]
[More Information Needed]
#### Training Hyperparameters
- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
#### Speeds, Sizes, Times [optional]
<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
[More Information Needed]
## Evaluation
<!-- This section describes the evaluation protocols and provides the results. -->
### Testing Data, Factors & Metrics
#### Testing Data
<!-- This should link to a Data Card if possible. -->
[More Information Needed]
#### Factors
<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
[More Information Needed]
#### Metrics
<!-- These are the evaluation metrics being used, ideally with a description of why. -->
[More Information Needed]
### Results
[More Information Needed]
#### Summary
## Model Examination [optional]
<!-- Relevant interpretability work for the model goes here -->
[More Information Needed]
## Environmental Impact
<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
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).
- **Hardware Type:** [More Information Needed]
- **Hours used:** [More Information Needed]
- **Cloud Provider:** [More Information Needed]
- **Compute Region:** [More Information Needed]
- **Carbon Emitted:** [More Information Needed]
## Technical Specifications [optional]
### Model Architecture and Objective
[More Information Needed]
### Compute Infrastructure
[More Information Needed]
#### Hardware
[More Information Needed]
#### Software
[More Information Needed]
## Citation [optional]
<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
**BibTeX:**
[More Information Needed]
**APA:**
[More Information Needed]
## Glossary [optional]
<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
[More Information Needed]
## More Information [optional]
[More Information Needed]
## Model Card Authors [optional]
[More Information Needed]
## Model Card Contact
[More Information Needed] | 5,845 | [
[
-0.048187255859375,
-0.045440673828125,
0.03167724609375,
0.0082244873046875,
-0.0239715576171875,
-0.02520751953125,
0.00882720947265625,
-0.047119140625,
0.0176239013671875,
0.049530029296875,
-0.05621337890625,
-0.051025390625,
-0.044219970703125,
-0.0081024169921875,
-0.0207366943359375,
0.09222412109375,
0.00019621849060058594,
0.008392333984375,
-0.02288818359375,
0.005702972412109375,
-0.035919189453125,
-0.043212890625,
-0.048309326171875,
-0.031585693359375,
0.0306854248046875,
0.0234527587890625,
0.041473388671875,
0.060028076171875,
0.050140380859375,
0.0198211669921875,
-0.0256500244140625,
-0.00921630859375,
-0.0287322998046875,
-0.027557373046875,
-0.01361083984375,
-0.01092529296875,
-0.0718994140625,
0.0092926025390625,
0.043853759765625,
0.047088623046875,
-0.01611328125,
0.04327392578125,
-0.0039825439453125,
0.038970947265625,
-0.045867919921875,
0.02264404296875,
-0.038970947265625,
0.006717681884765625,
-0.0155792236328125,
0.003421783447265625,
-0.0137786865234375,
-0.0070037841796875,
-0.003025054931640625,
-0.0504150390625,
0.0131072998046875,
0.0171356201171875,
0.0889892578125,
0.00974273681640625,
-0.0282440185546875,
-0.01470184326171875,
-0.057525634765625,
0.041900634765625,
-0.05078125,
0.0243988037109375,
0.033935546875,
0.031280517578125,
0.004547119140625,
-0.06304931640625,
-0.0301666259765625,
-0.00958251953125,
0.00006818771362304688,
0.0282440185546875,
-0.0190887451171875,
0.01021575927734375,
0.03948974609375,
0.0384521484375,
-0.03399658203125,
0.00023639202117919922,
-0.039154052734375,
-0.01678466796875,
0.06414794921875,
0.034515380859375,
0.01296234130859375,
-0.0211181640625,
-0.036224365234375,
-0.01324462890625,
-0.0289764404296875,
0.01284027099609375,
0.03857421875,
0.0289764404296875,
-0.04730224609375,
0.051055908203125,
-0.01617431640625,
0.04388427734375,
0.00701904296875,
-0.0007166862487792969,
0.0230865478515625,
-0.047882080078125,
-0.0249481201171875,
-0.01226806640625,
0.050811767578125,
0.037506103515625,
-0.0203094482421875,
0.0074462890625,
-0.024627685546875,
-0.01544952392578125,
0.0308380126953125,
-0.06817626953125,
-0.0200958251953125,
0.02850341796875,
-0.05426025390625,
-0.03167724609375,
0.0023899078369140625,
-0.075927734375,
-0.003803253173828125,
-0.0323486328125,
0.0264739990234375,
-0.0180816650390625,
-0.0287322998046875,
-0.005619049072265625,
-0.0277862548828125,
0.019927978515625,
0.020477294921875,
-0.0596923828125,
0.044036865234375,
0.042755126953125,
0.05511474609375,
0.005641937255859375,
-0.0131072998046875,
0.0022602081298828125,
0.007076263427734375,
-0.000004112720489501953,
0.05169677734375,
-0.022369384765625,
-0.04949951171875,
-0.00879669189453125,
0.01403045654296875,
0.00439453125,
-0.017364501953125,
0.0589599609375,
-0.0289764404296875,
0.013427734375,
-0.030853271484375,
-0.04779052734375,
-0.030364990234375,
0.0194549560546875,
-0.062225341796875,
0.0830078125,
0.01031494140625,
-0.0704345703125,
0.01983642578125,
-0.076904296875,
-0.0196533203125,
0.012786865234375,
0.01320648193359375,
-0.046478271484375,
-0.00560760498046875,
-0.01215362548828125,
0.0343017578125,
-0.0330810546875,
0.015899658203125,
-0.032989501953125,
-0.00559234619140625,
-0.0244293212890625,
0.00472259521484375,
0.08123779296875,
0.0281524658203125,
-0.01092529296875,
0.013641357421875,
-0.06549072265625,
0.007083892822265625,
0.0241546630859375,
-0.0212249755859375,
-0.002559661865234375,
-0.0168304443359375,
0.045501708984375,
0.0082855224609375,
0.0245819091796875,
-0.0308990478515625,
0.013916015625,
-0.0008754730224609375,
0.033050537109375,
0.04107666015625,
0.0208892822265625,
0.0210723876953125,
-0.03228759765625,
0.04644775390625,
-0.00124359130859375,
0.044586181640625,
0.0074615478515625,
-0.04681396484375,
-0.049163818359375,
-0.007495880126953125,
0.021636962890625,
0.041717529296875,
-0.02056884765625,
0.059906005859375,
-0.00504302978515625,
-0.07647705078125,
-0.0125579833984375,
0.00391387939453125,
0.0262298583984375,
0.052337646484375,
0.0292816162109375,
-0.02362060546875,
-0.0570068359375,
-0.06591796875,
0.0063018798828125,
-0.006805419921875,
0.0200653076171875,
0.03424072265625,
0.07098388671875,
-0.0266571044921875,
0.06201171875,
-0.048309326171875,
-0.00778961181640625,
-0.021636962890625,
-0.001499176025390625,
0.004985809326171875,
0.048187255859375,
0.049102783203125,
-0.072021484375,
-0.0137939453125,
-0.0219573974609375,
-0.0419921875,
0.01141357421875,
0.0028858184814453125,
-0.020172119140625,
-0.0085601806640625,
0.0189056396484375,
-0.05072021484375,
0.042449951171875,
0.036590576171875,
-0.028167724609375,
0.053802490234375,
-0.0124664306640625,
-0.006153106689453125,
-0.09002685546875,
0.0335693359375,
0.0135345458984375,
-0.00652313232421875,
-0.032135009765625,
0.0125579833984375,
-0.004238128662109375,
-0.0289306640625,
-0.04498291015625,
0.0584716796875,
-0.02423095703125,
0.001422882080078125,
-0.0185089111328125,
-0.01806640625,
0.0142822265625,
0.035369873046875,
0.016326904296875,
0.03778076171875,
0.035247802734375,
-0.056396484375,
0.0177001953125,
0.0242919921875,
-0.01134490966796875,
0.038238525390625,
-0.06402587890625,
0.0057525634765625,
-0.006130218505859375,
0.02972412109375,
-0.044830322265625,
-0.0283966064453125,
0.028411865234375,
-0.02703857421875,
0.02587890625,
-0.012054443359375,
-0.03948974609375,
-0.03729248046875,
-0.003025054931640625,
0.0230712890625,
0.045654296875,
-0.018463134765625,
0.041748046875,
0.0482177734375,
0.0188751220703125,
-0.016143798828125,
-0.038970947265625,
-0.005306243896484375,
-0.0287322998046875,
-0.031158447265625,
0.04034423828125,
-0.0166015625,
-0.004917144775390625,
0.009307861328125,
0.0156402587890625,
-0.033172607421875,
0.01512908935546875,
0.037109375,
0.019073486328125,
-0.00008797645568847656,
0.004535675048828125,
-0.01068878173828125,
-0.01068878173828125,
0.0115814208984375,
-0.00821685791015625,
0.0196990966796875,
-0.003940582275390625,
-0.0026836395263671875,
-0.05133056640625,
0.041412353515625,
0.037628173828125,
-0.01373291015625,
0.049591064453125,
0.057464599609375,
-0.061676025390625,
0.004207611083984375,
-0.03228759765625,
-0.0186920166015625,
-0.0308990478515625,
0.0281219482421875,
-0.020416259765625,
-0.030517578125,
0.047119140625,
-0.0022125244140625,
0.0010938644409179688,
0.06781005859375,
0.0411376953125,
-0.00786590576171875,
0.07421875,
0.068603515625,
0.0021610260009765625,
0.040618896484375,
-0.036590576171875,
0.003265380859375,
-0.0792236328125,
-0.032501220703125,
-0.060516357421875,
-0.0007867813110351562,
-0.046417236328125,
-0.0105743408203125,
0.00830078125,
0.0114288330078125,
-0.045318603515625,
0.047149658203125,
-0.043121337890625,
0.01023101806640625,
0.041412353515625,
0.01338958740234375,
-0.004650115966796875,
-0.0185089111328125,
-0.003650665283203125,
0.00499725341796875,
-0.052947998046875,
-0.045501708984375,
0.07977294921875,
0.051605224609375,
0.03466796875,
-0.007503509521484375,
0.051055908203125,
0.019012451171875,
0.0201568603515625,
-0.043853759765625,
0.03173828125,
0.00321197509765625,
-0.0745849609375,
-0.001220703125,
-0.0194244384765625,
-0.059967041015625,
-0.0004398822784423828,
-0.029144287109375,
-0.060516357421875,
0.018585205078125,
0.0234832763671875,
-0.03717041015625,
0.0293731689453125,
-0.049102783203125,
0.08795166015625,
-0.032562255859375,
-0.0213470458984375,
-0.007099151611328125,
-0.0435791015625,
0.0296478271484375,
0.0060577392578125,
0.0184478759765625,
-0.01102447509765625,
-0.003574371337890625,
0.06488037109375,
-0.0595703125,
0.07183837890625,
-0.03131103515625,
0.0276641845703125,
0.032501220703125,
-0.0285186767578125,
0.035430908203125,
-0.0021877288818359375,
-0.011932373046875,
0.03607177734375,
0.01666259765625,
-0.03607177734375,
-0.022918701171875,
0.04296875,
-0.06610107421875,
-0.01457977294921875,
-0.03564453125,
-0.03338623046875,
-0.00411224365234375,
0.030426025390625,
0.031768798828125,
0.017120361328125,
-0.016937255859375,
0.0086822509765625,
0.048919677734375,
-0.01561737060546875,
0.00914764404296875,
0.021026611328125,
-0.01032257080078125,
-0.036224365234375,
0.051910400390625,
0.0069580078125,
0.0117340087890625,
0.0199432373046875,
0.017120361328125,
-0.040130615234375,
-0.044677734375,
-0.033233642578125,
0.01033782958984375,
-0.050140380859375,
-0.01450347900390625,
-0.05645751953125,
-0.02496337890625,
-0.0423583984375,
0.0002465248107910156,
-0.03363037109375,
-0.010833740234375,
-0.0433349609375,
-0.01715087890625,
0.035308837890625,
0.042724609375,
-0.01207733154296875,
0.043365478515625,
-0.051177978515625,
0.00690460205078125,
0.007755279541015625,
0.0271759033203125,
0.005077362060546875,
-0.032562255859375,
-0.027801513671875,
0.01377105712890625,
-0.04010009765625,
-0.06512451171875,
0.019622802734375,
0.00006181001663208008,
0.043426513671875,
0.0253448486328125,
-0.001323699951171875,
0.049163818359375,
-0.0300140380859375,
0.0709228515625,
0.0235595703125,
-0.05889892578125,
0.050323486328125,
-0.033905029296875,
0.01206207275390625,
0.059234619140625,
0.04730224609375,
-0.0164947509765625,
0.012603759765625,
-0.07403564453125,
-0.066650390625,
0.0360107421875,
0.0118865966796875,
0.0188446044921875,
0.011505126953125,
0.050628662109375,
-0.01348876953125,
0.023651123046875,
-0.066162109375,
-0.0245513916015625,
-0.02154541015625,
0.0010175704956054688,
0.003276824951171875,
-0.020477294921875,
-0.019287109375,
-0.040496826171875,
0.0687255859375,
0.0147857666015625,
0.03839111328125,
0.007110595703125,
0.0177459716796875,
-0.000873565673828125,
-0.00841522216796875,
0.038909912109375,
0.040496826171875,
-0.04241943359375,
-0.02191162109375,
0.01207733154296875,
-0.043121337890625,
-0.01160430908203125,
0.01239776611328125,
-0.0245513916015625,
-0.0084686279296875,
0.01715087890625,
0.07318115234375,
0.015533447265625,
-0.0304718017578125,
0.0305328369140625,
0.003200531005859375,
-0.0250396728515625,
-0.039703369140625,
0.00275421142578125,
0.01207733154296875,
0.00019359588623046875,
-0.007350921630859375,
0.01319122314453125,
0.027862548828125,
-0.0390625,
0.008087158203125,
0.026947021484375,
-0.0447998046875,
-0.01294708251953125,
0.071044921875,
0.0311126708984375,
-0.0306243896484375,
0.043701171875,
-0.0203094482421875,
-0.0277557373046875,
0.07135009765625,
0.036773681640625,
0.06536865234375,
-0.001285552978515625,
0.0008025169372558594,
0.05438232421875,
0.025360107421875,
0.006603240966796875,
0.0283355712890625,
-0.00760650634765625,
-0.040130615234375,
-0.000514984130859375,
-0.0455322265625,
-0.034515380859375,
0.0245513916015625,
-0.07183837890625,
0.04730224609375,
-0.057037353515625,
-0.0256500244140625,
0.0219879150390625,
0.01953125,
-0.08148193359375,
0.03948974609375,
0.0084686279296875,
0.094482421875,
-0.0753173828125,
0.057525634765625,
0.0640869140625,
-0.0618896484375,
-0.068603515625,
-0.020477294921875,
0.01099395751953125,
-0.0465087890625,
0.0214385986328125,
0.0018835067749023438,
0.0147705078125,
-0.00684356689453125,
-0.046417236328125,
-0.05499267578125,
0.0989990234375,
0.002716064453125,
-0.04705810546875,
0.0034236907958984375,
-0.013671875,
0.03961181640625,
-0.041900634765625,
0.042755126953125,
0.0176239013671875,
0.04193115234375,
0.016998291015625,
-0.056396484375,
0.01348114013671875,
-0.0233306884765625,
0.0129852294921875,
-0.00862884521484375,
-0.060546875,
0.0628662109375,
-0.01314544677734375,
-0.0022735595703125,
0.00914764404296875,
0.031280517578125,
0.009521484375,
0.042388916015625,
0.035003662109375,
0.0582275390625,
0.0594482421875,
0.004749298095703125,
0.0999755859375,
-0.04034423828125,
0.042877197265625,
0.10101318359375,
-0.0131683349609375,
0.06451416015625,
0.0242156982421875,
-0.025848388671875,
0.022003173828125,
0.084228515625,
-0.0258331298828125,
0.02899169921875,
0.016754150390625,
-0.0016679763793945312,
-0.0182342529296875,
-0.0230712890625,
-0.044189453125,
0.02362060546875,
0.01161956787109375,
-0.043701171875,
-0.0135498046875,
-0.00968170166015625,
0.006732940673828125,
-0.02099609375,
-0.0282745361328125,
0.052001953125,
-0.0014715194702148438,
-0.031646728515625,
0.0166473388671875,
0.018768310546875,
0.0262908935546875,
-0.056396484375,
-0.015655517578125,
0.00408172607421875,
0.00005310773849487305,
-0.03436279296875,
-0.04150390625,
0.034271240234375,
-0.0021419525146484375,
-0.0400390625,
-0.01326751708984375,
0.046356201171875,
-0.00714111328125,
-0.057342529296875,
0.0272674560546875,
0.0168914794921875,
0.0237274169921875,
-0.00572967529296875,
-0.0887451171875,
0.01029205322265625,
-0.004974365234375,
-0.007198333740234375,
0.013397216796875,
0.0013303756713867188,
0.000621795654296875,
0.03948974609375,
0.048065185546875,
0.004314422607421875,
-0.01085662841796875,
0.0009775161743164062,
0.072021484375,
-0.0513916015625,
-0.039154052734375,
-0.031585693359375,
0.056610107421875,
-0.018585205078125,
-0.043182373046875,
0.049560546875,
0.060272216796875,
0.0662841796875,
0.0006546974182128906,
0.06561279296875,
-0.0169830322265625,
0.0312347412109375,
-0.0243377685546875,
0.0465087890625,
-0.053924560546875,
-0.00315093994140625,
-0.029388427734375,
-0.0687255859375,
-0.003032684326171875,
0.044036865234375,
-0.0198211669921875,
0.015716552734375,
0.0361328125,
0.047882080078125,
-0.01053619384765625,
0.024749755859375,
0.0080413818359375,
0.01303863525390625,
0.0108489990234375,
0.023956298828125,
0.03936767578125,
-0.053497314453125,
0.01776123046875,
-0.0457763671875,
-0.0250396728515625,
-0.0139923095703125,
-0.08026123046875,
-0.045166015625,
-0.049102783203125,
-0.052001953125,
-0.0316162109375,
0.0028171539306640625,
0.056793212890625,
0.08062744140625,
-0.056793212890625,
-0.0269775390625,
-0.0166473388671875,
0.0075531005859375,
-0.020751953125,
-0.017913818359375,
0.023895263671875,
-0.00135040283203125,
-0.0531005859375,
-0.0055084228515625,
-0.01255035400390625,
0.0214691162109375,
-0.022491455078125,
-0.01006317138671875,
-0.02105712890625,
-0.00356292724609375,
0.0306549072265625,
0.033660888671875,
-0.039886474609375,
-0.0181884765625,
-0.0150299072265625,
-0.00167083740234375,
-0.01093292236328125,
0.049346923828125,
-0.0209503173828125,
0.02783203125,
0.03179931640625,
0.0293731689453125,
0.05224609375,
0.0021190643310546875,
0.0244598388671875,
-0.0194091796875,
0.0084686279296875,
0.0238189697265625,
0.039459228515625,
0.01042938232421875,
-0.05438232421875,
0.041168212890625,
0.029144287109375,
-0.04437255859375,
-0.05694580078125,
0.0008959770202636719,
-0.09429931640625,
-0.005451202392578125,
0.08612060546875,
-0.00493621826171875,
-0.0258941650390625,
0.000027954578399658203,
-0.0157318115234375,
0.01044464111328125,
-0.0217132568359375,
0.037811279296875,
0.061187744140625,
-0.020477294921875,
0.0016155242919921875,
-0.048919677734375,
0.03802490234375,
-0.001934051513671875,
-0.07586669921875,
-0.01407623291015625,
0.039215087890625,
0.036865234375,
0.0080413818359375,
0.0421142578125,
-0.015899658203125,
0.01617431640625,
0.022247314453125,
0.035491943359375,
-0.00943756103515625,
-0.025115966796875,
-0.0270233154296875,
-0.0006151199340820312,
-0.00649261474609375,
-0.0274200439453125
]
] |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.